A New Epistemic Culture?  Wikipedia as an Arena for the Production of Knowledge in Late  Modernity

Lior Gelernter

Wikipedia is a multilingual internet encyclopedia which users can freely use or edit, and whose contents are written collaboratively by volunteers from all over the world. Over the last decade, its innovative mode of production allowed it to accumulate and organize unprecedented volumes of content and establish itself as a prominent resource in contemporary information society. At the same time, however, its reliance on un-vetted and unprofessional voluntary labor raises widespread doubts regarding its reliability and legitimacy as a source of information. This study explores the institutionalization and subsequent reception of Wikipedia as a distinctive, emergent form of epistemic culture – i.e. amalgams of standards and practices underlying the production and legitimation of knowledge (Knorr-Cetina, 1999). As such, it presents a strategic case allowing us to observe how new epistemic cultures are constructed, debated and institutionalized in practice. Current research tends to explain Wikipedia’s success in a-historical terms, using its current features as taken for granted explanatory factors. In contrast, this study draws attention to the contingent and contentious processes through which these features were formed, negotiated and institutionalized during Wikipedia's initial, founding phase using previously unavailable archives from that period. Moreover, while research has tended to neglect the context of Wikipedia's consumption, this study also examines Wikipedia's reception and use by a specific category of knowledge experts —namely, journalists—who use it routinely in the course of their work and have the capacity to propagate its legitimacy. Combining attention to both production and reception, I suggest can lead to a better understanding of Wikipedia's actual success and legitimation processes. Theoretically, this research reconstructs Wikipedia’s history and patterns of production and consumption through pragmatist lenses—emphasizing the importance of exploring concrete practices and objects rather than abstract beliefs or values. Thus, the argument centers on the ways in which actors creatively solve problems they encounter in the course of action, focusing on their use and translation of extant repertoires of practices and justifications. It draws upon insights from several pragmatically oriented theoretical streams, mainly related to Studies of Science and Technology (STS), such as Knorr-Cetina’s constructivism, Susan L. Star's interactionism and Actor-Network Theory (ANT). Opting for such a pragmatist approach allows me to both conceive of Wikipedia as a modern epistemic object, and outline actors’ modes of engagement with the practicalities of knowledge giving which sustain it. Methodologically, the first part of this study is based on a newly available database of almost all the web activity conducted in Wikipedia's formative first year, providing a comprehensive record of every edit entry and dispute in the field. The main analytic tool I use here is the “cartography of controversies” (Venturini, 2009), an Actor-Network Theory inspired approach which maps the locations and actions of different actors within a network, and traces how disputed claims become taken-for-grated and “black-boxed.” The second part, examining Wikipedia's reception, is based on in depth, semi-structured, interviews I have conducted with 20 journalists in Israel. The following, empirical section starts with two chapters linking Wikipedia to its immediate historical origins – the first of them tracing the failure of Nupedia as the first serious open internet encyclopedic project, and the second comparing this failure to the processes which eventually led to Wikipedia subsequent success. In an attempt to give it a solid basis of legitimacy, Nupedia's founders designed a production system based on rigid and restrictive practices of knowledge giving imported from the academic world, and relied on the participation of volunteer certified experts. Its failure to capitalize on the legitimacy of existing cultural practices, expressed in its inability to build a viable gift-economy attractive to the volunteers they sought, led its founders to try and supplement it with its mirror image, i.e. an open, almost unstructured project. The result was Wikipedia: a project aiming for production volume rather than legitimacy, and based on an essentially unrestricted contribution of knowledge. Subsequent chapters then show how Wikipedia's policies and standards were established from this chaotic origin in the midst of a series of conflicts and controversies. As I show through detailed analysis of several such controversies, the negotiation of these conflicts led to the emergence of number of key principles of coordination: the primacy of the traditional encyclopedic model as a "standard for standards"; the legitimacy of the Wiki cultural as a resource for schemes and practices; radical collaboration and lack of hierarchy in the writing and editing process; the reliance on effective consensus between editors to establish policy; the dialogical nature of knowledge construction; the reliance on references to other texts rather than on expertise and original research to establish knowledge; and finally, the preference for productivity and content quantity over control and quality. In addition, I demonstrate how various "conscription devices" (Henderson, 1991), such as policy pages, warning templates and software features, were erected to sustain these principles as well as to establish new forms of gift practices, many of which still stand at the basis of Wikipedia's current structures and practices. Emerging from this description is a distinct form of knowledge production, contribution and coordination displaying a relatively informal pattern of institutionalization, based on the erection of what I term "Privileged Passage Points," (PPPs) a concept coined to highlight the common features of several elements in Wikipedia – such as its policy pages and the authority of its founders—that funnel action without becoming obligatory. As Wikipedia encourages and relies on the voluntary contribution of knowledge even by anonymous or inexperienced users, and as its body of editors is extremely diverse, such flexibility was crucial in allowing it to attract gifts of knowledge from diverse contributors and function as a “boundary object” (Star and Griesemer 1989) for them. The last empirical chapter explores how knowledge experts use Wikipedia and assess its reliability, based on semi-structured interviews with Israeli journalists. Here, I found a routine and universal use of Wikipedia. Simultaneously, however, the interviews revealed a systematic taboo preventing journalists from publicly acknowledging having made use of Wikipedia. Wikipedia thus functions as a hazardous yet practical shortcut to the more fundamental sources of knowledge understood to reside elsewhere, in external authoritative sources indicated by links and references framing Wikipedia's text. Journalists tend to assess the validity of the information conveyed in Wikipedia articles basing themselves on a semiotic frame found in and around the latter, and used to attest to its own credibility and legitimacy. The result, I suggest, is a mode of consumption based on practical legitimacy, i.e. a limited form of legitimization embedded in the practices of Wikipedia's users, and granting article contents a limited yet observable measure of authority, as it takes into account and encapsulates both Wikipedia's de-facto legitimacy—in the sense that "everybody does that"—and lack of de-jure legitimacy. Finally, the conclusion discusses the implication of Nupedia and Wikipedia's cases for the study of contemporary knowledge-giving practices, as well as for a better understanding of the latter's relations with the vision of enlightenment and notions of rationality, science and expertise commonly associated with the very concept of modernity. As for the former, I find that Wikipedia's success is related to its ability to retain a flexible and open gift repertoire congruent with its extremely dynamic social and technological online context. In contrast with Nupedia's reliance on academic-like practices, Wikipedia built on gift-practices already prevalent in the online environment it operated in, adapted them to its innovative goals, and kept them flexible enough to support various types of giving. As for the relevance of the concept of modernity, Wikipedia may seem to challenge the accepted vision of modern knowledge as it is based on the decoupling of knowledge and expertise and of theoretical and practical legitimation, two amalgams standing at the very core of modern theories and methodologies. However, it does so in a way that also keeps assuming the legitimacy of modern epistemologies, and indeed, attempts to insert itself in the legacy of previous encyclopedias. Rather than a radical rift, then, Wikipedia presents a case of more nuanced and subtle transformation. By practically moving from obligatory to privileged passage points, and from essential to practical legitimacy, it constructs a new form of epistemic culture, which draws from earlier conceptualizations of knowledge, but re-aligns the notions of knowledge and expertise in novel ways.

A New Epistemic Culture? Wikipedia as an Arena for the Production of Knowledge in Late Modernity Lior Gelernter Department of Sociology and Anthropology Ph.D. Thesis Submitted to the Senate of Bar-Ilan University Ramat-Gan, Israel May 2013 This work was carried out under the supervision of Professor Ilana F. Silber of the Department of Sociology and Anthropology, Bar-Ilan University. Acknowledgements This dissertation would never have been realized without the guidance and help of various individuals who assisted me in preparing, compiling and – last but by no means least – submitting it. I would like to take the opportunity to thank some of them. To my advisor, Prof. Ilana F. Silber, for her committed and sensitive guidance, for continually challenging me to refine and improve my ideas and my work, and for her enduring faith and dedication. To the President's Fellowship (Milgat HaNasi), for their generous financial support. To Avi Cordova and Natan Sznaider, both of whom I am proud to call my teachers and mentors. To Iddo Tavory, a true friend and colleague who has read and re-read this work, for his encouragement and inspiration. To my friends and colleagues who have read, commented on and discussed parts of this work: Hagay Bar, Tama Halfin, Esty Katz, Stav Kaufman Raviv, Idit Livneh, Ayelet Oz and Tom Pessah. To the guides of the Ph.D. workshop, Prof. Larissa Remennnick, Prof. Ilana F. Silber and Prof. Ephraim Tabory, and for its participants, my fellow Ph.D. students, for their intellectual and moral support. To the founders and participants of Wikipedia, the free encyclopedia, without which the world, and my life, would have been poorer and duller. To my family and friends, who were there for me even when I was absent. Finally, to my lovely, adored and benevolent wife who endured me heroically through these long years of work, and to my children, Yotam and Ayala, who prove children can be cuter and more fun than anything you can find on the internet. Table of Contents Abstract ..................................................................................................................... i Chapter 1: Introduction ............................................................................................. 1 1.1 Wikipedia's Rise to Fame ............................................................................. 4 1.2 Review of the Literature ............................................................................. 10 1.3 Theoretical Framework ............................................................................... 29 1.4 Research Objectives ................................................................................... 49 Chapter 2: Methodology ......................................................................................... 53 2.1 Data Corpus 1: Web Archives .................................................................... 57 2.2 Data Corpus 2: Interviews .......................................................................... 61 Chapter 3: Nupedia ................................................................................................. 66 3.1 Historical Background ................................................................................ 67 3.2 The Encyclopedic Model ............................................................................ 71 3.3 The Mode of Production ............................................................................. 77 3.4 A Controversy: Modifying the Mode of Production .................................... 85 3.5 Conclusion ................................................................................................. 90 Chapter 4: Wikipedia's Founding Period ................................................................. 95 Chapter 5: Encyclopedising the Wiki Technology .................................................104 5.1 Phase 1: Setting the Ground Rules .............................................................109 5.2 Phase 2: Purifying Article Pages: the “Refactoring Debate” .......................116 5.3 Phase 3: The Neutral Point of View ...........................................................124 5.4 Conclusion ................................................................................................142 Chapter 6: Wikifying the Encyclopedic Form ........................................................146 6.1 Phase 1: Wiki is not Paper .........................................................................149 6.2 Phase 2: Wikipedia is not a Dictionary ......................................................152 6.3 Phase 3: What Wikipedia is Not ................................................................176 6.4 Conclusion ................................................................................................191 Chapter 7: Wikipedia in the Field of Israeli journalism ..........................................197 7.1 Wikipedia in the Journalistic Environment ................................................200 7.2 Wikipedia in Journalistic Practice ..............................................................204 7.3 Using Israeli Journalists as Interviewees ....................................................220 7.4 Conclusion ................................................................................................221 Chapter 8: Conclusion and Discussion ...................................................................225 8.1 Theoretical Implications ............................................................................231 8.2 The Gift of Knowledge ..............................................................................239 8.3 Wikipedia and Contemporary Epistemic Culture .......................................244 Bibliography .........................................................................................................248 Appendix: Primary Resources ...............................................................................272 ‫ תקציר‬........................................................................................................................ ‫א‬ Abstract Wikipedia is a multilingual internet encyclopedia which users can freely use or edit, and whose contents are written collaboratively by volunteers from all over the world. Over the last decade, its innovative mode of production allowed it to accumulate and organize unprecedented volumes of content and establish itself as a prominent resource in contemporary information society. At the same time, however, its reliance on un-vetted and unprofessional voluntary labor raises widespread doubts regarding its reliability and legitimacy as a source of information. This study explores the institutionalization and subsequent reception of Wikipedia as a distinctive, emergent form of epistemic culture – i.e. amalgams of standards and practices underlying the production and legitimation of knowledge (Knorr-Cetina, 1999). As such, it presents a strategic case allowing us to observe how new epistemic cultures are constructed, debated and institutionalized in practice. Current research tends to explain Wikipedia’s success in a-historical terms, using its current features as taken for granted explanatory factors. In contrast, this study draws attention to the contingent and contentious processes through which these features were formed, negotiated and institutionalized during Wikipedia's initial, founding phase using previously unavailable archives from that period. Moreover, while research has tended to neglect the context of Wikipedia's consumption, this study also examines Wikipedia's reception and use by a specific category of knowledge experts —namely, journalists—who use it routinely in the course of their work and have the capacity to propagate its legitimacy. Combining attention to both production and reception, I suggest can lead to a better understanding of Wikipedia's actual success and legitimation processes. Theoretically, this research reconstructs Wikipedia’s history and patterns of production and consumption through pragmatist lenses—emphasizing the importance of exploring concrete practices and objects rather than abstract beliefs or values. Thus, the argument centers on the ways in which actors creatively solve problems they encounter in the course of action, focusing on their use and translation of extant repertoires of practices and justifications. It draws upon insights from several pragmatically oriented theoretical streams, mainly related to Studies of Science and Technology (STS), such as Knorr-Cetina’s constructivism, Susan L. Star's interactionism and Actor-Network Theory (ANT). Opting for such a pragmatist i approach allows me to both conceive of Wikipedia as a modern epistemic object, and outline actors’ modes of engagement with the practicalities of knowledge giving which sustain it. Methodologically, the first part of this study is based on a newly available database of almost all the web activity conducted in Wikipedia's formative first year, providing a comprehensive record of every edit entry and dispute in the field. The main analytic tool I use here is the “cartography of controversies” (Venturini, 2009), an Actor-Network Theory inspired approach which maps the locations and actions of different actors within a network, and traces how disputed claims become taken-forgrated and “black-boxed.” The second part, examining Wikipedia's reception, is based on in depth, semi-structured, interviews I have conducted with 20 journalists in Israel. The following, empirical section starts with two chapters linking Wikipedia to its immediate historical origins – the first of them tracing the failure of Nupedia as the first serious open internet encyclopedic project, and the second comparing this failure to the processes which eventually led to Wikipedia subsequent success. In an attempt to give it a solid basis of legitimacy, Nupedia's founders designed a production system based on rigid and restrictive practices of knowledge giving imported from the academic world, and relied on the participation of volunteer certified experts. Its failure to capitalize on the legitimacy of existing cultural practices, expressed in its inability to build a viable gift-economy attractive to the volunteers they sought, led its founders to try and supplement it with its mirror image, i.e. an open, almost unstructured project. The result was Wikipedia: a project aiming for production volume rather than legitimacy, and based on an essentially unrestricted contribution of knowledge. Subsequent chapters then show how Wikipedia's policies and standards were established from this chaotic origin in the midst of a series of conflicts and controversies. As I show through detailed analysis of several such controversies, the negotiation of these conflicts led to the emergence of number of key principles of coordination: the primacy of the traditional encyclopedic model as a "standard for standards"; the legitimacy of the Wiki cultural as a resource for schemes and practices; radical collaboration and lack of hierarchy in the writing and editing process; the reliance on effective consensus between editors to establish policy; the dialogical nature of knowledge construction; the reliance on references to other texts rather than on expertise and original research to establish knowledge; and finally, the ii preference for productivity and content quantity over control and quality. In addition, I demonstrate how various "conscription devices" (Henderson, 1991), such as policy pages, warning templates and software features, were erected to sustain these principles as well as to establish new forms of gift practices, many of which still stand at the basis of Wikipedia's current structures and practices. Emerging from this description is a distinct form of knowledge production, contribution and coordination displaying a relatively informal pattern of institutionalization, based on the erection of what I term "Privileged Passage Points," (PPPs) a concept coined to highlight the common features of several elements in Wikipedia – such as its policy pages and the authority of its founders—that funnel action without becoming obligatory. As Wikipedia encourages and relies on the voluntary contribution of knowledge even by anonymous or inexperienced users, and as its body of editors is extremely diverse, such flexibility was crucial in allowing it to attract gifts of knowledge from diverse contributors and function as a “boundary object” (Star and Griesemer 1989) for them. The last empirical chapter explores how knowledge experts use Wikipedia and assess its reliability, based on semi-structured interviews with Israeli journalists. Here, I found a routine and universal use of Wikipedia. Simultaneously, however, the interviews revealed a systematic taboo preventing journalists from publicly acknowledging having made use of Wikipedia. Wikipedia thus functions as a hazardous yet practical shortcut to the more fundamental sources of knowledge understood to reside elsewhere, in external authoritative sources indicated by links and references framing Wikipedia's text. Journalists tend to assess the validity of the information conveyed in Wikipedia articles basing themselves on a semiotic frame found in and around the latter, and used to attest to its own credibility and legitimacy. The result, I suggest, is a mode of consumption based on practical legitimacy, i.e. a limited form of legitimization embedded in the practices of Wikipedia's users, and granting article contents a limited yet observable measure of authority, as it takes into account and encapsulates both Wikipedia's de-facto legitimacy—in the sense that "everybody does that"—and lack of de-jure legitimacy. Finally, the conclusion discusses the implication of Nupedia and Wikipedia's cases for the study of contemporary knowledge-giving practices, as well as for a better understanding of the latter's relations with the vision of enlightenment and notions of rationality, science and expertise commonly associated with the very iii concept of modernity. As for the former, I find that Wikipedia's success is related to its ability to retain a flexible and open gift repertoire congruent with its extremely dynamic social and technological online context. In contrast with Nupedia's reliance on academic-like practices, Wikipedia built on gift-practices already prevalent in the online environment it operated in, adapted them to its innovative goals, and kept them flexible enough to support various types of giving. As for the relevance of the concept of modernity, Wikipedia may seem to challenge the accepted vision of modern knowledge as it is based on the decoupling of knowledge and expertise and of theoretical and practical legitimation, two amalgams standing at the very core of modern theories and methodologies. However, it does so in a way that also keeps assuming the legitimacy of modern epistemologies, and indeed, attempts to insert itself in the legacy of previous encyclopedias. Rather than a radical rift, then, Wikipedia presents a case of more nuanced and subtle transformation. By practically moving from obligatory to privileged passage points, and from essential to practical legitimacy, it constructs a new form of epistemic culture, which draws from earlier conceptualizations of knowledge, but re-aligns the notions of knowledge and expertise in novel ways. iv Chapter 1: Introduction "The zeroeth law of Wikipedia - The problem with Wikipedia is that it only works in practice. In theory, it can never work" (Wikipedia, 2006) "Even if you don't like the idea of Wikipedia replacing the old knowledge hierarchies of the Encyclopaedia Britannica or the newspaper's archive, you find yourself using it. It just becomes a vital tool. However, you never cite it or link to it. So you can say Wikipedia is like porn – no one talks about it, but everybody uses it" (interview with Eyal, journalist) Wikipedia is a multilingual internet encyclopedia whose contents are accessible without charge under a free content license, and is composed collaboratively by volunteers from all over the world. It is overwhelmingly popular, consistently ranked as one of the 10 most popular websites worldwide and routinely linked-to from other websites and mass media venues as a source for general information and as means for substantiating knowledge claims. However, even though Wikipedia is woven into the very fabric of World Wide Web, and perhaps even into that of everyday life in the contemporary knowledge society, it remains controversial and often viewed with suspicion. The quotes above uncover two fundamental and related tensions that surround Wikipedia as an innovative source of information. As "the zeroth law of Wikipedia," (Wikipedia, 2006) often used by supporters and defenders of Wikipedia (cf. Cohen 2007, 2011; Jesus, 2010; Kelly, 2008; Lee and Karuga, 2010) exemplifies, Wikipedia's popularity presents a paradox: its reliance on the collaboration of myriad unnamed and unprofessional volunteers deviates so drastically from culturally sanctioned methods of encyclopedic knowledge production that even when it is proved a resounding success (in practice) it cannot be regarded as one (in theory). This does not mean that Wikipedia does not have its own theory or methodology of encyclopedic knowledge production. As I show in this dissertation, Wikipedia's accomplishments owe much to the success of its founders in establishing relatively effective, coherent and justifiable mechanisms of knowledge production and legitimacy mechanisms. Wikipedia began as an experimental project, based on an 1 unprecedentedly open and heterogeneous social structure, incorporating an innovative and risky technology, and lacking the ordering and legitimating mechanisms of previous encyclopedic projects. To succeed against skepticism, doubt and conflicts, its founders had to produce new knowledge production and legitimacy mechanisms that could be made commensurable with those of earlier encyclopedic projects. While these efforts proved immensely successful, the epistemic culture1 (Knorr-Cetina, 1999) which emerged as their result is inherently and consciously risk-laden and far from being completely trustable. As they chose to embrace the uncertainty and instability of their new creation, they willingly traded a measure of authority for increased productivity. The result was a new form of practical legitimacy, which grants knowledge with diminished authority as it takes into account and encapsulates both its usability and its liabilities, both its de-facto legitimacy and its lack of de-jure legitimacy. Thus, Wikipedia pays for its ability to harness unprecedented voluntary labor power by explicitly admitting that it cannot, in principle, produce certified knowledge and that the legitimacy of its contents is inherently unstable. The journalist’s quote with which I have opened the dissertation—taken from one of the interviews I conducted—demonstrates the ironic effects of this state of affairs on Wikipedia's reception. The widespread doubts regarding its reliability do not prevent it from being intensively used, even by the very ones who doubt it – as long as they don't have to acknowledge using it and thus to regard the knowledge it contains as certified. The origins and implications of this peculiar situation are the main subjects of this work. Using historical archives and interviews, I draw the general outlines of this new epistemic culture; I trace its institutionalization in Wikipedia's founding period, thereby also addressing more general patterns of institutionalization in contemporary networked knowledge society; and examine how it is reflected in Wikipedia's reception and consumption by knowledge experts – namely, journalists – who use it routinely in the course of their work. 1 Knorr-Cetina defines epistemic culture as those "amalgams of arrangements and mechanisms – bonded through affinity, necessity and historical coincidence – which in a given field, make up how we know what we know" (Knorr-Cetina, 1999, p. 1. italics in the original). 2 While Wikipedia is already the subject of a substantive research literature, this work contributes to our understanding of Wikipedia and its reception in three important ways: first, it explores the concrete controversies and deliberations through which these tensions were negotiated, allowing a better understanding of the processes which produced Wikipedia's epistemic culture. Second, it makes use of previously unavailable historical records to trace these processes of institutionalization which mainly occurred in its early, founding phase. Lastly, it studies the correspondence between the routine practices of both producers and consumers, thereby giving a fuller portrayal of Wikipedia as a cultural phenomenon. These contributions may allow for a deeper understanding not only of Wikipedia itself, but also of more general mechanisms underlying the institutionalization of knowledge production projects and their reception, especially in voluntary and/or unprofessional settings. While other studies emphasized the importance of flexibility and inclusiveness in contemporary networked knowledge society (Castells, 2001; Fuchs, 2008), this work studies how it is actually constructed and sustained, through concrete practices and objects, and through more generalizable network formations. Rather than assume or acknowledge Wikipedia's flexibility, I focus on the complex local processes through which its unique epistemic culture was produced and institutionalized, thus engaging flexibility as an end point of social processes, rather as their cause. A central theoretical contribution which stems from this analysis is the development of the concept of Privileged Passage Points. As opposed to Obligatory Passage Points, “loc[i] that could shape and mobilize the local network” and “have control over all transactions between the local and the global networks" (Bijker and Law, 1992, p. 31), Privileged Passage Points have less power and allow for greater variability, though they succeed in effectively controlling the network in most cases. This relative weakness allows them the whole network to change and adapt faster to changing circumstances and allow for practical compromises between conflicting actors. These features make them especially effective in situation involving relatively unpredictable voluntary participation and contribution, as they preserve participants' sense of free (or at least relatively unconstrained) giving. 3 A second conceptual contribution of this work is a conceptualization of Wikipedia's unorthodox mode of access: in attempting to explain Wikipedia's peculiar consumption patterns, I develop the concept of practical legitimacy. This concept designates a form of legitimation which grants knowledge (or objects, or persons) a diminished yet observable measure authority, as it takes into account and encapsulates both its usability and its liabilities, both its de-facto legitimacy (in the sense that "everybody does that") and lack of de-jure legitimacy (observable thorough repeated utterances expressing doubts and criticism). This concept may prove helpful in accounting for the epistemic culture of digital, widely distributed knowledge production projects, or even become the basis for a better understanding of undertheorized legitimacy formations that fail to achieve a full measure of legitimacy, yet are still effective. This introductory chapter will be comprised of four parts: the first provides a brief overview of Wikipedia and its history. The second presents the burgeoning literature about Wikipedia, focusing on studies that touch on Wikipedia's knowledge production methods, its content, and its reception. The third part outlines the theoretical framework underlying this dissertation. I conclude with an outline of the dissertation in its entirety. 1.1 Wikipedia's Rise to Fame Wikipedia, which celebrated its tenth anniversary in January 2011, is a multilingual internet encyclopedia whose articles are constantly written and edited online by internet users from all over the world. Anyone interested can edit almost any article without restrictions2, using a nickname or remaining completely anonymous. Wikipedia's contents are free, published under a copyright license that allows it to be used or reproduced freely, demanding only that it be cited as a source.3 2 Some articles, susceptible to vandalizing or biased edits, are protected or semi-protected, allowing only veterans to edit them. 3 Wikipedia was originally published under a license called GNU-FDL (abr. for Gnu Free Documentation License). In &oldid09 it migrated to a new very similar license, CC-BY-SA. Basically, both licenses dictate that reuse of content should be accompanied by an attribution to its source, and that derivative works produced using it should be distributed under the same license. See Wikimedia (2009) 4 Wikipedia's website has a uniform structure, based on a clear and schematic design: each page, associated with a specific encyclopedic article4, has two main "tabs" that can be switched by a click of a mouse button: the Article tab containing its content and the Talk tab containing a discussions and commentary regarding the article. Three additional tabs provides alternate ways of accessing the contents of the article or discussion: the default Read tab presents the most recent version of the content; the Edit tab presents the content in a relatively simple word processor format that allows users to alter it, almost without restrictions5; and the View history tab contains all previous versions of the content and various details about them6. This structure renders Wikipedia's production extremely open and transparent, and simultaneously cater to the needs of various users: information seekers in search of a specific fact as well as flaneurs in the realm of knowledge; Students looking for help in their homework as well as professionals interested in a general impression of a subject matter in the periphery of their field; passive readers as well as active editors. Behind this simple and inviting interface, however, lies a huge and complex sociotechnical structure, which governs the countless contributions made to Wikipedia everyday and ensures its stability and growth. Behind the scenes, edits are monitored, users’ actions are judged and rewarded, policies are determined and enforced, and sophisticated technological tools are developed. These mechanisms are the product of long series of processes of institutionalization and transformation, where standards and practices were developed over a long period of time. Wikipedia began as an offshoot of an earlier "open-source"7 encyclopedic project named Nupedia. Both projects were founded by Jimmy Wales, a dot-com 4 This description refers to Wikipedia's main namespace, containing encyclopedia articles. Wikipedia is divided to several "namespaces", each containing another category of pages. Namespaces currently include: Main, User, Wikipedia, File, MediaWiki, Template, Help, Category, Portal, Book, Education program, TimedText, and Module. Each namespace has a corresponding Talk namespace (see below). The basic structure described below is the same in each of these namespaces. 5 Some of Wikipedia's more sensitive or vandalizing-prone articles are protected or "semi-protected", allowing only specific categories of users to edit them, but their amount is negligent in relation to Wikipedia's overall size. 6 The View history tab also contains notes for each version its editor's username or IP, the exact date and time it was created, and a short summary written by the editor. It also contains research tools, such as a comparison between versions and various search and statistics options. 7 Formed in 1998, the open source movement was a commercially driven effort to promote collaborative software development and permissive user license agreements. It was gaining notability and success at the time, and initial attempts were made to translate its success from the realm of software to that of internet content. 5 entrepreneur, and Larry Sanger, a young philosophy PhD and internet enthusiast whom he hired to manage them. Wales was inspired by the Free and Open Source Software (FOSS) movement, which advocated the use of distributed peer-production (Benkler, 2006) to create software which can be copied, re-used and modified. The Gnu/Linux operating system and the DMOZ web directory project, which attempted to copy the FOSS principles to the realm of content production, both gained recognition and success at the time were prominent examples that these ideas can work. Nupedia, launched March 10th 2000, was a rigorous and serious project, with an academic advisory board and painstaking editorial process. However, it soon proved unsuccessful—only a dozen articles were published within its first year of operation, due to the relatively small number of active volunteers, slow article submission pace and complex editorial process. In January 2001, frustrated by this meager output, Sanger and Wales announced a new initiative: an experimental project where articles can be written and edited without any limitations. Sanger named it Wikipedia, a portmanteau of Wiki – the technological infrastructure for collaboratively producing hypertext pages used in the project – and encyclopedia. It was, as he later noted, "a silly name for what was at first a very silly project" (Sanger, 2005). When Wikipedia went live on January 15, 2001, it contained nothing more than a few test pages. Wikipedia's founders, however, quickly amassed a vibrant community of volunteers (as most of Nupedia's participants declined to join it), created standards and practices for this highly unorthodox setting, and modified the Wiki infrastructure to fit aims it was never meant to serve. The newly fledged "Wikipedians" capitalized on both their personal knowledge and the abundance of content and knowledge available on the web to make Wikipedia useful for readers (though still very limited in its contents) in merely a few months. Before long, it proved so successful that Wales moved his focus of interest away from Nupedia and ultimately shut it down. After only a couple of months, the first non-English Wikipedias were set up, and the volunteer population began growing exponentially. Wikipedia's community's success was especially surprising when considering its heterogeneity: The community was comprised mostly of Americans, but it had a large minority of participants from other countries, mainly Europeans. It included 6 supporters from a variety of fringe ideologies: libertarians, socialists, anarchists and techno-utopists, as well as more mainstream vies. Areas of interest were also diverse, spanning such realms as statistics and mathematics, political theory, philosophy, computer and programming, and popular culture and current events. In light of this heterogeneity, and the lack of established practices, Wikipedia's participants soon began setting some policies and ground rules to foster cooperation and to ensure that the content produced in Wikipedia would promote the vision of creating an encyclopedia. The success of these efforts was far from trivial, as Wikipedia didn't have an editor-in-chief, advisory board, formal policy documents or participant selection mechanisms. Policy decisions were generally entrusted with the "Wikipedia community," determined through deliberation and "rough consensus," although the project's founders had a privileged position in these deliberations, and at times insisted on setting policy even against vocal opposition. During Wikipedia's first year, Sanger (who was officially employed as Nupedia's editor in chief and had no formal position in Wikipedia) took the lead role in steering Wikipedia's community, attempting – and generally succeeding – to establish policies and guidelines that will keep Wikipedia's content is as close as possible to that of traditional encyclopedias, in terms of style and tone8. However, Wikipedia's authority structure remained highly flexible and lenient – a leniency expressed in the fact that one of its major principles was (and still is) the "ignore all rules" policy, dictating that "If a rule prevents you from improving or maintaining Wikipedia, ignore it" (Wikipedia, 2012a). Following Sanger's departure in the beginning of 20029, Wales took the leading role in the project, assuming a position often referred to as that of a "benevolent dictator”10 (or "constitutional monarch," as he preferred. Cf. Reagle, 2010a, chapter 6). In this capacity, he held such prerogatives as introducing software and policy changes and banning users. However, he seldom used these capacities, relying on community deliberation and consensus for most decisions, and generally 8 The topics considered relevant for the encyclopedia, however, were way beyond the scope of traditional ones, as articles relating to such topics as popular culture and geek culture were intensively produced. 9 Sanger was laid off Bomis, who was in financial difficulties at the time. He continued to frequent the project from time to time, but subsequently lost his position of authority in it. 10 A term often used in FOSS projects to describe the privileged position of a project's founder in setting policies and steering the developer community. 7 reduced the use of his special position over the years. In 2003 he transferred the rights to Wikipedia from Bomis, his company, to the newly erected Wikimedia foundation, a non-profit organization which operated it ever since.11 At the same year, he erected the Arbitration Committee, a panel of editors12 responsible for deciding on disputes that the community was unable to resolve. In the following years, Wales continued to give up his privileges such as vetoing appointments and blocking users. Today, Wales' formal authority is limited to a sit in Wikimedia's board of trusties, a ceremonial role in appointing new members to the arbitration committee (after they have been elected by the community), and special "founder" privileges in Wikipedia's software, which grant him access to unique technical capabilities, such as removing the privileges of other users. Over the years, Wikipedia’s hierarchy structure became more layered and institutionalized. The emerging status-groups were based mainly on degrees of privileges in Wikipedia's software. As the project grew ever larger, and as new technical tools emerged to help maintain it, more and more users were entrusted with such tools, receiving a position of "administrators."13 While Wikipedia is a highly flexible project, it is by and large institutionalized, as the technical and social changes it continuously undergoes have relatively little effect on its general model and methods of operation.14 Wikipedia's contributor-base grew immensely over time, and it now has more than 80,000 regularly active editors, 35,000 of them in the English language Wikipedia, by far the largest and most productive15. The contribution pattern also changed over time, as more contributions were being made by more casual users.16 Moreover, Wikipedia had come, over time, to rely heavily on "bots," software scripts 11 As well as sister projects created later such as Wikitionary, Wikiquote, Wikisource, Wikinews etc. The first members volunteered, or were "Drafted" by Wales. Subsequent members were annually elected by the community (Wikipedia, 2012c). 13 Subsequently, a whole set of user access levels was devised, including such positions as "bureaucrats," "stewards" and "CheckUsers," each endowed with a special set of privileges. Advancement in the hierarchy was initially based on personal acquaintance and reputation, but later a formal election became routinized and institutionalized. 14 This description is only true to the English language Wikipedia; Wikipedias in other languages does not comply to the same polices, and there is a large degree of variance in the way they operate. 15 All the quantitative data presented in the following paragraphs was extracted from Wikimedia (2012), an automatically generated aggregation of statistical information concerning Wikipedia, unless noted otherwise. 16 As Kittur et al., (2008) noted, "Although Wikipedia was driven by the influence of 'elite' users early on, more recently there has been a dramatic shift in workload to the “common” user." 12 8 that automated much of the work of assembling, editing and safeguarding in Wikipedia. These programs perform such functions as blocking vandalism; linking similar articles in different languages; importing data from external resources and converting it to articles; and assisting Wikipedians by pointing to articles in need of editing and suggesting possible changes. currently, some 39 percent of edits done in all the various language Wikipedia's except for the English language one are bot edits (Niederer and van Dijck, 2010). Even in the English language Wikipedia, which has the lowest percentage of bot edits, they account for almost 10%. These processes allowed Wikipedia to grow enormously over the years. After a long period of exponential growth, its growth is now linear – but still extremely fast, with 12,760 new articles and over 370,000 edits per day in December 2011, (876 new articles and over 100,000 edits in the English Wikipedia). Wikipedia's total size is almost unimaginable: spanning more than 8 billion words in 20 million articles written in over 280 languages, (roughly 4 million in the English version alone), it is the largest and most diverse encyclopedia ever produced. In terms of exposure, Wikipedia is by far the most popular online reference work, and one of the web's most popular websites. It is ranked by Alexa as the world's 5th most popular website (Alexa, 2011) and ComScore estimates its global reach at about 30% of global internet user population, with more than 365 million unique visitors each month, (Wikimedia, 2011). According to a recent PEW survey, it is used by 53% of all adult internet users in the U.S.A. and almost 70% of users with college education or higher (Zickuhr and Rainie, 2011). The status of Wikipedia in terms of its size and usage are somewhat marred by Wikipedia's ambivalent reception. It is often attacked as unreliable, biased, and damaging to contemporary culture. In 2005, it was attacked by the Association for Computing Machinery's Committee on Computers and Public Policy, which claimed that it “cannot attain the status of a true encyclopedia without more formal contentinclusion and expert review procedures” (Denning et al., 2005). Jaron Lanier (2006) claimeds that the importance accorded to Wikipedia's "online collectivism" as means of knowledge production leads our culture on the path to "digital Maoism" which crushes the individual. This is a dangerous phenomenon, as "History has shown us again and again that a hive mind is a cruel idiot when it runs on autopilot. Nasty hive 9 mind outbursts have been flavored Maoist, Fascist, and religious, and these are only a small sampling. I don't see why there couldn't be future social disasters that appear suddenly under the cover of technological utopianism." Similarly, Andrew Keen accused wikipedia of fostering a "cult of the amateur" which degrades expertise and damaging contemporary culture (Keen, 2007). It was even suggested by popular TV host and comedian Stephen Colbert that Wikipedia creates its own Wikiality – a "truth by consensus," detrimental to reason and enlightenment. Wikipedia’s influence, innovative technology and unorthodox production processes places it at the heart of questions and dilemmas regarding the production and consumption of knowledge in contemporary society. It is thus no wonder than that Wikipedia became the center of a growing body of research. In the next part, I outline the major themes and disagreements emerging within this body of knowledge. 1.2 Review of the Literature There are three main themes prevalent in the extant literature on Wikipedia: (a) its content as a cultural product, (b) its usage and influence as an information resource, and (c) the dynamics underlying its production as a voluntary collaborative knowledge production project.17 The first two of these strands demonstrate the overall success of Wikipedia's aspirations, as they demonstrate its success in producing an effective encyclopedia (though it is arguably inferior to more conventional commercial information resources), and its diffuse and widespread use. The last strand, the most sociologically oriented and relevant of the three, aims to explain this success by analyzing the social structures production dynamics that enabled it. Thus, I'll describe the first two relatively briefly, and then move on to discuss the literature about Wikipedia's production at length. 17 In addition to the studies presented here, there are also works dealing with Wikipedia that will not be included in this review. These are mainly works from the point of view of network analysis who analyze Wikipedia's network correspondents with various mathematical models. Cf. Almeida, Mozafari and Cho (2007), Bellomi and Bonato (2005), Buriol et al. (2006), Capocci et al., (2006), Holloway, Božicevic and Börner (2007), Korfiatis, Poulos and Bokos (2006), Zlatic et al., (2006). Another type of works is review essays in information science publications that describe Wikipedia's interface and techno-functional makeup, without attempting to analyze it. Some (such as Mckiernan, 2005; Voss, 2005) are completely descriptive, while others compare it to other electronic knowledge management systems (Kolbitsch and Maurer, 2005, 2006; Wagner, 2006). In the following review, I will only describe these works in cases where they have a direct bearing on the content of the dissertation. 10 1.2.1 Wikipedia as a Cultural Product There is general agreement in the literature that Wikipedia, as a cultural product, retains most of the features prevalent in traditional encyclopedias, though there are disagreements regarding its relative quality as such. The results of studies assessing Wikipedia's accuracy, style and coverage indicate that Wikipedia's articles on traditional encyclopedic topics are generally similar to traditional encyclopedias in terms of their style and content. However, they tend to disagree on the merits of Wikipedia articles in comparison with other encyclopedias or resources, with some regarding Wikipedia as inferior to them and other positing it as equivalent In an oft-cited study, Giles (2005) found Wikipedia's science-related articles to contain only a slightly higher rate of mistakes than the Encyclopaedia Britannica's, although they were generally inferior in terms of style.18 Similarly, Rosenzweig (2006) found that Wikipedia's accuracy in articles about history related topics was similar to that of the Encarta online encyclopedia19. Brown (2011), who reviewed political science related articles in Wikipedia,20 found that "Wikipedia's political coverage is often very good for recent or prominent topics but is lacking on older or more obscure topics." Finally, Halavais and Lackaff (2008) compared Wikipedia's coverage in comparison with three specialized print encyclopedias. They found that the percentage of articles that had no parallels in Wikipedia was 37% for a poetry encyclopedia, 21% for linguistics and 18% for physics.21 Relatively extensive attention had been given to Wikipedia's coverage of health issues, which was found to be generally good but inferior to professional resources. This finding was consistent across several studies with different topics and methodologies: it was the conclusion of a study comparing Wikipedia's coverage of 18 This study was widely cited both in the press and in scholarly articles (Benkler, 2006; ChandlerOlcott, 2009; Chandler and Gregory, 2010; Mercer, 2007) as evidence that Wikipedia is generally reliable not inferior to Britannica. Britannica publicly contested the methodology and results of this study (Encyclopaedia Britannica, 2006), but nature rejected their criticism (Nature, 2006). 19 However, both were inferior to those in American National Biography Online. 20 He examined articles concerning specific candidates, elections, and officeholders. 21 It should be noted that the use specialized encyclopedias renders this comparison problematic, as their coverage is generally wider than that of general encyclopedias. In addition, the data from Wikipedia was gathered in 2006, making it somewhat outdated 11 medical drug information to that of Medscape Drug Reference22 (Clauson et al., 2008); a study comparing Wikipedia's coverage of osteosarcoma with that of US's National Cancer Institute (NCI) website23 (Leithner et al., 2010), and a study in which experts reviewed Wikipedia's articles on four psychiatric disorders (Mercer, 2007). More recently, however, a study which examined articles concerning depression and schizophrenia found that "the quality of information on depression and schizophrenia on Wikipedia is generally as good as, or better than, that provided by centrally controlled websites, Encyclopaedia Britannica and psychiatry textbook" (Reavley et al., 2011). Similarly, Schweitzer (2008) found Wikipedia's coverage of psychological topics to be comprehensive. Finally, some studies were attempted to assess the style of Wikipedia's articles, usually by comparing it to the style of other encyclopedias or information resources.24 Giles (2005) found Wikipedia's natural science articles were less clearly styled and less structured than those of the Encyclopaedia Britannica's, though their accuracy was relatively similar. Similarly, Rosenzweig (2006) found the prose style in biographies of historical figures was significantly inferior to that of American National Biography Online (written by expert historians), noting that "overall, writing is the Achilles' heel of Wikipedia".25 Other studies found Wikipedia's style to be on the same level with that of previous encyclopedias: Emigh and Herring (2005) found that Wikipedia is "largely indistinguishable stylistically from the expert created Columbia Encyclopedia," and Elia (2006) found that Wikipedia's articles style is similar to that of Britannica Online. Finally, Bell (2007) found that Wikipedia articles are similar to parallel ones found in Britannica in terms of readability and grammatical textual composition. To conclude, the disagreements about Wikipedia's quality hide a deeper consensus, about its status as an encyclopedia and its usability as one. To paraphrase the old proverb, considering Wikipedia's unorthodox production methods the wonder 22 The authors found that "Wikipedia has a more narrow scope, is less complete, and has more errors of omission than the comparator database." However, they also noted that it's useful and that the quality of information improved significantly over the preceding 90 days. 23 The authors found that "the quality of osteosarcoma-related information found in the English Wikipedia is good but inferior to the patient information provided by the NCI" 24 The emphasis in such comparisons is mainly on similarities of "quality", and it seems that many of them are not critical enough of their research methods and categories of comparison. 25 However, he also found the style of said articles to be similar to their equivalents in the Encarta encyclopedia. 12 isn't how good of an encyclopedia it is, but that it is an encyclopedia at all. As Wikipedia defines itself as a "work in progress" (Wikipedia, 2012b) any shortcomings are only to be expected. The fact that similar standards can be used to measure it alongside respectable information resources is undisputed in this strand of research, and the very fact that it is often found equal, attests to the project's overall success. 1.2.2 Wikipedia's Usage and Influence A second strand of research looks into Wikipedia's reception rather than its content. This set of studies examined the extent of Wikipedia's use in various contexts and fields, primarily in higher education. The general consensus emerging from these studies is that Wikipedia is extensively used, even though it is often shunned by those in positions of authority (such as university professors, teachers, etc.). Moreover, they indicate that the growing use of Wikipedia tends to undermine the authority structure characteristic of institutionalized realms of knowledge. Several studies that examined Wikipedia's use in academic setting found that it was widely used (Eijkman, 2010; Head, 2007; Head and Eisenberg, 2010; Lim, 2009; Pechacek, 2007; Schweitzer, 2008), and that its use increased over time (Judd and Kennedy, 2010). While most of these studies address students' use of Wikipedia, a study that focused on academics, found that many of them use Wikipedia extensively, though cautiously (Eijkman, 2010). Similarly, the use of Wikipedia to search for medical information is not restricted to patients: Spence (2009) claimed that roughly half of the doctors used Wikipedia to look for medical information. While several studies found that teachers often raise concerns about Wikipedia's use, to the point of discouraging its use (Chandler-Olcott, 2009, Chandler and Gregory, 2010; Eijkman, 2010), some teachers see it as a unavoidable part of contemporary education environment (Maehre, 2009; McPherson, 2006; Rosenzweig, 2006), and some even advocate its use, or report using it successfully as part of their teaching apparatus (Barton, 2005; Caverly and Ward, 2008; Chandler-Olcott, 2009; Chandler and Gregory, 2010; Eijkman, 2010; Nix, 2010; Pollard, 2008; Tarasiuk, 2010). A number of articles point out the increasing and routine use of Wikipedia in both judicial decisions (Peoples, 2009; Murray and Miller, 2010) and law review 13 articles (Baker, 2009).26 In both cases, Wikipedia was cited primarily on extra-judicial issues, as a source for definitions of terms or for references to technology, pop culture, or current events. All of the writers mentioned above share reservations about citing Wikipedia, pointing to the evolving nature of its articles and reliability problems as central causes of concern. Studies that examined Wikipedia's influence finds that its popularity tends to undermine the authority structures of institutionalized realms of knowledge. Barton (2005) claims that Wikipedia changes students' academic writing, and that it takes part in the creation of a critical public sphere that undermines established knowledge hierarchies. Similarly, Eijkman (2010) found that academics believe that Wikipedia disrupts the conventional academic power-knowledge arrangements. Lih (2004a), who analyzed Wikipedia's effects on mass media, found that it has played a significant role in the spread of a participatory form of journalism which undermines the global media ecology (Lih, 2004b, 2004c). Rosenzweig (2006), who described Wikipedia's impact on history teaching in higher education, concluded that it undermines the authority of professional historians and offered possible ways to mitigate this effect. Finally, a few researches attempted to study how Wikipedia's users assess the quality of its articles. Chesney (2006) asked expert academics to rate the quality of Wikipedia articles (and of Wikipedia in general) under tow conditions: half of the participants were assigned an article in their area of expertise, and half were assigned an article outside it. While no difference was found in the assessment of Wikipedia in general, it was found that on average, participants who were assigned articles in their areas of expertise rated it higher then those who were assigned articles outside it. Metzger, Flanagin and Medders (2010) used undergraduate students' focus groups to study how individuals assess the credibility of online information sources. They found that they tend to rely on the opinion of others, often through the use of social networking sites or reputation systems, or on heuristic processes – reputation, endorsement, consistency, expectancy violation, and persuasive intent. Many of their participants questioned Wikipedia credibility, and it was generally considered less 26 Murray and Miller (2010) found that Wikipedia was cited on more than 500 federal or state cases (as of September 2009), and Baker found that Wikipedia was cited in some 500 American law review journals, including 29 of the 30 top ranked law reviews. 14 reliable than information from "real people". Yaari, Baruchson-Arbib and Bar-Ilan (2011) also studied how students evaluate Wikipedia articles. However, in contrast with Metzger et. Al. (2010), they observed users while they evaluated articles. They found that the most frequently mentioned attributes for determining article qualities were not unique to Wikipedia: students used such attributes as the article's length, the presence of external links, and Wikipedia templates (such as "featured article"), most of them similar to ones often mentioned in other information quality studies on the web. To conclude, these studies (as well as the usage statistics described above) indicate that Wikipedia, in terms of its reception, is a highly popular information resource, and keeps gaining ground. Over the previous years, it became a major force in the field of knowledge distribution, exceeding the expectation of its founders. This achievement, as well as its ability to present itself as a genuine encyclopedia, indicates that despite its unorthodox and innovative production, it can and needs be considered a highly successful project. Thus, in the next and central part of this review, I'll outline theories that attempt to explain its success by attributing it to its production processes. 1.2.3 Wikipedia as a Voluntary Collaborative Project Studies dealing with Wikipedia as a knowledge production project usually begin by describing its success, and then attempt to explain the mechanisms or processes that enabled it.27 In general, they agree that the keys to Wikipedia's success are its ability to harness the knowledge and efforts of many volunteers with diverse areas of interest, and its innovative technology and model of collaborative knowledge production which coordinates their actions. Its technology and production model, as these studies point out, compensates for the lack of expertise and control inherent in its voluntary production system by virtue of their ability to accommodate radical forms of participation and collaboration whereby each participant (and theoretically, each reader) can easily and quickly supplement or correct the contribution of any other participant (Bryant, Forte and Bruckman, 2005; Emigh and Herring, 2005; 27 This one, it should be noted, is no different. 15 Konieczny, 2009; Kuznetsov, 2006; Lih, 2004a, 2009; Viégas, Wattenberg and McKeon, 2007). Beyond this broad agreement, there are important disagreements about critical characteristics of this model, resulting in two general approaches in the literature. The first is a network-centered approach focusing on "crowdsourcing" or "the wisdom of the crowds" (Surowiecki, 2004), which highlights Wikipedia's openness as the main principle behind its ability to harness voluntary contributions from huge numbers of casual users and active editors whose work is relatively unconstrained and unreciprocated. A second, community-centered approach focusing on "the wisdom of the few" or "the wisdom of chaperons" (Wilson, 2008), emphasizes the importance of Wikipedia's core community of very active contributors who produce most of Wikipedia's content, effectively governing it by setting and enforcing standards and policies, and taking part in an internal gift-giving culture sustaining their social relations. These approaches are not unique to the research of Wikipedia, but are rather related to broader conceptualizations and analyses of FOSS projects (or "digital commons") and online communities, especially those focusing on knowledge-sharing and voluntary contributions28. Similar to the distinction made above, some conceptualizations stress the highly distributed and somewhat individualized aspects of online participation and sharing (Benkler, 2006; Bruns, 2008; Lessig, 1999; Shirky, 2008, 2010; Surowiecki, 2004; Tapscott and Williams, 2006; Weinberger, 2007, 2011; Rainie and Wellman, 2012), while others stress the communal context of such practices, immersed in an internal gift-giving culture (Barbrook, 1998; Bauwens, 2005; Bergquist and Ljungberg, 2001; Giesler, 2006; Jenkins, 2006; Kollock, 1999; Rheingold, 2000). A similar contention relates to the issue of reciprocity: while some writers (cf. Barbrook, 1998, 2000; Barbrook and Cameron, 1996; Bruns, 2008; Raymond, 1998; Tapscott and Williams, 2006) stress the importance of symbolic exchange in this context, manifested in the existence of a "gift exchange" structuring expectation for recognition or symbolic returns, others stress the non- or even antiutilitarian nature of such contributions (c.f. Rheingold, 2000; Bauwens, 2005), highlighting rather the ideological basis in such acts. However, recent findings point 28 This is hardly surprising, as many of these works used Wikipedia itself as an important example for such settings – in fact, this is true to every such work I found which was published after 2004. 16 out that at least in the context of Wikipedia, these conflicting attitudes are not as mutually exclusive as they may seem, and that both these two modes of social structures and practices have to be taken into account in understanding its operation. Thus, new "technology-centered" explanations try to uncover how Wikipedia exerts control over various types of contributions by pointing to the cooperation between human and non-human actors (such as protocols and software) that together make up the Wikipedia's sociotechnical ensemble29. In what follows, I describe these approaches in the context of research regarding Wikipedia. 1.2.3.1 Network Centered Approach: the Wisdom of the Crowds Some of the most notable interpretations of Wikipedia posit it at the center of a new model of content production based on very large scale voluntary and decentralized (or "networked") collaboration. A term often used to denote this approach is "the wisdom of the crowds"30 (coined by Surowiecki, 2004, and used, for example, in Arazy, Morgan and Patterson, 2006, Kittur and Kraut, 2008; Kittur et al., 2008): it explains Wikipedia's success as a ability to attract a very large and diverse body of volunteers, each producing only a small part of the work in an area she is familiar with, and collaboratively correcting each other's shortcomings. While some researchers in this approach acknowledge the importance of highly active regular users, the emphasis is on the great number and diversity of relatively small individual contributions31. In any case, the difference between the productivity of various participants is usually constructed in quantitative rather than qualitative terms. Writers often invoke the concepts of "the long tail" or "power-law distribution" (Benkler, 2006; Kittur and Kraut, 2008; Kittur et al., 2008; Shirky, 2008): concepts that point to the fact that many participants contribute very small amounts of content, while a few contribute large amounts – and that this pattern is shared by many online phenomena as is thus symptomatic of the internet's technological infrastructure. 29 On the notion of sociotechnical ensemble and sociotechnical systems see Law, (1991), Bijker and Law (1992), Bijker (1993, 1995a, 1995b). 30 It should be noted that a similar term coined by Rheingold (2003) in the headline of his book Smart Mobs, meaning pretty much the same, did not catch as well. 31 Sunstein, for example, claimed that "an essential part of the answer [to the question why Wikipedia is so successful, L.G], is that large numbers of knowledgeable people are willing to participate in creating Wikipedia, and whatever errors they make usually receive massive correction, simply because so many minds are involved" (Sunstein, 2006, p. 151, italics not in the original). 17 In this approach, giving and sharing knowledge is predominantly seen as emerging from the individual acts of isolated actors. Several studies highlight the importance of intrinsic motivations (Okoli and Oh, 2007) such as "fun" (Nov, 2007), "altruism" and "self direction" (Oreg and Nov, 2008) or "educating humanity" and "making a difference" (Kuznetsov, 2006). While self-interest (appearing in such expressions as gaining experience, exposure, and social connections) was also found relevant in such studies, altruistic motivations were found to be more dominant. In any case, such studies usually hold a methodological individualist point of view, as the social or communal side of giving is usually underemphasized in them. Finally, this approach is often related to some soft version of technological determinism which posits that social media technology, informational (rather than physical) goods and/or excess leisure facilitates such acts of giving, making Wikipedia a prominent model for this new form of production (Benkler, 2006; Shirky, 2008, 2010; Tapscott and Williams, 2006). Benkler (2006), one of the major theorists of collaborative online projects, coined the term "peer production"32 to describe this model, and posited Wikipedia as a prominent example of its achievements. Benkler describes it as "a new modality of organizing production: radically decentralized, collaborative and nonproprietary; based on sharing resources and outputs among widely distributed, loosely connected individuals who cooperate with each other without relying on either market signals or managerial commands" (Benkler, 2006, p. 60). Following similar ideas developed earlier by Lawrence Lessig (1999), he praised this model as part of a major cultural and economic revolution fostering a new networked information economy based on giving and sharing rather than economic exchange or organizational hierarchies. This new modality, whose expansion is dependent on new information technology, enhances freedom, democracy and critical culture, and is based on autonomy and selffulfillment wishes of individual contributors.33 Tapscott and Williams (2006) re-articulated this conceptualization in more economic terms. In contrast with Benkler's emphasis on networked organization, they claim that Wikipedia functions so well because it operates as a free market where 32 His original term was actually "commons-based peer production", emphasizing the giving and sharing aspects of this mode of production, but the shorter version has taken roots in the relevant discourse. 33 However, Benkler also emphasized the importance of Wikipedia's norms and community. See below. 18 masses of contributors act more or less individually. The market is perceived as faster and as more reliable than a centrally controlled system and as more efficient the more participants it includes. In outlining this model of peer production, they claim that it depends on the following conditions: a) the object produced is cultural or informational; b) the work of producing it can be "chunked out into bite-size pieces that individuals can contribute in small increments and independently of other producers"; and c) the costs of assessing and integrating those pieces must be low (Tapscott and Williams, 2006, p. 70). Wikipedia is portrayed as benefiting from all three conditions.34 Its openness, and the large user population it encourages, allows it to utilize an "almost infinite wealth of talent, energy, and insight that far exceeds what Britannica's closed model can muster" (ibid). A similar economic approach is advocated by Clay Shirky (2008, pp. 109-142, 2010), who used Wikipedia as a prominent example for a model for new online social tools that utilize the "cognitive surplus" created by expanding leisure for collaborative production. He claims that "Wikipedia's lack of managerial direction makes it easier for the casual contributor to add something of value; in economic terms, an open social system like Wikipedia dramatically reduces both managerial overhead and disincentives to participation" (Shirky 2008, P. 130), thus solving the potential “freerider problem” inherent in such projects. In allowing for differing levels of participation, Wikipedia encourages a power-law distribution of contributions, meaning that a minority is responsible for most of the contributions, but an extremely large population of low-level users is responsible for many others – and often more substantial ones. While Shirky acknowledges the importance of highly active contributors, he hardly addresses such issues as community norms and project wide policies, focusing on the levels of individual contribution and local interaction. Weinberger also stresses the distributed nature of knowledge production in Wikipedia, as well as in the internet in general. As suggested by the subtitle of his Too Big to Know (Weinberger, 2011) – “Rethinking Knowledge Now That the Facts Aren't the Facts, Experts Are Everywhere, and the Smartest Person in the Room Is the 34 While Tapscott and Williams (2006, p. 72) acknowledges the importance of the "hard-core group of about five thousand Wikipedians who gladly accept responsibility for the large variety of tasks that keep Wikipedia humming," they view Wikipedia's advantages as stemming mainly from its open model which allows anyone to edit it, and the masses of contributors involved in mainly individual work. 19 Room” – he sees Wikipedia as a network whose knowledge surpasses that of the individuals in it: “As knowledge becomes network, the smartest person in the room isn’t the person standing at the front lecturing us, and is in the collective wisdom of those in the room. The smartest person in the room is the room itself: the network that joins the people and ideas in the room… It’s not that the network is becoming a conscious superbrain. Rather, knowledge is becoming inextricable from – literally unthinkable without – the network that enables it. [p. xiii] A central point in this argument is that Wikipedia (and similar projects’) success in building a successful network stems from its distributed leadership pattern and decentralized or network decision-making (pp.170-171). Lastly, reformulating this standpoint from a more organizational position Konieczny (2009, 2010) maintains that Wikipedia has an extremely decentralized governance structure. This structure, which he argues is very close to that of an adhocracy, is characterized by "flat hierarchy, decentralization, little managerial control, and ad-hoc creation of informal multidisciplinary teams" Konieczny (2010). This structure, the open technology underlying it, and its heterogeneity, makes Wikipedia a unique case of organization immune to the iron law of oligarchy. While Konieczny does not preclude oligarchy as a future option, he thus holds it to be structurally unlikely. As he state, "Among the millions of editors and hundreds of wiki organizations, there is too much diversity for the formation of a conscious oligarchy" (Konieczny, 2009). As is evident, this approach is immersed in a technological utopian ethos that accompanied the growing interest in the trend known as "web 2.0" (O'Reilly, 2007) – the emergence of websites (such as Youtube, Flicker, Wikipedia, etc.) based on the contributions and interactions of many individual users, and lacking central authority.35 In this reading, Wikipedia is envisioned as a veritable spearhead of this trend, celebrated as a restructuring of the internet in a way that fosters participant democracy and subverting central authority. However, this approach soon came under criticism, first and foremost by Wikipedia's participants themselves. This reaction, based on a mixture of academic and independent research, stressed the fact that most 35 For a history and critique of technological utopianism see Segal, 1985, Tenner, 1996, Hayden, 2009. 20 of Wikipedia's edits are made by a very small minority of contributors. This criticism was related to, and significantly promoted, a second interpretation of Wikipedia, focusing on "the power of the few" (Kittur et al., 2008) or "the wisdom of the chaperons" (Wilson, 2008). 1.2.3.2 Community Centered Approach: the Wisdom of the Few While the approach described above stresses the importance of Wikipedia's openness, decentralization, heterogeneity and networked structure, this second strand of research attributes Wikipedia's success mainly to a relatively small and cohesive community of dedicated and experienced editors, and the culture of giving and reciprocity associated with it (cf. Reinhardt, 2003; Bryant et al., 2005; Rafaeli, Hayat and Ariel, 2009, O'Sullivan, 2009, Reagle, 2010a). This community not only produces much of its content and guards it against disruption; it is also responsible for Wikipedia's policy formation, technological development and maintenance. Influenced by theories of virtual communities (Rheingold, 2000) and communities of practice (Lave and Wenger, 1991, Wenger, 1998), the studies employing this approach usually highlight issues such as the effect of communal identity and norms on motivation and practices, the sense of commitment to the community and project, and socialization of new participants. As noted above, the basic finding underlying this approach is the disproportional quantity and quality36 of work done by a relatively small group of regular contributors. The initial assertion by jimmy Wales in the early days of Wikipedia, based on a research he conducted using raw data from Wikipedia was that more than half the edits made by logged-in users in Wikipedia are made by only 2.5% of these editors (Wales, 2005a). Later, refining his claims, he asserted that "over 50% of all the edits are done by just.7% of the users … 524 people. … And in fact the most active 2%, which is 1400 people, have done 73.4% of all the edits" (quoted in Swartz, 2006). Later studies seem to reinforce this assertion. Ortega and Gonzalez-Barahona (2008), for instance, found that in each of the ten largest language versions of Wikipedias, less than 10% of the registered users are responsible for more than the 36 While there is no clear measure for quality in Wikipedia, a number of studies used retention rate (i.e. the percentage of an edit which survived subsequent edits) as an approximation. 21 90% of the total number of edits. Anthony, Smith and Williamson, (2007), who checked the retention rate of content in the French and German Wikipedias, found that content added by veteran registered users tends to stay in Wikipedia more than new or unregistered users.37 Similarly, Priedhorsky et al., (2007) found that frequent editors dominate what people see when they visit the English Wikipedia, as more of their words are retained in subsequent edits than those of irregular ones, especially in popular articles. They also claim that this domination is increasing over time. The importance of Wikipedia's core community of very active editors is further highlighted in studies about Wikipedia's response to vandalism and disruptions, such as the insertion of false or biased content, replacing or supplementing legitimate content with obscenities or gibberish, and deleting of whole article pages. Ciffolilli (2003) claimed that Wikipedia's success in dealing with vandalism stems from a reputation system based on contribution to the project that underlies a consensual community and authority structure geared towards improving useful content and deleting unwanted contribution. Similarly, Lorenzen (2006) contended that Wikipedia's community succeeded in developing robust production norms and sophisticated mechanisms for detecting and warning against potentially problematic behaviors and assessing their nature, as well as effective sanctions to deter problematic editors. He concludes that "there is a strong and viable reaction to problematic behavior at Wikipedia. The community is quite capable of defending itself from obvious vandals." As this approach stresses the social context of production processes in Wikipedia, it attempts to describe how this context enhances participants' productivity and motivations. The most essential aspect of such explanations is probably the emphasis on the importance of social norms and their enforcement by the Wikipedia community. This aspect is highlighted in many accounts of Wikipedia, (Bryant et al., 2005; Emigh and Herring, 2005; Farrell and Schwartzberg, 2008; Pentzold, 2011; Priedhorsky et al., 2007; Rafaeli, Hayat and Ariel, 2009; Reagle 2010a, 2010b; Viégas, Wattenberg and McKeon, 2007), and is most often credited as a direct cause to its success. Such accounts usually highlights such norms as cooperation, consensus building, assuming good faith (especially Bryant et al., 2005; Reagle 2010a, 2010b), 37 The only category with similar retention rates was the rare "Good Samaritan" experts, who contribute only once. 22 adhering to the stylistic tropes of encyclopedia articles, and especially to a neutral point of view, assuming a critical point of view towards knowledge and attempting to debate it in a rational and matter-of-fact manner (Benkler, 2006; Emigh and Herring, 2005; Farrell and Schwartzberg, 2008). A second essential feature of this approach is the importance of Wikipedia's community to the identity and motivation of contributors. In contrast with studies emphasizing "intrinsic" motivation (see above in the previous section), most researches in this approach stress the importance of community related motivations (Rafaeli, Ariel and Hayat 2007, Rafaeli, Hayat and Ariel, 2009; Schroer and Hertel, 2009). Thus, while Kuznetsov emphasized the important of intrinsic motivations, he also found that 15% of Wikipedians report that they contribute in order “to give back to the Wikipedia community" (Kuznetsov, 2006). Ciffolilli (2003) posited that Wikipedians were moved to contribute by their desire to gain reputation within the community. Similarly, Anthony et al., (2007) claim that Wikipedia "zealots"38 are motivated primarily by their wish to acquire reputation and to belong to Wikipedia's community. Reagle's Good Faith Collaboration (2010a), the most extensive ethnographic account of Wikipedia to date, describes its social norms and community processes at length, presenting it first and foremost as a community driven project. His explanation of Wikipedia's success is essentially a description of the beneficial features of its community. As the title of his book suggests, the most prominent of these is its "good faith collaboration culture": a collection of prosocial norms which reflect a trusting attitude towards contributors (whether they are familiar or not) and their actions. A second feature is its nature as an "open content community" characterized by its focus on open (as in open-source) content; the transparency of its rules and processes; the integrity of the processes; the evaluation of person and contribution based solely on their merit; and the right to "fork"39. A third feature is the community's reliance on consensus for most of its decision making processes, and the various community mechanisms used to establish and create such consensus. Finally, the last feature is 38 Who are responsible to most of the high quality contributions in Wikipedia, see above. Reagle calls this right "non-interference". Forking, the right and ability of dissatisfied community members to copy existing content and use it as the basis for a new or derivative project, is considered a basic right in all FOSS projects. 39 23 the "authorial" leadership pattern embodied in Jimmy Wales, based on merit and trust, and enacted through civility and humor. A similar approach was offered by Axel Bruns (2008), who coined the term produsage to explain its essence40. In his insightful analysis of Wikipedia, based on concepts such as collective intelligence and knowledge spaces developed by Pierre Lévy (1997), Bruns claims that it is Wikipedia's innovative community structures and processes that drive its success: Content in Wikipedia evolves under the influence of policing and development not by individual actors operating in inherent isolation from one another, but by communities organized through emerging and increasingly complex yet fluid heterarchical structures; far from being unspecified, a great number of implicit and explicit conventions, rules, and norms… govern – on a ad-hock basis – this continuing communal process…" (Bruns, 2008, p. 128) Thus, while stressing Wikipedia's "complex and fluid community," he highlights "community standards" as the main factor for improvement in quality. Bryant et al., (2005) focus on the socialization process of Wikipedians, examining how Wikipedians move from a state of Legitimate Peripheral Participation (LPP) to full community involvement, taking on more roles and responsibility41. They contend that the socialization process transforms participants' perceptions, motivations and actions, as their participation becomes more central and frequent: "They identify the site, not as a random collection of articles, but as a community of co-authors who play distinct roles and have distinct talents as they build a resource. They move from a local focus on individual articles to a concern for the quality of the Wikipedia content as a whole and the health of the community." It should be noted that the community approach does not stand in total contrast with the network approach, but rather they are a matter of emphasis. Benkler's description, for example, stresses the importance of decentralization and openness but 40 Burns claims that in virtual communities such as blogs, second life and Wikipedia the very separation between producers and consumers is inapplicable. His alternative term, produsage, "[…] highlights that within the communities which engage in the collaborative creation and extension of information and knowledge […], the role of ‘consumer’ and even that of ‘end user’ have long disappeared" (Bruns, 2008, p.2). 41 However, Panciera, Halfaker and Terveen (2009) rejected Bryant's finding, claiming that "Wikipedians are born, not made": that is, the writing pattern of users that subsequently became dedicated users was distinguishable from their early days in the project, and did not change much over time. Further evidence is needed to determine the validity of these conflicting claims. 24 at the same time argues that Wikipedia's success is dependent on "self-conscious social-norms-based dedication to objective writing" and "self-conscious use of open discourse, usually aimed at consensus" (Benkler, 2006, p. 72). This combination, he asserts, prevents it from having to use software based editing restrictions and sanctions and allows it to exercise greater freedom than similar projects. In his latest book, the Penguin and the Leviathan (Benkler, 2011), he moved closer to the community approach, further emphasizing the importance of Wikipedia's norms and communal mechanisms, and stressing the crucial role of moral commitments to such a community structure which fosters and sustains them (Benkler, 2011 pp. 152-159). Similarly, Bruns (2008) does not only focus on community enforced norms and standards, but at the same time highlights the "complex yet fluid heterarchical structures" underlying it, and Rainie and Wellman (2012) focus on the heterogeneous networked in which "networked individuals" in Wikipedia chose to take part. Similarly, while most studies concerned with motivations for giving side more closely with one of the approaches, they usually also acknowledge the relevance of motivations related to the other approach. While the community approach is indispensible in attempting to understand Wikipedia's success, recent research seems to indicate that it is gradually losing ground. Recent findings suggest that Wikipedia's workload, initially carried out mainly by a core community of Wikipedians is now increasingly skewed towards more casual users, or "crowds" (Kittur et al., 2008). Other studies found that Wikipedia's governance is becoming increasingly decentralized (Forte and Bruckman, 2008; Forte, Larco and Bruckman, 2009). Moreover, it seems that Wikipedia's core community began suffering from erosion, as it was suggested that editor retention rates are dropping (Ortega, 2009). These changes brought about a third approach which attempted to transcend previous dichotomies and focus on Wikipedia's sociotechnical makeup. 1.2.3.3 Technology Centered Approach: "The Technicity of Content" This approach attempts to go beyond the two mentioned above, as it highlights the conditions that allow the co-existence of both individual and group-oriented 25 participation in Wikipedia. Thus, it explains Wikipedia's success by pointing to its ability in to harness various types of contributors and contributions, with divergent and often conflicting motivations. As a growing number of studies began to address multiple types of contributions (Anthony et al, 2007; Kittur et al, 2008) and motivations for contributing (Kuznetsov, 2006; Rafaeli and Ariel, 2008), this approach attempt to point at specific features in Wikipedia which fosters this diversity, manifested in its technological makeup or design. While the approaches outlined above focus on Wikipedia's human resources, Niederer and van Dijck (2010) offer a fresh approach as they focus on the sociotechnical makeup of the website. Wikipedia, they claim, exerts "protocological control": a mode of control that is at once social and technical based on standards of action for using specific technology. At the same time, they stress the importance of non-human content agents, such as bots42. Contesting the dichotomy between the network and community approaches, they contend that Wikipedia's functioning depends on its ability to exert effective control on all contributors, both casual and very active one. Wikipedia is "a gradually evolving sociotechnical system that carefully orchestrates all kinds of human and non-human contributors by implementing managerial hierarchies, protocols and automated editing systems."43 Niederer and van Dijck (2010) focus on the sociotechnical makeup of Wikipedia and on the constant collaboration of human and non-human agents in defining the quality standards in Wikipedia. Building on the work of Galloway (2004) they contend that Wikipedia's production process is neither "open" nor "communitarian," but rather mediated through modulated disciplinary control. Contributors – both humans and non-humans – enjoy different permission level to Wikipedia's tools, and their contributions are regulated through a system of rewards (manifested in access to higher permission) and restrictions (manifested in blocking or removing user permissions). Moreover, Wikipedia's content is increasingly "co- 42 Bots are software scripts that automatically handle much of the work of assembling, editing and safeguarding in Wikipedia. 43 In addition, Niederer and van Dijck claim that the pattern of relations between humans and bots is indicative the robustness of various language Wikipedias, as smaller language Wikipedias exhibit a larger percentage of bot edits and less human interactions with them; in contrast, "a richer variety of bot activity, largely subservient to human edit activity, could be considered an indicator of a large and lively language space." 26 authored by bots" – non human content agents – whose contribution is crucial to Wikipedia's activity. Niederer and van Dijck (2010) were not the first to address the importance of technology in Wikipedia – for example, Priedhorsky et al., (2007) claim that Wikipedia's ability to counter vandalism is increasingly dependent on the use of antivandalism bots. Similarly, Viégas, Wattenberg and McKeon (2007) note that Wikipedia's "featured articles" process depends on technological aspects of the Wiki and Wikipedians' creative use of them. However, they are the first to conceptualize it as the crucial element sustaining its overall success. While this approach goes a long way towards explaining Wikipedia's success, I would contend that it has two major shortcomings: first, it fails to explain how such technological and protocological mechanisms were constructed in the first place. Second, I contend that it misses a crucial element in Wikipedia's success: the fact that it owes its success, at least in part, to the relative lack of control over the contribution of its participants, which allows it to maintain an ethos of "free giving" at the expense of accuracy and reliability. To summarize this subchapter, studies about Wikipedia generally agree that it is commensurable with more traditional encyclopedias and that it is widely and increasingly used in various cultural fields. However, they differ in their explanations of the underlying social structure mechanisms responsible for this success. In this study, I shall attempt to answer some of the lacks in the literature, while providing a more general model for networked production and legitimation. First, the attempt to assess Wikipedia as a cultural product by comparing its content to that of traditional encyclopedias is limited in its contribution to understanding what Wikipedia is and how it functions. Such studies fail to address the ways people actually use Wikipedia. Despite all the differences between the studies reviewed above, they all share a common methodological problem: they study a very small group of articles, namely those that have parallels in more traditional encyclopedias or reference works. However, with more than 3.5 million articles in the English Wikipedia, it seems that the overwhelming majority of articles have no parallels that can be used for comparison and thus remain unstudied. This problem is 27 even greater when considering the actual uses of Wikipedia: a study of Wikipedia's traffic found that the most visited category of articles was entertainment,44 rating topics that have no parallel in traditional encyclopedias higher than those that do (Spoerri, 2007). Thus although it is certainly interesting to compare Wikipedia entries to parallel entries in traditional encyclopedias, it may be more productive to examine the shared practices, standards and policies that shape the bulk of Wikipedia's articles, including those that have no parallel in traditional encyclopedias. Instead of focusing on the standards set by traditional encyclopedias, studying these policies help uncover the important differences that make Wikipedia's usable and popular, such as its readiness to dwell on fleeting and concrete subjects that are absent from traditional encyclopedias, its use of references and external links, and the way criticism, interpretation and reservations are manifested in it.45 In order to examine Wikipedia's characteristics as a cultural product, I examine how Wikipedia articles in general are structured, and how their features render Wikipedia at the same time very similar and very different from other encyclopedias. Taking a pragmatist position, this direction will be supplemented by a focus on how Wikipedia's articles are used in everyday life, rather than in the controlled environment of a review process. I will thus demonstrate how these unique features are central to Wikipedia's use, and argue that these features are more relevant to the needs and prevalent practices of Wikipedia's users than its accuracy or readability. This direction is especially warranted considering that studies that assess Wikipedia's social impact failed to describe how Wikipedia is incorporated into the shared and routine practices in established fields, changing their micro-level formations rather than their macro-level structures and ecology.46 Moreover, they do not attempt to describe the alternative structure of authority over knowledge that Wikipedia 44 Subsequent categories were politics and history, geography, sexuality and science. Entertainment, together with politics and history accounts for more than half of the most popular articles. 45 Foucault (2001) studied the emergence of criticism and hermeneutics as form of writing in the discourse that bred Diderot and D'alamber's Encyclopedie, while Rosenberg (2001) pointed to the far reaching significance of cross references in it. 46 A rare exception can be found in Lim (2009), who describes how students use Wikipedia, claiming that they do so more often for personal purposes than for academic ones, and generally tend not to use if for homework assignments and reference search. However, the small sample (it used an online survey conducted only in one public university), as well as the use of broad categories of use, implies that further research is needed. 28 advances (except for generally claiming that it is more "open" or "collaborative"), and how it is received. Thus, I use interviews with journalists – a milieu previously studied in this context – to examine how they incorporate Wikipedia into their everyday routine, and how this incorporation reflect a distinctive structure of authority over knowledge. The central contribution of this dissertation, however, lies in the attempt to account for Wikipedia's success by analyzing the institutionalization of its social structures and dynamics. Previous accounts of Wikipedia's success, described above, attribute it to either "the wisdom of the crowds," to the "community norms," or to sociotechnical mechanisms. Leaving aside theses differences, these explanations all share a common problem: they neglect to address the fact that Wikipedia had to gain considerable initial success in order to attract crowds; that "community norms" are not a given but an achievement – and an extremely difficult one at that; and that the technology it incorporated had to undergo considerable development and changes to accommodate the needs of the project (rather than the opposite direction of influence underlined by some of the studies above). Although important, these theorizations thus begin with a relatively static, “finished,” system. An explanation of Wikipedia's success has to account for its ability, starting in its early days, to achieve these accomplishments. Thus, using previously unavailable archives containing almost all of Wikipedia's activity in its first year, as well as other resources, I examine the creation and institutionalization of Wikipedia's epistemic culture, essential to explaining its subsequent success. 1.3 Theoretical Framework The theories underlying the literature regarding Wikipedia thus have two main shortcomings: first, they are usually a-temporal in their orientation; second, in focusing on Wikipedia's success they fail to conceptualize its peculiar mixture of popularity and lack of legitimacy. The tendency to neglect Wikipedia's time-bound aspects is most evident in the widespread descriptions of Wikipedia as a kind of structured totality: a social structure with relatively stable features which explain its success, and are approached from a synchronic rather than diachronic perspective. They present selected features of Wikipedia (e.g. social structure, culture, technology) 29 as the causes for its success, without explaining how these features themselves were erected. The goal of this work, however, is to better understand how such features are constructed, through a more dynamic and processual approach. Thus, I explore the contribution of the pragmatist tradition, which dealt with these issues intensively, and three related strands in the social sciences: the symbolic-interactionist perspective, social studies of science and technology, and pragmatic sociology. Subsequently, I briefly outline two additional theoretical issues to be addressed in this work: the charitable giving of knowledge and the relations between contemporary information society and modernity. 1.3.1 Pragmatism Pragmatism is a philosophical perspective which emerged in the United States in the end of the nineteenth century. Its main proponents were Charles S. Pierce, William James, John Dewey and George Herbert Mead. These thinkers shared an emphasis on experience and practices, and rejected attempts to build overarching systems of abstractions that would fix the world in one "true" form. In contrast with Cartesian rationalist philosophy, their basic view of the world posited it as being in a constant state of flux. As James put it: "The essential contrast [between pragmatism and rationalism] is that for rationalism reality is ready-made and complete from all eternity, while for pragmatism it is still in the making, and awaits part of its complexion from the future. On the one side the universe is absolutely secure; on the other it is still pursuing its adventures" (James, 2008 [1907] p. 108) It is up to human beings to make sense and give a concrete order of this world – not once and for all, but in the context of their specific environments, needs and situations they encounter. Deeply influenced by Darwinism and rejecting traditional idealist or dualist philosophies, they believed that "humans are problem solvers and the function of thought is to guide action in the service of solving practical problems that arise in the course of life" (Gross, 2009). When no such problems exist, human behavior is generally guided by habits (or beliefs, concepts that were usually constructed as synonymous) and runs along smoothly. However, when action is obstructed, humans experiment and attempt to generate new practices and meanings. 30 Pragmatism was thus, perhaps more than anything else, a theory of knowledge (or more generally, culture) and its production. Its basic premise is that knowledge is actively constructed, and thus always local, contingent and contextual. It is intimately woven into of the everyday situations that make up our lives. Knowledge, meanings and objects are the result of practical needs, and are manifested in practical outcomes. As Pierce’s original formulation of the pragmatist maxim states: "[T]here is no distinction of meaning so fine as to consist in anything but a possible difference of practice… Consider what effects, which might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of those effects is the whole of our conception of the object" (Pierce, 2012 [1878], p. 119. italics added L.G.). As their emphasis on experimentation betrays, the pragmatist thinkers were generally committed to the notion of scientific enquiry. However, they tended to generalize it in ways that lead to a highly contextual and often pluralistic view of knowledge production. Knowledge production, claims Dewey (1941), is always done through a concrete inquiry borne out of concrete problems, and its results (i.e., knowledge) are contingent upon its procedure, or means. These means in themselves have no universal properties, and should be suited both to the object of study and to the concrete circumstances in which the inquiry takes place.47 These means tend to start out as rather "vague and coarse," but progress over time, as the problem itself takes on a clearer shape and is gradually articulated. Thus, the pragmatist conception of knowledge is inherently social – a feature extensively developed both in Peirce’s notion of "a community of inquiry” (Pierce, 2012 [1878]),48 and in Mead’s seminal Mind, Self and Society (Mead, 1967[1934]). This theory of knowledge production is highly useful in attempting to understand Wikipedia's temporal trajectory. As I show throughout this work, Wikipedia's history follows a very similar trail, by which the object of knowledge and the means to construct it are gradually articulated and co-constructed, and habit and creativity alternate as problems lead to re-articulations of both the project and its 47 Dewey (1941) notes that means "are formulated in discourse, i.e., as propositions, and…as means they have the properties appropriate to means (viz., relevancy and efficacy-including economy)." I take the word efficacy to relate (at least amongst other issues) to the social circumstances that condition them. 48 This notion was later elaborated and broadened by Dewey (1944 [1916]). 31 goals. This approach is also highly relevant to understanding how Wikipedia is incorporated into knowledge production practices of consumers, as its use is conditioned upon and conditions the concrete environments in which it operates. Moreover, this view helps redirect the focus from the overly theoretical and judgmental question of Wikipedia's legitimacy as a project and as a cultural object (cf. Sanger, 2009; Wray, 2009) to the actual practices its producers and consumers perform in their attempt to construct it as legitimate. As the above makes clear, the present study has deep affinities with pragmatic thinking. However, the pragmatist tenets presented are philosophical heuristics rather than a method of social inquiry. I thus move now to describe three sociological legacies that are inspired by pragmatism, and from which I draw heavily in this work: the symbolic-interactionist perspective, social studies of science and technology, and pragmatic sociology. 1.3.2 Symbolic Interactionism Symbolic interactionism is perhaps the most obviously pragmatist form of sociology (Shalin, 1986; Gross, 2007). According to Shalin, the basic tenets of symbolic interactionism – the philosophical perspective on reality as being in the state of flux, the sociological view of social structure as emergent process, the methodological preference for participatory modes of research and the ideological commitment to ongoing social reconstruction - are derived from pragmatist thought. George Herbert Mead, influenced by Dewey's pragmatism, began translating it from a philosophical school to a social research endeavor, a task later continued by his own student, Herbert Blumer, who coined the term (which he, himself, called a “barbaric neologism”) "symbolic interactionism". Mead emphasized the social aspects of pragmatic behavior: objects, meaning and knowledge are created intersubjectively, through social acts and social processes (see Mead, 1967 [1934], especially pp. 6882). Blumer later expanded this insight, positing three basic premises for the new school: "The first premise is that human beings act toward things on the basis of the meanings that the things have for them. Such things include everything that the human being may note in his world— physical objects, such as trees or chairs; other human beings, such as 32 a mother or a store clerk; categories of human beings, such as friends or enemies; institutions, as a school or a government; guiding ideals, such as individual independence or honesty; activities of others, such as their commands or requests; and such situations as an individual encounters in his daily life. The second premise is that the meaning of such things is derived from, or arises out of, the social interaction that one has with one's fellows. The third premise is that these meanings are handled in, and modified through, an interpretative process used by the person in dealing with the things he encounters" (Blumer, 1969, p.2) An important concept in the symbolic-interactionist tradition, developed to compensate for its initial orientation towards micro-level, face to face phenomena, is that of the social wholes, or rather social Worlds. A social world is a collective of individuals with shared commitments, who act in relation to one another, and who gather in particular sites to perform at least one primary activity, using a technology (means of carrying out the relevant activities) and gradually evolving formal organizations. Social worlds are highly fluid, constantly changing due to conflict, competition, negotiation, or exchange (Strauss, 1978). The social world perspective dealt intensively with the boundaries between social worlds, and their intersection. These intersections are related to the existence of arenas: spaces where practical issues are negotiated and fought over by representatives of the various social worlds and sub-worlds; their dynamics are usually those of struggles over power, authority, and legitimacy. As I demonstrate throughout this study, understanding Wikipedia as such an arena is a highly productive move. Indeed, Wikipedia can be seen as an extremely important arena, whose uniqueness lay not only in the wide variety of social worlds potentially related to it, but also in the ease and immediacy in which members of worlds interested in their representation can make it their arena. The results of these conditions required Wikipedia to produce innovative mechanisms to regulate and contain the dynamics of these encounters. Especially important in this respect is the interactionist research done in the domain of science and technology studies, and especially the framework developed by Susan Leigh Star (Star and Griesemer, 1989, Bowker and Star, 1999, Star, 2010). However, as this work is related to a different strand of research also relevant to this study, namely that of social studies of science and technology, I will now describe the context for these developments in this latter field. 33 1.3.3 Science and Technology Studies The last thirty years saw a tremendous increase in science and technology studies, and a major shift in their theoretical orientations. These new studies contested both the traditional internal understanding of science as an autonomous phenomenon developing according to its own logic and the macro-sociological and externalist analysis of the sociology of knowledge, which held that it should be explained against the backdrop of “social” factors.49 Such an approach, based on the distinction between society on the one hand, and science and technology on the other, was clearly problematic in a society where the two were increasingly enmeshed. Thus, both the traditional sociology of science developed by Robert K. Merton50 (which refrained from externalist explanations of the actual methods and content of science) and David Bloor's51 "strong program" (which attempted to attribute the success of scientific theories to external social factors, mainly in terms of group interests) were rejected. In their place came various constructivist and contextual currents, such as laboratory studies, social studies of scientific knowledge (SSK), science and technology studies (STS) and science, technology and medicine studies (ST and MS) – all of which attempted to study the production of technology and scientific knowledge "from within," as an ongoing achievement– a project which retraced many the steps taken by pragmatist philosophers. While most of these currents were not informed by pragmatist philosophy or symbolic interaction initially, they were, as 49 Thus, Marx explained that "the ruling ideas are the ideas of the ruling class" (Marx, 1913, pp. 11-12), Durkheim explained the emergence of perception categories as a result of social processes (Durkheim and Mauss, 2010 [1903]; Durkheim, 1995 [1912]) Weber (1930) illustrated how Protestant ethics gave rise not only to capitalism, but to the growing rationalization of modern societies, and Mannheim contended that "mental structures are inevitably differently formed in different social and historical settings" (Mannheim, 1936, p. 238), focusing on the effects of differences in class and generation. 50 Following Weber, Merton (1973) pointed out the Puritan origins of modern science and went on to offer a comprehensive account of the "scientific community" as a social system complete with its norms (namely, Communalism, Universalism, Disinterestedness, Originality and Skepticism ), institutions (such as the peer-review system), gift-exchange like reward system, social organization and stratification. The thrust of criticisms against him focused on his claims regarding the role of norms in science – which now interpreted as a "justificatory ideology" (Mulkay, 1976, p. 654) – and his assertion that the development of scientific methods and knowledge themselves lie beyond the realm of social research. For a discussion of attitudes towards Merton in the new sociology of science, see H.M. Collins (1982); Gieryn (1982a, 1982b); Knorr-Cetina (1982); Krohn (1982); Mulkay and Nigel (1982). 51 Bloor (1976, 1986; Barnes, Bloor and Henry, 1996) criticized Merton's sociology of knowledge for exempting the actual content of scientific knowledge from social inquiry, and claimed that sociology of knowledge should be based on a principle of symmetry, which posited that all knowledge claims – whether true or false – should be explained through social structures and interests. This "strong program" was later criticized for its own lack of symmetry, as it took a realistic-absolutist position with regards to social categories and objects (classes, power relations etc.), while taking an opposing relativist position regarding scientific categories and objects. 34 Shapin noted, ”wholly compatible with pragmatism, and, by extension, with strands of academic sociology-those of Mead, Blumer, and their progeny-that drew inspiration from James and Dewey" (Shapin, 1995). While highly divergent in their orientation, these approaches share a commitment to describing and explaining concrete practices in concrete sites of knowledge production through both ethnographic "laboratory studies”52 and historical accounts53 (For a general review see Shapin, 1995). They contest the traditional distinctions between "the social" and the "scientific" or "technological," and instead strive to account for the concrete relations between all the relevant entities, whether human or non-human, without attributing primacy or agency beforehand. One such approach that guided me through my research was Karin KnorrCetina's constructivist and contextual framework to understanding knowledge production. One of the pioneers in laboratory studies (Knorr-Cetina, 1981), KnorrCetina went on to study sites of knowledge production with some properties that are very similar to Wikipedia's, such as the CERN laboratory for particle physics, global financial markets, and international terrorist networks. These fields are characterized by collaboration on a massive scale by geographically dispersed participants, a lack of a clear and explicit hierarchy, decentralized authority structure and a reliance advanced computation and communication technology. In her writings, Knorr-Cetina developed several concepts that will be useful in understanding Wikipedia as a knowledge production project. One such concept is that of epistemic cultures, developed against the prevalent view of a science (or the scientific community, or method) as a single, unitary system of knowledge production. In her seminal book, Knorr-Cetina (1999) claims that science studies should distinguish between different modes knowledge production prevalent in different scientific fields. Each such field has its own epistemic culture – defined as those "amalgams of arrangements and mechanisms – bonded through affinity, necessity and historical coincidence – which in a given field, make up how we know what we know" (Knorr-Cetina, 1999, p. 1, italics in the original). Such 52 See, e.g. Knorr-Cetina (1981, 1999), knorr-Cetina and Mulkay (1983), Lynch (1985), Latour and Woolgar (1986), Latour (1987), Pinch (1986), Traweek, (1988). 53 See, e.g. Harwood (1993), MacKenzie (1981, 1990), Pickering (1984), Schaffer (1996), Shapin, (1994, 1996), Sahpin and Schaffer (1985), Warwick (1993). 35 cultures have distinct patterns of authority and organization, different relevant objects and subjects of knowledge, and different mechanisms mediating between them. Understanding the construction and institutionalization of such a culture in Wikipedia is one of the primary goals of this work. This may help shed light on the processes by which "amalgams of arrangements and mechanisms" are bonded together in practice – an issue Knorr-Cetina hardly addresses. Another concept important for understanding Wikipedia's social makeup is that of global microstructures, "forms of connectivity and coordination that combine global reach with microstructural mechanisms that instantiate self-organizing principles and patterns" (Knorr-Cetina, 2005). Global microstructures are conceptually opposed to formal organization, as they are based on features characteristic to face to face interactions such as trust and acquaintance, rather than on clear structure and hierarchy. Lacking formal organizations' rational structuring of social relations, they are inherently unstable and unpredictable. Despite this semichaotic character, these structures tend to be highly effective and sustainable thanks to extensive use of strategies of amplification and augmentation, which allow them to mediate conflicting goals and magnify the outcomes of a given effort. This is done both by using technological resources, such as computers or digital communication, and by using social mechanisms, such as outsourcing certain functions to the external environment, thus remaining structurally "light" and more adaptable. In my research, I attempt to uncover the construction of such strategies, focusing on scopic media, "centering and mediating device[s] through which things become assembled and from which they are projected." Such technological aids help coordinate actions in the global microstructure: "like an array of crystals acting as lenses that collect light, focusing it on one point, such mechanisms collect and focus activities, interests and events, and project them in identical fashion to dispersed audiences" (Knorr-Cetina, 2005). A second approach, or rather a broader current which guided me is the ActorNetwork Theory (ANT), developed by Bruno Latour, Michel Callon and John Law. According to ANT, the world is comprised of an array of actors (or actants, in Latour's terms), both human and nonhuman, material54 and conceptual, which 54 In one of his famous articles Latour (1992) analyses the door closer as an actor. 36 generate sociotechnical networks. Actors are constantly formed and defined through their interrelations with one another. They have no constant, stable and immanent properties outside these relations. Actors build diverse relationships with each other, thus influencing each other's interests and action. In this process, various actors with divergent viewpoints and interests are brought together and enlisted through processes of translation so that their interests and character will enable cooperation. The network durability and stability is determined by it density, the number of elements and associations it encompasses. The more durable it is, the harder it is to doubt it – in a sense, the more essential it becomes. In this account, knowledge is produced through chains of translation which move objects and concepts across the network, each stage making the object either more "material" or "conceptual." These stages are always reversible, allowing objects to be reexamined, reconceptualized, and restructured. The durability of knowledge is thus a function of the integrity of the chain as a whole – cut any link, and the whole array collapses, preventing knowledge from flowing through it (Latour, 1999). When the attempts at enlistments and translations succeed – usually through trials of strength against other possible networks (Yonay, 1994) - and a strong, stable network is created, one may speak of the creation of a new object, or “fact.” In this stage, movement in the network is so smooth and obvious, that it becomes opaque, creating a blackbox, or institution. The attribution of something as "knowledge" thus designates the institutionalization of a network that is so dense that it can be taken for granted. All actors are not equal in this process: at times, certain actors succeed in positioning themselves as "Obligatory Passage Points" (OPPs), through which all actors have to pass through to achieve their goals. They thus succeed in becoming "spokespersons" and representing others,55 funneling the actor-network and aligning the actions of all the other relevant actors (Callon, 1986, Latour, 1988, p. 43, 2005, p.31). This model was later extended by Susan Leigh Star (Star and Griesemer, 1989, Bowker and Star, 1999, Star, 2010), who connected the social studies of science and technology with a symbolic-interactionist social world's perspective (see above). Star 55 Representation here is used in the widest sense of the word: it can refer to just as well to a politician representing a group of people through his "leadership", a scientist representing objects through signs, or an example representing a larger phenomena 37 moved the focus to situations where several social worlds have different claims to the object of studies, thus preventing the creation of a single OPP. In this situation, "several obligatory points of passage are negotiated with several kinds of allies".56 The primary results of these negotiations are methods standardization, the coordination of the activities of various allies to make their products compatible through a common standard, and the creation of boundary objects.57 Boundary objects are "[scientific] objects which both inhabit several intersecting social worlds…and satisfy the informational requirements of each of them… [they] are both plastic enough to adapt to local needs and the constraints of several parties employing them, yet robust enough to maintain a common identity across sites" (Star and Griesemer, 1989: p. 393). The notion of boundary objects was further developed by Kathryn Henderson (1991), who analyzed Engineering sketches and drawings, claiming that they serve both as boundary objects and as conscription devices, which enlist and organize group participation.58 Following Star, I emphasize the interplay between policies enabling method standardization and boundary objects related to them. However, while in Star's model the actors involved in their creation represent distinct social worlds, the participants in Wikipedia's case overtly refused representation: they are more individualized, their relevant identities are harder to pin down, and their constant influx demanded everrecurring acts of translation. These challenges were compounded by Wikipedia's spatial make-up: Wikipedia was (and to a large extent, still is) a network of text pages, each dedicated to an encyclopedic subject. Its radical collaboration process meant that in principle, at least, alignment had to be achieved regarding each individual page, as each was a boundary object in itself, edited by various actors with potentially conflicting interests. This situation created in Wikipedia an epistemic culture that is even more flexible and polysemic than the ones described by Star and her co-writers. Having no OPP's that can commit a large amount of allies to a 56 Another important aspect of this move is the attribution of agency mainly to human actors who do the actual negotiations, rather than distributing it equally to humans and non-humans. See below, chapter 2. 57 The term has since been applied in a wide range of topics, and became prevalent in the field (cf. Flichy, 2007a, 2007b; Fujimura, 1992; Henderson, 1991). 58 The concept of conscription device is very close to Knorr-Cetina's (2005) conceptualization of scopic media. However, while the later refers to a wide variety of coordination devices based on common visibility, the term "conscription device" refers specifically to devices bridging between different social worlds, designating them as boundary objects. In this work I use both terms, preferring in each case the one most suitable to the circumstances. 38 common pattern of action or generate stable methods standardization, it had to generate softer measures of social coordination, less authoritative and strict than those found in scientific contexts, but still effective enough for the project to function59. While the above-mentioned theoretical approaches will serve as the basis of my study, they lack an element which I find extremely important: the connection between knowledge and moral issues. The production and reception of information involve serious moral issues, which makes some networks more plausible than others, by virtue of their alignment or Interessement with the moral preferences of relevant actors. Such preferences play a very important role in knowledge production, as they set standards that help determine what practices and object may be considered legitimate (or rather, legitimable). I thus turn to discuss French pragmatic sociology, which helps address such issues. 1.3.4 French Pragmatic Sociology The new French pragmatism (or "pragmatic sociology") was formulated by Luc Boltanski and Laurent Thévenot in the eighties, as an opposition to Bourdieu's dominance in French sociology, and further developed in the following decades along with other researchers.60 It is inspired both by philosophical pragmatism and by Garfinkel's ethnomethodology, in itself closely related to pragmatist assumptions (Emirbayer and Maynard, 2011). While description of this school usually focus on their analysis of "regimes of justification" (Boltanski and Thévenot, 2006) in the context of this work, its most relevant aspect is its analysis – especially by Laurent Thévenot – of pragmatic regimes governing the engagement with the world (Thévenot, 2001a). Emphasizing the importance of practice and social action rather than social structures, Thévenot claims that human beings constantly need to shift their mode of engagement with the world. Environment, by this account, is not a "given," mute background, but rather a dynamic and multifarious reality whose relevant aspects are 59 This preference was also related to a broader cultural context: that of the egalitarian and extremely liberal ethos of the hacker culture, influential in the relevant social milieus (Lih, 2009). 60 Notable works in this tradition include, among others, Boltanski and Chiapello (2005); Boltanski and Thévenot (2006) and Lamont and Thévenot (2000). 39 dependent on actors’ mode of engagement, which are at the same time resistant and changing. Engagement with the environment is governed by pragmatic regimes: "social devices which govern our way of engaging with our environment inasmuch as they articulate two notions: a) an orientation towards some kind of good; b) a mode of access to reality" (Thévenot, 2001b). Thévenot posits three such regimes: familiarity, regular planned action, and justification. The regime of familiarity is based on an aspiration towards personal and local convenience in a familiar milieu. It is completely individualized, as actors engage objects in idiosyncratic ways, attempting to adjust them to their bodily existence and gestures. The regime of regular planned action sets in as soon as other actors engage the first: it is based on an aspiration towards successful conventional action, in which all actors can function in a "good working order." The regime of justification is the most formal, and the most studied of the three: it is based on an aspiration towards collective conventions of the common good. It is used when a dispute erupts, and regular planned action breaks down. In such cases, actors deploy one of several orders of worth, all of which have their own conception of the good to be aspired.61 Each of these orders measure worth differently, and each has its own standards to evaluate people, objects and practices. Thévenot's analysis of the way orders of worth are negotiated is of central importance to this work, as it focuses on critical situations (Thévenot, 2001a, 2002a, 2002b): situations that have no predetermined order of worth, as was Wikipedia in its early history. Critical situations are those that are not subject to a single mode of evaluation and require social negotiation of its overarching principles of coordination. Thévenot suggests that coordination in such situations is heavily dependent on a process of form investment – that is, creating equivalences across time and space between various practices, objects, persons, etc (Thévenot, 1984). Its successful completion enables divergent parties to cooperate in such critical situations, and eventually surpass them. The importance of this analysis lies in its emphasis of external cultural and moral factors, incorporated to the situation through this process in the form of preexisting cultural models and justifications. However, in contrast to Thévenot I 61 Boltanski and Thévenot (2006) originally describe six such orders of worth (or ‘cités’) – the inspirational, the domestic, the civic, the cité of fame, the market, and the industrial cité (later, other regimes, such as “the regime of the project” were added). 40 emphasize the importance of habitual activity even in such critical situations: following the original pragmatist school, I observe the ways in which habit and creativity alternate in processes of problem-solving and institutionalization (Joas, 1993, 1996; Gross, 2009). Moreover, while Thévenot addresses general modes of coordination or orders of worth, my focus here is on much more localized cultural models and modes of coordination62, focused primarily on the moral aspects of knowledge production and consumption, encapsulated under the notion of epistemic virtues (Wood, 1998; Daston and Galison, 2007): “norms that are internalized and enforced by appeal to ethical values, as well as to pragmatic efficacy in securing knowledge” (Daston and Galison, 2007, p. 40). While Daston and Galison may use a language that is too normative for my taste63, their analysis of the historical succession of epistemic virtues informing scientific representation is highly compelling, as they also show how different and contrasting epistemic virtues are constructed and coexist. It serves as a reminder that knowledge production has its own field of moral concerns, intimately intertwined with practical ones. Thus, I explore how Wikipedia's producers and consumers cope with this combination of practical and moral concerns by constructing or implementing relevant epistemic virtues. 1.3.5 The Gift of Knowledge and Contemporary Culture Having outlined the general theoretical consideration that guided me in my research, I now go on to briefly present two additional theoretical issues to be addressed in this work: first, the question of social embbededness in gift practices; and second, the relations between contemporary information society and modernity. 1.3.5.1 Wikipedia and the Gift of Knowledge Wikipedia success in harnessing voluntary, un-remunerated contributions highlights the connection between knowledge and gift-giving. This is far from innovative, as this 62 For an analysis of the embbededness of models in a concrete epistemic culture, see Breslau and Yonay (1999), Yonay and Breslau (2006). 63 Mulkay (1976) offered the paradigmatic critique of this stance. 41 connection was present since at least the ancient Greeks, and the more so since Enlightenment, which accorded knowledge an emancipatory quality and saw its dissemination as a moral obligation inherent to the virtues of "men of letter" (Foucault, 2001; Hénaff, 2010a; Lyotard, 1984). However, Wikipedia's success in developing innovative and unorthodox gift practices where its predecessor Nupedia failed, suggests that the ability to harness this moral obligation to the creation of a free online encyclopedia is far from self-evident. In what follows, I outline the literature regarding gift giving in general, and specifically regarding knowledge-giving, relevant to explaining this difference. Mauss's (1990 [1925]) seminal essay on the gift presented it as a complex phenomenon, riddled with internal tensions such as interestedness/disinterestedness, freedom/obligation, reciprocity/competition, etc. However, this attention to the gift's complexity was coupled with a relative neglect of the gift's diversity, as Mauss emphasized the basic similarities in its varied historical and expressions (Silber, 2007). This almost essentializing outlook is most clearly manifested in Mauss's insistence on the three obligation involved in the gift (to give, to receive, to reciprocate), and especially in his emphasis on the universality of reciprocity. While this emphasis had a significant effect on later conceptualizations of the gift, the last few decades saw writers challenge this view, claiming that retribution is not a necessary part of the gift's pattern of action, and that our view of the gift should be more open to historical and cultural variations. Mauss's emphasis on the importance of reciprocity was immensely influential in the social sciences. This direction was taken to its extremes in the social exchange theory (Blau, 1964; Homans, 1958), in Levi-Strauss' (1987) structural anthropology and in Bourdieu's genetic structuralism (1977, 1998)64. In contrast with this emphasis on reciprocity and exchange, recent years had seen a growing number of writers who highlighted unilateral, non-reciprocal forms of giving where the obligation to reciprocate played a lesser part, or was absent altogether. Following Parry's (1986) research on the ideal of a "pure" gift in Brahmanic Hinduism, which described unilateral gifts in the context of transcendental 64 While Bourdieu's theory is considered more critical than the others, he followed the same trail as he supplemented Levi-Strauss' structural theory of gift exchange with an emphasis on its temporal structure, intimately linked to its strategic aspects and the hidden interests underlying it. 42 religions, much of the subsequent writing on the subject focuses on the sacred or religious aspects of such gifts (Godelier, 1999; Laidlaw 2000; Silber, 1995, 2000; Tarot, 2000). However, Titmuss (1971) stressed a similar theme in his discussion of the modern and predominantly secular phenomenon of blood donations as a unilateral “gift to strangers.” Hénaff (2010a, 2010b) also discusses the "gracious giving," a gesture "without any expectation of reciprocation and without any association with a situation of scarcity," without connecting it necessarily to religious themes. Similar tensions regarding the role of reciprocity in gift giving are also reflected in the burgeoning literature about knowledge production in general, and more specifically about FOSS production. The literature about gifts in knowledge production settings such as science and the academia traditionally emphasize interest, reciprocity and exchange. In this context, the scientific community is constructed as a gift-exchange system (or gift-economy), a feature apparent both in the gift-relations involved mentoring of new scholars (Cronin, 1995; Savolainen, 1999; Hénaff, 2010a), and in the gift economy of scientific publications (Berthoin-Antal and Richebé, 2009; Hagstrom, 1982). The latter phenomenon was used to contrast academic life and the market, as scientists are not paid for their publications, but rather rewarded with recognition and prestige, often expressed through citations65. Thus, the concept of gift economy in this context is almost always linked with personal interest in prestige and acknowledgement. Similar claims were made regarding other knowledge intensive production environments, such as the high-tech industry (Barbrook, 1998). Closer to our subject matter, the concept of gift economy became popular in studies of peer-production (Benkler, 2006) settings, such as online communities (Barbrook, 1998; Rheingold, 2000; Jenkins, 2006), open-source projects (Bergquist and Ljungberg, 2001) and file sharing networks (Giesler, 2006). Many of these studies stress the communal context of such practices, claiming that acts of giving online usually follow the pattern of generalized exchange, where reciprocation comes from the community as a whole rather than to the specific beneficiary (Bauwens, 2005; Bergquist and Ljungberg, 2001; Giesler, 2006; Jenkins, 2006; Kollock, 1999; Rheingold, 2000). Some posit the act of giving as the result of values and norms prevalent in the immediate social context of the actors, often described as a "gift 65 However, Hagstrom also notes that prestigious professors still publish new works, attributing due to their commitment to the scientific community. 43 giving-culture," which constructs sharing and openness as virtues (Kollock, 1999; Benkler, 2011). Other writers focus more on the individual interest of the actors, stressing the importance of expectation for returns or social acknowledgement and prestige (Bruns, 2008; Raymond, 1998; Tapscott and Williams, 2006). However, both sides agree in their view of the act of gift as embedded in some form of exchange, whether generalized or restricted, and on the interest of the actors in the act of giving – whether articulated in individualistic or communal terms. Thus, both sides generally reject the notion of a free or pure gift, and stress the importance of symmetry and recognition in the gift. This connection presumes the relevance of a consistent social identity of social actors, who can gain recognition and reputation from this environment. At the same time, other conceptualizations of peer production or produsage (Bruns, 2008) are closer to the idea of the "pure" or "free" gift. Such conceptualizations downplay the importance of community, reciprocity, and recognition in online giving, and emphasize rather broader cultural factors and intrinsic motivations for giving (such as ideology or fun), and help address the prevalence of anonymous giving in online contexts, a phenomenon which defies explanation relying on the relevance of recognition. Some writers stress the non- or even anti-utilitarian nature of such contributions in online communities or FOSS projects, (c.f. Barbrook, 1998, 2000; Rheingold, 2000; Bauwens, 2005), while others merely emphasize the free, de-centered and individualized (or networked) aspects of online giving and sharing (Benkler, 2006; Bruns, 2008; Lessig, 1999; Shirky, 2008, 2010; Surowiecki, 2004; Tapscott and Williams, 2006). While this emphasis does not necessarily preclude the existence of individual or communal interest in the act of giving, it does deny the centrality of individual identity and belonging. So how are we to approach these seeming complexities, variations and contradictions? One productive way may be to address them as expressing the simultaneous existence of different modes or dimensions of giving. Indeed, such move has been highly influential in studies of the gift. In recent decades, research had moved from trying to explain away its complexity to rather trace it diversity. This move is at the heart of what Silber (2007) calls "the typological turn": the appearance of studies such as Vandevelde (2000), Caille (2001), Graeber (2001), or Hénaff 44 (2010a, 2010b) who suggest types or dimensions of gift-giving practices as categories of analysis. However, as this typological orientation still generally fails to address the unique combinations of forms and practices of giving in specific contexts, Silber suggests the idea of a gift repertoire (Silber, 2007): "the available store of ideals, practices, institutions, priorities and dilemmas, foundational stories, ideological and even cosmological theories of giving coexisting and made available in a specific historical and cultural settings. Such repertoires may vary not only in the options which they provide for thinking, practicing and contesting giving but also in their very scope, richness and internal diversification". Thus, this work aims at demonstrating how attempts at creating a free online encyclopedia took into account and at the same time shaped such a repertoire of acts of knowledge giving. The concept of a repertoire of gift practices may prove useful in attempting to understand Wikipedia's success and Nupedia's failure: rather than focusing on the motivations of contributors, it opens the perspective to the wider context of their actions, shaped at least in part by the project itself. This attempt may also help us understand better how gift repertoires are shaped and constructed in a concrete and limited setting, and the effect of their dissemination. As most studies of the gift tend to analyze it strictly in broad, macro-level contexts such as whole societies or civilizations, this may prove of some contribution not only to the study of Wikipedia, but also to the literature on the gift. Following the review of the literature about giving to Wikipedia described above I focus on distinguishing between two relevant modes of online knowledge giving. Group-oriented giving may be either communal or exchange-oriented (or both), but in any case involves practical manifestation of the shared social context of givers, and the expectation for acknowledgement and recognition, from others, as well as some sort of reciprocity. This mode is dependent on the existence of relatively stable social relations, and personal identities which can be recognized and reciprocated. On the other hand, individual giving is at work when the giver is not engaged in long terms social relations with others related to his gift and/or does not have a consistent social identity which is socially recognized and accumulates reputation. This mode is dependent on the existence of mechanisms for publishing 45 content anonymously to a public which would accept and consume it as knowledge. It also depends on an ideology that constructs such an act as worthy. However, both types of giving should not be viewed as general, abstract types only, but rather as historically and culturally embedded, as both the social structure and the cultural setting in which the act of giving described here is embedded are culturally historically specific. Thus, I now go on to discuss the more general social and cultural setting in which Wikipedia exists – namely, that of contemporary knowledge society. 1.3.5.2 Wikipedia and Conceptualizations of Contemporary Culture The wider implications of Wikipedia's epistemic culture may be better understood against the backdrop of theories analyzing contemporary societies as knowledge societies, or information societies. There is a widespread consensus that in the last few decades, the production of knowledge and information moved to the heart of social, economic, political and cultural structures and processes. Beginning in the 1970s, a plethora of concepts and ideas emerged to describe this transformation, often by contrasting it to its predecessor, namely industrial society. Such concepts include, among others, the postindustrial society (Touraine, 1971; Bell, 1973), knowledge society (Bell, 1973; Böhme and Stehr, 1986), the information age (Castells, 1996, 1997, 1998), information society (Drucker, 1993), postmodern society (Lyotard, 1984), late modernity (Giddens, 1991), reflexive modernity, (Beck, Giddens and Lash, 1994), second modernity (Beck, 2000) and after modernity (Beck and Sznaider, 2010). At the same time, however, there remain highly contentious disagreements regarding the question if change signals a transformation within modernity, or rather a break from it. Some commentators advocate the continued relevance of the concept of modernity66, at least if used to connote an existence of a common cultural center, a sense of historical continuity with the vision of enlightenment, and the centrality of rationality, science and expertise (cf. Beck, 2000; Eisenstadt, 2001; Giddens, 1991, 1994; Habermas, 1984). On the one hand, Other notable commentators, which 66 Such commentators often identified through their use of the terms late or second modernity. 46 maintain that contemporary society represents a break from modernity, with new and dissenting knowledge production structures, processes and legitimation principles may help come to terms with this paradox (c.f. Baudrillard, 1983, 1994; Jameson, 1991; Lyotard, 1984). In the context of the latter body of literature, it has become commonplace to claim that the last decades were characterized by an erosion in the metanarratives of enlightenment, the dominance of cultural centers and the monopoly of experts over knowledge (Best and Kellner, 1997, J. Collins, 1989; Featherstone, 1989, 1991; Jameson, 1991; Lash, 1990; Lash and Urry, 1994). The resultant postmodern culture is characterized by an erosion of the aura (Benjamin, 1968) related to knowledge production and the emergence of a wide range of intermediate social categories which undermine the coupling between knowledge production and expertise by producing unprecedented stocks of signs and representations67. An integral part of this process is the transgression of cultural boundaries, the waning of enlightenment's metanarratives68, and the decline of regimes of signification underlying modern culture69. While this approach can help make sense of some of Wikipedia's unorthodox features such as the de-coupling of knowledge and expertise70, it does not account for those of its basic premises and motivations which are still manifestly related to ideals of enlightenment and modern culture – such as the construction of knowledge dissemination as a moral act. Thus, a more nuanced approach, which stops short of asserting a complete breakdown of modern cultural categories, but which also goes beyond the emphasis 67 Such categories include, among others, the "new cultural intermediaries" (Bourdieu, 1984; Featherstone, 1991; Lash and Urry, 1994), consumption and lifestyle brokers (Appadurai, 1986, p. 54, 1990), the devout computer enthusiasts who played a major part in the development of the computer and internet industry (Castells, 2001); and media fans (Jenkins, 1992; Abercrombie and Longhurst, 1998). 68 Such Meta-Narratives, which include the progress of humanity, the triumph of reason and enlightenment, the liberation of the subject etc. are central to the construction of knowledge production and dissemination as moral acts. See e.g. Lyotard (1984). 69 The extreme versions of this argument go as far as questioning the principles underlying reality itself. Baudrillard (1983, 1994) claims that the accelerated production of signs and information results in an implosion whereby all the dichotomies and boundaries of late capitalism collapse, along with meaning, knowledge, subjectivity, society and reality itself. 70 The coupling of knowledge and expertise in modernity was highlighted by Giddens (1990), who claimed that modern society is increasingly saturated with expert systems which provide ontological security. Wikipedia's success, despite the fact that it cannot generate the very trust that underlies expert-systems, suggests that others social processes are at work here. 47 on continuities with it, may be more promising. Such an approach, advocated by writers such as Laermans (1992), Lash (1994) and Knorr-Cetina (1999, 2005) claims modern culture's centers, distinctions and meta-narratives did not completely disappear. Rather, they are now supplemented by various subcultures, communities and network which flourish at its margins. Such cultures are increasingly global in scope, and their participants act as local experts or prosumers (Toffler, 1980. For a critique of the term see Comor, 2011), as they foster immanent standards, practices and justifications. Such standards and practices have varying degrees of affinity to modern culture, but in any case are local in character, failing to achieve widely acknowledged legitimacy (Laermans, 1992; Lash, 1994). Their expertise is embedded in an intersubjective, micro-level trust, rather than formal standards or organizations. Such structures are increasingly (though not exclusively) built in and through the internet, as it allows people around the world to transcend their immediate environment and create the mediated intersubjectivity that they require. Knorr-Cetina (1999, 2005), which studied several such cultures, stresses the important structural and "textural" differences between them, as well as the great variability in their relations with modern cultures71. Wikipedia seems to fit very well with this type of theories, as it depends on internal standards and intersubjective mechanisms, and is similar in many of its aspects to other subcultures studied in this theoretical context. However, Wikipedia occupies a unique position it this respect: while this approach stresses the blurring between consumers and producers and the importance of internal or local standards and practices, Wikipedia's success involved reaching diverse crowds for whom these standards are external and readymade. Thus, assessing the extent to which specific form investments72 made by Wikipedia's participants correspond with central tenets of modern culture may help assess its viability and relevance in contemporary information society. 71 In her empirical studies, encompassing diverse topics such as high energy physics (Knorr-Cetina, 1999), financial markets (Knorr-cetina and Brugger, 2000), and terrorist networks (Knorr-Cetina, 2005), she focuses on the use of information technologies for social coordination, and the temporal aspects of this coordination. 72 As noted above, critical situation where standards and practices are unclear demand "form investment" in prevalent cultural models to coordinate expectations and actions. 48 1.4 Research Objectives The production and institutionalization of a new epistemic culture in Wikipedia and its dissemination, processes previously unstudied, are the focus of this research. In the context of production, I describe how various people, technologies and cultural models were amalgamated to bring forward an innovative way of constructing and legitimizing knowledge in Wikipedia. In the context of consumption or use, I describe how Wikipedia is used by a specific group of knowledge experts – namely journalists – and how they incorporate and put into practice the epistemic mechanisms underlying it. In other words, the objective of this research is to examine: how was Wikipedia's unique epistemic stance formed, and how is it reflected in its everyday, practical use? After the next chapter, in which I describe the methodology used to answer these questions, the first empirical chapter (chapter 3) describes the concrete environment in which Wikipedia was born, embodied in its predecessor, Nupedia. Founded by the same team – Larry Sanger and Jimmy Wales – Nupedia was an open internet encyclopedia project, which was based on production by experts and a rigorous production process. Its epistemic culture was based on two strong Obligatory Passage Points (Bijker and Law, 1992; Callon, 1986; Latour, 1988): the Encyclopaedia Britannica, which was constructed as the authoritative standard for its content; and the ethos of expertise, coupled with a reliance on formal academic credentials, as the basis of the projects authority structure and production methods. While its production process incorporated innovative elements, such as the use of an "open review" process in which anyone who wanted to could critique articles, it ultimately remained highly conservative in its epistemic assumptions and methods. This approach, rooted in an attempt to endow the project with traditional legitimacy (Weber, 1978, pp. 266-231), proved ill-suited, as production pace was excruciatingly slow. Attempts to alter it failed, as the Obligatory Passage Points erected in it prevented any significant changes in the production process. Thus, when Sanger and Wales learned about the Wiki technology, the decided to start an independent project, named Wikipedia, free from the constraints exerted by Nupedia's 49 OPP's. Wikipedia thus began without an overarching epistemic culture. The creation and institutionalization of this culture is the subject of the next three chapters. The next chapter (4) is a relatively brief chronology of Wikipedia's founding period in its first year or so, which outlines its emergence and institutionalization as a knowledge production project. In that, it also lays the ground for the next two chapters, which focus on events occurring in the described period. The next two chapters describe the institutionalization of standards and policies governing the production of articles in Wikipedia. Such policies were socially constructed and institutionalized through a relatively regular pattern of actions, involving a conflict or controversy over the question what is the legitimate conduct in a certain case; the formation of a more or less binding general policy relevant to other hypothesized cases; subsequent conflicts over the application (or at times even legitimacy) of this policy; the institutionalization, through the enactments of power and authority, of a policy which is regularly re-articulated and expanded; and the development of social or technological enforcement mechanisms to uphold it. They thus became Privilege Passage Points in Wikipedia, governing effectively, though not completely, its production processes and epistemic mechanisms73. Chapter 5 focuses on the creation of radically collaborative mode o production in Wikipedia, and the construction of two Privilege Passage Points that were essential in fitting it to the established model of encyclopedia articles. The first of them created a special category of Wiki pages referred to as article or subject pages, where writing is to be styled after traditional encyclopedia articles. It also specified that discussion of articles should not take place in the article pages themselves. This policy was the most essential in domesticating the Wiki technology as it set a precedent that content policies should be based on traditional encyclopedic standards, even if those contrast with standard Wiki operation. The second major content policy, neutral point of view, was created originally as means to enable the joint work of participants with conflicting views; it dictated that that the opinions of all parties to the disputes should be presented in the article in a manner they can agree on. Later on, it transcended its 73 As opposed to Obligatory Passage Points, “loc[i] that could shape and mobilize the local network” and “have control over all transactions between the local and the global networks" (Bijker and Law, 1992, p. 31), Privileged Passage Points have less power and allow for greater variability, though they succeed in effectively controlling the network in most cases. 50 initial context and was extended to comprise a general standard for representing knowledge in the project. Chapter 6 picks up the theme of the previous one, but focuses on the institutionalization of policies establishing standards for the inclusion or exclusion of content, rather than collaborative production. The first of them stated that Wikipedia is not a dictionary, was responsible for affixing the notion that Wikipedia's goal is more specific than producing an innovative reference work, or a general information resource. Wikipedia was constructed through it as an encyclopedia, first and foremost, with unique characteristics derived from the history and conventions of the genre. This articulation paved the way to further policies dealing with types of content not prevalent in traditional encyclopedias. These policies were grouped to form the second such policy, What Wikipedia is not, which became a general standard distinguishing between legitimate and illegitimate content. Setting the standard for legitimate content via negativa, it was a readymade tool for an orientation of new editors on content policies, and was quickly adapted to answer this need. As time went by, new types of content were gradually added to it, as they appeared in Wikipedia and were then de-legitimized as anti encyclopedic. The policies outlined in chapters 5 and 6 became the backbone for Wikipedia's epistemic governance, and shape till this very day the implicit contract between Wikipedia and its users. They construct Wikipedia's users as active, critical and engaged participants, rather than passive consumers, offering them all available information regarding the article's production and possible disagreements or conflicts regarding it. The construction of its consumers as competent judges allows Wikipedia to demand of them to individually assess the quality of its articles, which are deemed inherently suspicious in terms of their authority. Chapter 7 sets to explore the everyday manifestations of this mode of consumption through interviews with Israeli journalists about their use of Wikipedia. Based on these interviews, I constructed a model of journalistic use of Wikipedia comprised of five stages: Problem Google search Consulting Wikipedia Triangulation Re-articulation. In this process, the journalist moves from her immediate practical needs and environmental constraints following a chain of references (Latour, 1999) leading to a hypothesized origin of information, until she 51 reaches a point where following further references becomes impossible or unpractical. In this process, Wikipedia serves as a Privileged Passage Point – a practical shortcut to knowledge that originated elsewhere. Over time, however, It gradually tends to develops into a practical end-point in this process, as users tend more and more to rely on the validity of the information it contains, based on evidence found in its articles that further valid references were involved in their compilation. As users increasingly tend to suffice with information found in Wikipedia articles, consuming them thus involves judgmental and critical practices meant to establish the validity of the sought after information. The result is a mode of consumption based on practical legitimacy: a form of legitimation which grants knowledge (or objects, or persons) a diminished yet observable measure authority, as it takes into account and encapsulates both its usability and its liabilities, both its de-facto legitimacy and lack of de-jure legitimacy. 52 Chapter 2: Methodology To account for the institutionalization of Wikipedia's epistemic culture, I used two distinct corpuses, one addressing its internal formation and the other its external implications. The first corpus was assembled by retrieving relevant data from web archives containing almost all the actions and interactions conducted within Wikipedia's founding period of 2000-2002. Out of this data, I reconstructed the controversies and social dynamics that led to the institutionalization of various central features of Wikipedia's epistemic culture. The second corpus was comprised of twenty interviews conducted in 2010 with knowledge retrieval and production experts – i.e., journalists – about Wikipedia and the way they use it. The data gathered was used to assess how this emergent structure was implicated in the lives of its users, effecting society at a larger scale. The methodological orientation of this work is informed primarily by studies related to the tradition social studies of science and technology. As noted earlier, such studies attempt to trace the concrete interactions and associations which occur in sites of knowledge production, and describe how they bring together heterogeneous elements which together shape its methods and content. Specifically, I am most influenced by the methods developed in the ANT school. ANT is notoriously ambiguous about issues of methodology: Latour (2004) is famous for his directive "just describe". However, it does have some methodological imperatives, most explicitly manifested in the "cartography of controversies" approach, which is a "practical" version of ANT. Its most systematic account was articulated by Tommaso Venturini (2009), on which I rely in its description. A controversy, in the language of ANT, is "every bit of science and technology which is not yet stabilized, closed or 'black boxed'" (Macospol, 2007, cited in Venturini, 2009). Controversies tend a number of common features, and their research is supposed to reflect them74: First, controversies involve heterogeneous actors – human, social, cultural, natural, technological, etc. the research should reflect this 74 The list below is somewhat different from the articulation appearing in Venturini (2009), reflecting my personal emphases and preferences. 53 heterogeneity rather than reduce it to one "explaining" variable. Second, they are extremely dynamic, with ever-changing composition, and ever shifting relations and alliances. The research must not posit given, predetermined "subjects," as any network of objects can congeal into an actor, and any actor can dissolve or mutate. It is the controversy and its evolvement that should stand in the heart of the study. Third, they involve conflicting articulations – and thus force the researcher to be attuned to local definitions of the situation, rather than forcing his own. Fourth, controversies are essentially conflictual events, as different articulations clash, merge, and mutate to fit or contradict each other. Thus, social order and social hierarchies are always implicated in such situation – not as determinants, but as results or achievements (see Latour, 1986). The study should focus on moments of conflicts and their resolution, not assuming that their outcome is determined by preexisting "social power," but describing how they are manifested concretely in the situation – via translations, compromises, coercion etc. Controversies usually share a direction, or a common drive: that of reducing complexity of interactions and creating stable entities. In this sense, observing controversies is watching a world in the making. The researchers should thus be attuned to processes where boundaries and distinctions are created, to the processes of congealment which create order from chaos, simplicity from complexity. This approach is extremely useful in the case of Wikipedia, a subject shrouded in controversies – both internal and external – from its inception, but gradually institutionalizing and gaining stability and robustness. Wikipedia satisfies all the criteria of a "good controversy": it is "hot," as fierce disagreements and conflicting visions of the project and of knowledge production frequently light up in it – and did so even more often, and with greater consequences within the project's founding period. While many of the original internal controversies had cooled down, it is possible, through archives and edit-histories, to approach them as if they were current. In addition, Wikipedia is still a controversial issue in the public at large, an issue that will be discussed in the dissertation's final chapter. The controversies surrounding its creation are well bounded, both thematically (the conceptualization and building of a free encyclopedia) and geographically – almost all the controversies related to Wikipedia's inception can be located in two websites 54 (Nupedia and Wikipedia) and two mailing lists (the Nupedia-l and Wikipedia-l mailing lists). Other nods are usually traceable through links and referrals from these sources. This feature also means that the debate was generally open and public, meaning that it can be easily observed. How do we go about researching the controversies regarding Wikipedia? The basic attitude is that of maximizing the effect of the actors, and preferring descriptions that follow their actions rather than explanations that reduce them or interpret them through foreign factors articulated beforehand. Venturini (2009) offers three general maxims: "1) you shall not restrain your observation to any single theory or methodology; 2) you shall observe from as many viewpoints as possible; 3) you shall listen to actors’ voices more than to your own presumptions." To a large extent, these maxims reflect the ideas basic attitude of anthropology, and the idea of "thick description" (Geertz, 1973) – indeed, Latour originally rejected sociology and referred to himself as an anthropologist75 – but back away from it focus on "culture" and the idea of interpretation, promoting instead a "realist" stance which rejects phenomenology and its emphasis on human intentionality. In terms of the actual process of research, the cartography of controversies offers a variety of "lenses" through which the researcher can pick and organize his materials. First, to map the web of references: that is, to trace statements and practices until one feels he reached the boundaries of the arena, and then to organize them into relatively coherent positions and discourses (or literatures) in which they support each other. Second, to locate the actors – that is, any object in the field that produces statements and practices, or has a visible effect on their articulation76. Third, to locate the networks of actors: not as stable entities, but rather as ever changing series of connection which are being tied and untied as the controversies unfold. Fourth, to reconstruct, as best he can, the relevant cosmoi (Venturini, 2009) or panoramas (Latour, 2005, 183-190) – i.e., visions, totalities, ordered and idealized versions of the world – which inform controversies and clash in their process. Finally, to describe the 75 For a discussion of the affinities between ANT and anthropology, see Latour, (1993, pp. 91-109). His reconsideration of his relations with sociology is one of the main goals of Reassembling the Social (Latour, 2005, especially pp. 3-17, 37-42). 76 Actors, it should be remembered, may be of any sort: human (such as a participants in the controversy), natural (such as a phenomena discussed in it), cultural (such as models and practices that serve as precedence or cautionary tale), technological (such as the medium thorough which it is carried), etc. 55 cosmopolitics – the common world emerging from the meeting of different networks and cosmoi, and regulates their relations. These lenses were highly valuable in the research process of this work, and were reflected in the research and writing of this dissertation (see below). I chose to divert from the ANT tradition in two main ways: first, in the attribution of agency and attitude towards interpretation. While ANT emphasizes the need to attribute agency to all actors in the field, I focused almost entirely on the agency of humans. That is not because I disagree with the general point: on the contrary, it is precisely because I believe that this cardinal point in ANT was demonstrated successfully enough and adapted to the point that demonstrating it once again is of no great consequence. More than anything else, this is a stylistic choice. The convention of referring to humans as actors and to objects as implicated in their actions rather than acting on their own comes more easily (but not "naturally") to me. Thus, in my field notes and in my organizing summaries I did just that: refer to human as actors, and while emphasizing the effects of non-human entities such as models and technological artifacts, I did not refer to them as purposeful or intentional, but rather concentrated on how they fit in with human beings and their actions – which they, of course, have a visible and undeniable effect on. In writing my report, I chose to leave this organizing structure as it was – not the least because I believe that in most cases it makes for a better, more coherent read. In a more principled argument, I may add that this orientation is not "natural" – but rather informed by works in the tradition of symbolic interactionism, especially those made in the social studies of science and technology. Such works, while acknowledging the importance of material environment, nonetheless focus on humans and their interactions, highlighting the importance of social (in the narrow sense) definitions of the situation. That preference is closely related to my second divergence: namely, an emphasis of moral concerns implicated in the building of networks. While this is not a clear break with the ANT tradition, it is certainly a shift in emphasis. Closer to the pragmatic sociology point of view, I observed closely the justifications and definitions of the good which was implicated in the relations of (human) actors to others. Attempting to go beyond indentifying cosmoses, I attempted to reconstruct the 56 epistemic virtues (Wood, 1998; Daston and Galison, 2007) which inform and justify the actions of individuals, imbue the form and content of objects, and dictates standards that qualify them (Thévenot, 2001b, 2002a, 2002b; Star and Griesemer, 1989, Bowker and Star, 1999, Star, 2010). To account for the institutionalization of Wikipedia's epistemic culture, I used two distinct corpuses, one addressing its internal formation and the other its external implications. The first corpus was assembled from web archives containing almost all the actions and interactions conducted within Wikipedia's founding period of 20002002. The second corpus was comprised of twenty interviews conducted in 2010 with knowledge retrieval and production experts – i.e., journalists – about Wikipedia and the way they use it. 2.1 Data Corpus 1: Web Archives The first part of the research was comprised of tracing controversies whose decision shaped Wikipedia's epistemic culture, both on the World Wide Web and in mailing lists. By the time the research began, these controversies were, to a large extent, decided and blackboxed, and the relevant discussions were had to be excavated and retraced. By using various web archives, I was able to reproduce with very few gaps the relevant arenas and networks, in a way that allowed me to conduct my research in manner close to a virtual or online ethnography (Markham, 1998, Hine, 2000). As some of the methods I used were innovative, I will describe them in some length. 2.1.1 Assembling the Corpus The first step was to locate the relevant locations where controversies took place, and build a corpus of relevant material. I began my explorations by reading previous research that touched upon Wikipedia's inception (most notably Lih, 2009 and Reagle, 2010a), as well as reports and memoirs written by Wikipedians at later dates (most notably Sanger, 2005 and Wikipedia, 2012d). These resources contained links and references to much of the archives I later used to build my corpus. Reading 57 them, I soon realized that to trace the main controversies at their origin, my research had to go further back in time and include Nupedia, the project from and against which Wikipedia developed. Thus, I used four main archives to construct my corpus: 1) The Nupedia-l mailing list, the main mailing list that was used to coordinate and discuss the project. This archive was constructed by Joseph Reagle and kept at his website (Reagle, 2006). 2) Internet Archives containing several versions of Nupedia's website, accessed through the Internet Archive's Wayback Machine (Wayback Machine, 2011a). The fact that the website was routinely re-archived (though in uneven intervals) allowed me to observe how the resolution of controversies was reflected in it, and shaped its content. 3) The Wikipedia-l mailing list, the main mailing list that was used to coordinate and discuss the project until November 2002. 4) Several archives containing almost all the edits made in Wikipedia: a) Wikipedia's edit histories, saved automatically to the website, documenting each and every change made to its content. They are easily accessible, enabling an almost complete reconstruction of Wikipedia's history. There are some gaps in this history: histories of deleted pages, or pages that were moved or renamed, are missing or hard to find. The greatest drawback of this archive is that as a result of software issues, the edit histories of Wikipedia's first year are almost entirely missing, as are parts of its second year. This lacuna in the documentation of Wikipedia's founding period, which limited the research regarding Wikipedia's institutionalization up till now, was overcome by use of other archives. b) Two archives containing early versions of Wikipedia, the first in the Internet Archive Wayback Machine (Wayback Machine, 2011b) and the other in "Nostalgia Wikipedia" (Nostalgia Wikipedia, 2012), a reconstruction of Wikipedia as it was in December 2001. While far from 58 comprehensive, they contained valuable – though limited – edit histories saved in the periods missing from Wikipedia's edit histories (as they appear on Wikipedia's website). c) A complete, chronologically ordered, backup of all Wikipedia edits from its inception on January 2001 until mid-August of that year77. It contained the time of every edit and the identity (whether in the form of a system-recognized account nickname or an IP address) of its maker, the name of the page edited and the content added or deleted from the Wiki. As the original archive was stored in a huge, barely usable text file, I recoded it to fit the needs of this research: I edited the raw text files it contained to build a Microsoft Excel spreadsheet placing each edit in a separate row (the final file contained 58,018 such rows). This allowed me to both chronologically reconstruct the development of Wikipedia, and sort it according to various features, such as editor or page. A Bibliographical note: As this archive is not easily accessible (or processed, as it appears in the form of a very large Microsoft Excel file), I gave references to the previous sources whenever possible, even in cases when I originally used this archive to uncover the relevant information. In cases when it was not possible, the reference points to the Wikipedia Backups Archive and to the unique timestamp automatically generated for each edit in the archive. This allows the reader to find the relevant location through a simple search in the archive, which can be downloaded from the following URL: http://liorgimel.wordpress.com/2011/05/18/old-Wikipedia-edits- from-2001. 2.1.2 Method of Analysis After assembling the relevant corpus, I reconstructed the main controversies through which Wikipedia was negotiated and constructed, using methods very similar 77 This archive was considered lost due to software issues relating to the Wiki software used at the time, until it was found and extracted by Tim Starling in 2010. See Starling (2010) for the original announcement of the discovery. 59 to those practiced in the mapping of controversies. My analysis of the data included the following stages: 1) Reading the mailing lists and noting in a separate file the main discussions and controversies, thus creating a tentative timeline of controversies. For each discussion, I also noted the relevant links which were mentioned in it, thus extending the network of references to external locations (mainly, but not exclusively, to Nupedia or Wikipedia). 2) Accessing Nupedia's archives and noting web pages that were connected to discussions in the mailing list. These connections were not limited to links, but rather included the contents of the pages. In cases when standards, rules, or practices appearing on a webpage were subject of mailing list discussion, I added it to the files containing the relevant mailing list threads. 3) Accessing Wikipedia's backup of edits histories from its first few months. All controversies or discussions regarding standards, rules, practices or epistemology were saved, each in a different file, with all the relevant documents that were linked to. Discussions that had corollaries in the mailing list were merged with them to form a single file. 4) Arranging and describing the controversies. Using the previously mentioned files, each discussion was arranged and analyzed diachronically. As what I was looking for was a longitudinal process, I analyzed each concrete controversy as a social drama (Turner 1967, 1968, 1974), in which contingent and local interactions determine the outcome78. Furthermore, the diachronic outlook allowed me to observe how each resolution of a controversy conditioned (with varying degrees of success) the outcome of later ones. 5) Tracing the networks. Network of relevant actors were identified for each controversy. As virtually all entries were attributed to specific, identifiable actors and contained explicit references and links to relevant objects, it was relatively easy to construct the networks operating in the field. The networks I observed were highly mobile, realigning themselves to address each 78 In contrast with turner, however, I did not always construct the dramas as results of "cracks" in the social structure, as "structure" is the goal of this work, not its hypothesis. 60 controversy and changing their configurations over time. At times these changes were very minor; at other times they were revolutionary. 6) Following the process of translation. I followed closely the processes of translation and enlistment that took place as the controversies unfold, paying special attention to the visions, models and epistemic virtues implicated in them. 7) Assessing the outcomes. Finally, I observed the results in terms of the emergent epistemic culture (Knorr-Cetina, 1999): the cosmopolitics that emerged and determined how knowledge ought to be assembled and which actors and objects are qualified to partake in its production. 2.2 Data Corpus 2: Interviews The second part of the research was comprised of semi-structured interviews – a highly flexible tool, allowing for new and unexpected issues to emerge. This strategy is based on the assumption that research questions should never be completely closed and that this openness should be reflected in research methods. Moreover, flexibility is essential in attempting to come close to the interviewee's world (Berg, 1998). 2.2.1 Assembling the Corpus The interviews had two main purposes: first, to explore interviewee's perceptions of Wikipedia and its production, with special emphasis given to their attitudes towards its validity and legitimacy as a source of information. Second, to understand if, how and in what occasions journalists used Wikipedia. Originally, I assumed that these two directions would later intersect; I hypothesized that there would be a correlation between interviewees' perceptions of Wikipedia and their use of it. However, after I conducted a few pilot interviews to chart the main aspects of the topic and test my questions and understanding of the field (cf. Kvale, 2008, pp. 46-47) it became clear to me that such a correlation did not exist, and that journalists 61 used Wikipedia regardless of their perception of it and whether they trusted it or not: as all interviewees concurred, consulting Wikipedia it was already a firmly instituted journalistic research method. At the same time, it became just as clear that there was a strict taboo on mentioning Wikipedia as a source. I thus changed the interview to reflect my interest in this apparent contradiction. The interviews began with a general explanation of the research and its aims, and an invitation to ask any question the interviewees may have about it. In its final version, the interviews were comprised of four batches of questions: first, a number of general questions meant to identify the interviewees' position in the field of journalism and their background. The decision to open this kind of questions was not incidental: apart from being informative, they also help build rapport (Alvesson, 2003). The second batch focused on questions regarding the interviewees knowledge about Wikipedia, such as what do they know of the way its managed and constructed, how familiar are they with aspects of the website outside the scope of article pages (such as article history and discussion), etc. The third and largest batch related to journalists' relation with Wikipedia: first in general, i.e. how widespread is the use of Wikipedia in their workplace; when, how and under what circumstances it is legitimized as a research tool in the workplace. Then I moved on to more personal questions: how much they use it, when, for what purposes, how essential it is to their workflow, etc. when setting up interviews, I specifically requested interviewees to try and remember – and write down, if possible – actual occurrences when they use Wikipedia, and a substantial portion of this part was devoted to discussing and analyzing it. In the fourth and final batch, the questions focused on attitudes towards Wikipedia, focusing on its perceived reliability, on justifications for using it, on the legitimacy of using it, etc. I finished every interview by requesting interviewees to talk about how they felt about Wikipedia in general. This question often proved highly fruitful, as interviewees described their views and opinions at length and quite openly. The selection of interviewees was based on maximum variation sampling, a purposive sampling method which "aims at capturing and describing the central themes that cut across a great deal of variation" (Patton, 2002, PP. 234-235). To maximize variation, I picked journalists in various Israeli news venues: I interviewed 62 journalists from Israel four major daily papers, Yediot Aharonot, Maariv, Ha'aretz and Israel Hayom; from two of Israel's three financial newspapers, namely Globes and Calcalist, and three major Websites, Ynet and Walla positions. I also attempted to cover various typical roles in the journalistic world, those of reporters, editors, copyeditors and proofreaders. Finally, I aimed at interviewing journalists in various specialties: domestic news; foreign news; sports; finance; science and technology; culture and entertainment. This variation allowed me to reconstruct shared practices that cut across local or role-dependent differences. To reach interviewees I used a modified snowball method where I asked each interviewee for more contacts, specifically soliciting contacts that fitted the venues, roles and specialties I was aiming for. Twelve of the interviewees were men, and eight were women. Most of them were in their late twenties or early thirties, the youngest bring 25 and the oldest 38. This does not represent the full diversity of the journalistic field, as interviewing older journalists may have produced different result. However, it does represent the generations which came of age professionally at a time when computer and internet technology were already commonplace, or on their way to Become commonplace. All interviews were recorded and transcribed, with the permission of the interviewees. I also took some notes, highlighting certain aspects of the interview for later contemplation. 2.2.2 Method of The Analysis In the spirit of works done in the realm of human-computer interactions (Heath, Knoblauch and Luff, 2000; Guimares, 2005; Rutter and Smith 2005; Turkle, 1995, 1996; Wellman and Haythornthwaite, 2002), I used the data collected in the interviews to understand the way humans (in this case, journalist) construct technology through their everyday practices, and how they interact with, changing both their practices and identity, and the technology itself (or at least adjusting it to their needs). 63 I analyzed the interviewees' descriptions of how they use Wikipedia, to reconstruct the chains of translation which incorporated Wikipedia. From a very early stage, it was evident that these were very similar, and no systematic deviations were emerging. I thus built a model of how journalists use Wikipedia which was gradually refined as more data was added in subsequent interviews, until new interviews made no impact to it. To build it, I used a process of analytic induction (Robinson, 1951; Denzin, 2007), whose steps are: 1) A rough definition of the phenomenon to be explained is formulated. 2) A hypothetical explanation of that phenomenon is formulated. 3) One case is studied in light of the hypothesis, with the object of determining whether or not the hypothesis fits the facts in that case. 4) If the hypothesis does not fit the facts, either the hypothesis is reformulated or the phenomenon to be explained is redefined so that the case is excluded. 5) Practical certainty can be attained after a small number of cases have been examined, but the discovery of negative cases disproves the explanation and require a reformulation. 6) This procedure of examining cases, redefining the phenomenon, and reformulating the hypotheses is continued until a universal relationship is established, each negative case calling for a redefinition or a reformulation. (Denzin, 2007, p. 134) In the building of this model, I put great emphasis on the epistemic practices and standards described by the interviewees, as I tried to trace the methods journalists used to assess the credibility and value of information in Wikipedia. In addition, I tried to understand how these practices are incorporated in the larger context of journalistic workflow, rather than focusing on them in isolation (see Heath, Knoblauch and Luff, 2000). Apart for reconstructing journalists' practices, I used the interviews to observe how journalists frame their use of Wikipedia in a way that is coherent and justifiable. I used their descriptions, narratives, proverbs and metaphors to reconstruct cosmos or epistemic virtue relevant to their field of work. 64 2.2.3 The Use of Israeli Journalists as Interviewees: a Caveat As noted above, my interviewees were all Israelis working in Hebrew written venues. This may limit the generalizability of the findings presented above. As Nicholas Johns' (2011a, 2011b) studies about the diffusion of the internet to Israel during the eighties and early nineties suggest, the institutionalization of new media is highly dependent on individual and state level machinations. This means that further research is needed to understand the differences between the ways journalists in different countries use Wikipedia. Moreover, the most relevant version of Wikipedia for the needs of my interviewees was the Hebrew one, while this study's assessment of Wikipedia is mainly based on the dominant English language version. This problem is compounded by the fact that Wikipedias in different languages do not comply to the same polices (with a few exceptions, including the Neutral point of view policy, considered binding in all Wikipedias), and there is a large degree of variance in the way they operate. However, I would like to note that all interviewees, without exception, noted that they frequently use the English Wikipedia, some of them more often than they did the Hebrew one. Moreover, they report similar patterns of use and evaluation in different language versions, and more often than not use the English version as a higher authority than the Hebrew one. Thus, I believe that their reports may be used to construct a pattern of action generalizable to audiences of various language versions of Wikipedia. This belief, however, obviously demands further research to affirm. I will discuss these issues further below in chapter 7. 65 Chapter 3: Nupedia Cultural and epistemic changes often have modest origins. In this chapter I trace the very early stages of such a change, tracing the concrete and gradual microlevel processes involved in it and attempting at the same time to uncover the older, more stable parts of the relevant cultural repertoires (Lamont and Thévenot, 2000) whose assembly allows it to emerge. This assembly is constrained by the need to use familiar and legitimate cultural resources to cope with what Wuthnow calls the problem of articulation: "if cultural products are not articulate closely enough with their social settings, they are likely to be regarded by the potential audiences of which these settings are composed as irrelevant, unrealistic, artificial, and overly abstract, or worse, their producers will be unlikely to receive the support necessary to carry on their work; but if cultural products articulate too closely with the specific social environment in which they are produced, they are likely to be thought of as esoteric, parochial, time bound, and fail to attract a wider and more lasting audience" (Wuthnow, 1989, p.3) This problem limits the range of possible networks, leading to a preference of networks that succeed in enlisting already established actors and objects that can impart their legitimacy to the network. This was especially true for Nupedia: as a project based on the volunteer labor of (mainly academic) experts, it was designed and constructed by its founders specifically to bestow it with legitimacy and respectability, to make it attractive to them. Thus, I focus my attention on attempts by the project's founders and contributors to qualify objects and practices as legitimate in terms of more general cosmos (Venturini, 2009) or order of worth (Boltanski and Thévenot, 2006, Thévenot, 2001b). This chapter is divided into four subchapters: the first is a brief historical overview, based mainly on the chronology presented in (Sanger, 2005). The second subchapter describes Nupedia's encyclopedic model, which was intimately linked with that of the Encyclopaedia Britannica. The third subchapter analyzes Nupedia's mode of production, which was conservative in its assumptions about knowledge 66 production and based primarily on the ethos of expertise. The final subchapter depicts a social drama that occurred as a result of attempts to change Nupedia's mode of production. This drama uncovers the dialectic moved that took place in Nupedia, which began as a very conservative project but gradually began to downplay the reliance on experts – a trend that ultimately led to the creation of Wikipedia. 3.1 Historical Background Nupedia was initiated and operated by Bomis, Inc., a dot-com company founded in 1996 and owned by Jimmy Wales, Tim Shell and Michael Davis. Bomis was a relatively small company (with ten employees in October 2001), which ran a web portal (bomis.com) and made some profit from banner ads located in it. Nupedia was one of many ideas developed in this self-proclaimed "laboratory for experimenting with new ideas about the modeling of information, about new systems of online community and collaboration" (Bomis, 2006). It was supposed to be an "open source encyclopedia," produced by volunteers and generating income through banner-ad placement79. The presentation of Nupedia as the product of a commercial venture may be deceiving: Wales, the driving force both behind Bomis and behind the Nupedia project,80 previously used Bomis's servers for various purposes without financial motivation, such as providing hosting for Objectivist81 and Libertarian websites and databases. However, as I shall demonstrate, its roots in this commercial setting did condition its features and development, tying its fate with those of its owners, and the market in which they operated. In January 2000 Wales offered Larry Sanger to plan and spearhead the project82. Sanger was a Ph.D. student specializing in epistemology in Ohio State 79 Formed in 1998, the open source movement was a commercially driven effort to promote collaborative software development and permissive user license agreements. It was gaining notability and success at the time, and initial attempts were made to translate its success from the realm of software to that of internet content. 80 As he himself put it, "it is sort of funny to me when I read people discussing 'Bomis' in the abstract. Ultimately, that just means me, since I'm the final decision making authority. Bomis is not a big abstract company; it's just the people that you know. A down home mom and pop dot-com. :-)" (Wales, 2001a). 81 I.e. supporting the ideas and philosophy of Ian Rand. 82 Wales was acquainted with Sanger through online forums dedicated to discussing Ayn Rand's ideas, such as the #AynRand IRC channels and alt.philosophy.objectivism Usenet group (a contemporary 67 University at the time, and Wales was apparently looking for a like-minded philosopher to lead Nupedia. This seems to be at least partly a move which was meant to grant the project respectability – as is apparent from his insistence that Sanger will finish his dissertation as soon as possible. Sanger accepted the job and a month later moved to San Diego, where Bomis was located. He began working on the project in the beginning of February. The first copy of the Nupedia.com website in the internet archive is dated to March 4th (Wayback Machine, 2011a), but the project wasn't officially launched until March 9th. At that date the first messages were sent to the project's mailing list (Reagle, 2006), which was the main site of activity in his early days. First steps were taken towards publicizing the project and drawing volunteers: a day later, on March 10th, an article about the project appeared on PC World magazine (Gouthro, 2000), and mails were sent to mailing lists and newsgroups soliciting volunteers. If the date March 10th rings a bell, it is not incidental, and not entirely unrelated to our subject matter: it was the day the NASDAQ Composite index reached his highest point, after doubling its value over the year before due to "wild stock speculation and freewheeling venture capital investment that resulted in the often ludicrous overvaluation of sketchy internet companies" (Long, 2007). A day later, it started its long and hard decline, marking March 10th as "the day the bubble burst". This unfortunate economic coincidence was posthumously credited by some, including Sanger himself, as the main reason for Nupedia's failure. In the first months of the project, most of the efforts were directed at attracting volunteers and finalizing the guidelines for the project (Sanger, 2000a). On May 10th, Nupedia's news page reported that over 1800 members already joined the project (Nupedia, 2000a). By December 15th, the number rose to over 2500, including 15 subject area editors and 81 peer reviewers (Nupedia, 2000b). Throughout the first year of the project, most of the activity was carried out through various mailing lists. The majority of these lists functioned as review groups of specific encyclopedic topics, while some others were dedicated to general issues such as policy formations, copyediting etc. By June, the first drafts of articles were submitted to the topical account can be found in Wilcox (1994). Such forums provided the main social medium in which Nupedia – and to a large extent also Wikipedia – developed. 68 mailing lists for review (Nupedia, 2000a). At the same time, an advisory board comprised of Nupedia's editors and peer reviewers was set up and began discussing the project's guidelines, which were composed and edited primarily by Sanger. They did not change much as a result of these discussions. Version 2.1 of the guidelines was released in May (Nupedia, 2000c), and version 3.2, which was highly elaborate and relatively stable, was released in June (Nupedia, 2000d). By June, there were active review groups83 in the fields of biology; classics; computers; engineering; health sciences; literature; music; psychology; and zoology. An article assignment system was in place on the website and about twenty articles have been assigned, mostly on topics related to music and classical studies, the most active and populated of Nupedia review groups (Nupedia, 2000a). A great deal of the activity was still centered on recruitment: a flyer was posted on the website and mailing lists, and participants were requested to post it, both online and offline (Sanger, 2000b). By July, copyediting groups84 began their activity and started working on submitted articles and various technical improvements were carried out by programmers hired for the project, including an "anonymous remailer" system to enable blind peer review. For the first time, small incentives were offered to participants partaking in article production: editors, writers of at least one article and copyeditors of at least five were offered T-shirts and coffee mugs with Nupedia's logo and motto85. Copyeditors were also offered reimbursement for their reference books once they copyedited five articles (Nupedia, 2000a). According to the "newest articles" webpage (Nupedia, 2001a), it seems that the first article, "Atonality," was published in July 25th, and three more followed in August. By that time, however, the project's founders already began to raise concerns about the rate of production. Despite the continued development of the still unfinished technical system behind Nupedia and the founders' promise of long term commitment, maneuvers were soon set in motion in an attempt to accelerate production, most of them attempting to do so by flexing Nupedia's structures and processes: in August, 83 To qualify as active and begin assigning articles, a topic needed at least an interim editor and two peer reviewers. 84 There were two such groups, one for American English and the other for British English. 85 "Suppose scholars the world over were to learn of a serious online encyclopedia effort in which the results were not proprietary to the encyclopedist, but were freely distributable under an open content license in virtually any desired medium. How quickly would the encyclopedia grow?" (Campeu, 2000) 69 Sanger suggested active solicitation of brief introductory articles from non-experts and the introduction of a new "general" review group for fields that did not yet have an active review group (Sanger, 2000c, 2000d); in September, the whole production system migrated from mailing lists to a web based system (Nupedia, 2000b); and in November, a new policy was formed to enable and encourage the incorporation of open content materials produced outside Nupedia (Sanger, 2000e).86 At the same time, more moves were made to enlist new volunteers: in December, a contest was declared in which monetary rewards going up to 250$ were proposed to participants recruiting new members. However, these moves proved ineffective as by the end of the year only 12 articles were published, and an overall of 53 drafts were submitted to review process (Nupedia, 2000b). In January 10th, 2001, another such move was announced, when Sanger posted on Nupedia-l mailing list a mail with the heading "Let's make a Wiki!" inviting participants to examine the innovative Wiki technology, and offering to open one for Nupedia (Sanger, 2000f). Although initially planned as a side project, it was soon transferred to its own domain, Wikipedia.com. While Nupedia kept operating, it soon started dwindling – at least partly as a result of Sanger spending more and more time on the fast growing new website. Later, even after being fired from Bomis in February 2002, Sanger made several attempts to revive it by simplifying its editorial mechanism, introducing a modified Wiki platform (named "Nupedia chalkboard," see Sanger, 2001a) or using materials from Wikipedia for Nupedia (a project named "sifter," see Sanger, 2001b), but he ultimately failed. Nupedia only published 13 articles in 2001, none at all in 2002, and while an attempt to revive the project resulted in publishing two articles in April 2003 (Nupedia, 2003), it was the project's dying breath. After a server malfunction the website went down in September 2003, and with that ended project Nupedia. When it ceased to exist, it had produced a total of 27 articles (three of them being brief versions of longer ones), and had less than 60 articles that had undergone peer review. At the time, Wikipedia had long passed the 100,000 articles benchmark. 86 While the first two changes were implemented almost immediately, the later were not: the web-based system was not completed until the end of the year, and as far as I can tell, the use of external materials never got off the ground. 70 In what follows, I focus on two specific aspects of Nupedia which together defined its epistemic culture: its encyclopedic model and its mode of production. I track the "literatures" they are referencing to (Venturini, 2009), and describe how each of them was inscribed in Nupedia's mechanisms and practices. I then go on to describe how they interacted in a critical situation that involved the attempt to modify Nupedia's mode of production. 3.2 The Encyclopedic Model The goal of the Nupedia project was first and foremost the creation of an encyclopedia. However, questions such as what is an encyclopedia, and what kind of encyclopedia Nupedia is going to produce were seldom discussed, indicating that it held an unproblematic position in the project. Practically, Nupedia's main features relied on the general model of traditional paper encyclopedias – and specifically the Encyclopaedia Britannica. Indeed, the few texts and discussions that do deal with the completed work seem remarkably conservative, as features such as the overall structure of the encyclopedia, the writing style and the representation of knowledge were all explicitly modeled, to a large extent, on those of previous encyclopedias. In the few discussions where alternatives were raised concerning these features, the project's founders' position leaned – and tipped the scale – in favor of sustaining the traditional model. By and large, while most aspects of the Nupedia project were repeatedly articulated, explained and justified, its encyclopedic model was considered relatively unproblematic. Nupedia's founders didn't feel they had to explain it or coordinate expectations concerning it. Moreover, there was almost no demand from readers and participants for the presentation and explanation of that such a vision. There were even fewer controversies regarding it. This attitude seems to indicate two things: the robustness of the encyclopedic model as it developed in the 20th century, at least amongst the academic milieu which the bulk of participants belonged to; and the general adherence of Nupedia's founders and participants to this model. Thus, the epistemic underpinnings of the encyclopedic model remained relatively unarticulated, operating within Nupedia mostly in the form of underlying schemas or habits that are 71 "relatively unconscious, in the sense that they are taken-for-granted mental assumptions or modes of procedure that actors normally apply without being aware that they are applying them" (Sewell, 1992). In what follows, I use documents and self-descriptions found on Nupedia's website to reconstruct the taken-for-granted encyclopedic model underling Nupedia. As noted, Nupedia's founders hardly attempted to describe in advance the goal of their project, and where they did address it they did so in very indistinct terms. A characteristic example can be found in the "about" page, dedicated to an overview of the project. Explaining the need for a new encyclopedia, it constructs Wikipedia as different from other encyclopedia only in terms of its quality: "Our goal is to grow Nupedia indefinitely, to set a new standard for breadth, depth, timeliness, and lack of bias, and in the fullness of time to become the most comprehensive encyclopedia in the history of humankind" (Nupedia, 2000e) Nupedia's aspired quality is constructed in this text with regards to parameters that are implicitly constructed as central to encyclopedias: breadth, depth, timeliness, lack of bias and comprehensiveness. This list of common measures was used on several other occasions (see cf. Nupedia, 2000d, 2000f), indicating its importance to the project's self-presentation. In all these instances, the meaning of these parameters is not explicated, indicating that they are perceived as self evident with regards to the readers. The universal character of these parameters point to their meaning in broader terms: they create equivalences (Thévenot, 2001a, 2002a, 2002b) between Nupedia and an ideal model of an encyclopedia and construct a relation of commensurability between Nupedia and previous encyclopedias. Through these equivalences, Nupedia's founders indicate that it aims for the same objectives as traditional encyclopedias, and is different only in terms of its improved quality. An important implication of this construction is that it presents the reader with a vague but plausible vision of the project's outcome: Nupedia is going to be similar to existing encyclopedias, only bigger and better. As the content of the encyclopedia was yet to be produced, this vision encapsuled a definition of the situation in which the epistemic stance, the structure and the type of content in Nupedia are seen as taken for granted, and the main challenges are rather practical in nature. 72 Similarly, Nupedia's FAQ page (Nupedia, 2000g) presents an answer to the same question worded in greater length: "why is there a need for another encyclopedia? Aren't there already quite enough excellent encyclopedias?" This wording is very telling, as it positions Nupedia in a competitive relation with "excellent" traditional encyclopedias, and assumes it should justify its existence with regards to them. The answer, describing the "ways in which Nupedia has the potential of becoming unique," states that: Nupedia may well end up with a broader base of contributors than any encyclopedia in history. Moreover, it is hypertext, and online. Hence it is in principle infinitely expandable. So, our ambition for Nupedia is for it to become the largest -- broadest, deepest, and most up-to-date -- encyclopedia in history. Where Britannica might have 1000 words on a given relatively obscure subject in the history of science, Nupedia might, in the fullness of time, have a treatise. This text is a clear example of form investment (Thévenot, 1984), as it proposes to judge Nupedia with regards to parameters that are implicitly constructed as central to encyclopedias: breadth, depth, timeliness, lack of bias and comprehensiveness, indicating its importance to the project's self-presentation. The universal character of these parameters point to their meaning in broader terms: Nupedia is constructed as better, not different. It is subject to the same regime of justification (Boltanski and Thévenot, 2006) and adheres to the same epistemic virtues (Daston and Galison, 2007).87 This quote uncovers another important theme: the centrality of the Encyclopaedia Britannica in the self-presentation of the project. While the question is worded in a way that compares Nupedia with "excellent encyclopedias" in the plural, the only such encyclopedia mentioned is Encyclopaedia Britannica. This mention asserts Britannica as the model against which Nupedia judges its outcome. The reference to Britannica in this paradigmatic point is not accidental, as it played an important role in the construction of Nupedia's vision and policies. Nupedia's founders had in the Encyclopaedia Britannica a highly valuable resource in their attempt to make their vision a reality, as it helped narrow the interpretive openness of the abstract and implicit "traditional encyclopedic model." Britannica's hegemony in 87 This is especially evident in the illustration given for these consequences: the hypothetical topic is a traditional encyclopedic one (history of science), and Nupedia's advantage in this case is articulated as a matter of size only. 73 the realm of encyclopedias meant that virtually all those interested or taking part in the project knew it, and knew of its dominant status in the realm of encyclopedias, so that it could embody a culturally sanctioned model for general encyclopedias. As such, it could be used as a concrete point of reference or “test” (Boltanski and Thévenot, 2006) that narrows the interpretive possibilities and can be consulted in times of doubt88. Nupedia's founders were quite explicit about using the Britannica as a model for the Nupedia encyclopedia. On several occasions, they named Britannica specifically as the benchmark for quality they were trying to surpass, while neglecting to mention any possible changes to Britannica's structure and style. This position is salient in an article published in the pc-world journal the very day Nupedia was launched, which states that: "Aimed mainly at consumers, Nupedia considers Britannica its main competition. Britannica, which has been around for more than 200 years, recently made all of its content free on the Web. Sanger acknowledges that Britannica is a mighty force to reckon with, but maintains Nupedia will be a contender. 'Eventually we're going to out-Britannica Britannica,' he says. 'By that, I mean we'll do all the things that they do, only we'll do them better and in more depth.'" (Gouthro, 2000) Apart from setting a concrete model for the encyclopedia, Britannica influenced the Nupedia's encyclopedic model in a more subtle way. As it allowed free access to its content on its website, Nupedia's founders couldn't claim that they offer a new internet service, and thus had to find other justifications for the project's necessity. They used two such justifications: the first, demonstrated above, was that Nupedia would be better than Britannica. The second was that it would develop a substitution to Britannica's paper encyclopedia universally available only for just the cost of printing. This was a major rhetoric strategy used to mobilize volunteers by constructing the project as one of moral implications. This vision constructs Nupedia first and foremost as a not-for-profit substitute to Britannica in the realm of paper publications, and not as an internet centered initiative. These aspirations, in turn, conditioned its model, as they reduced the feasibility of features that exceeded the limitations of traditional desk encyclopedias. And indeed, Nupedia generally adhered 88 Moreover, there were already several online encyclopedias active at the time, including Microsoft's encyclopedia Encarta and the Encyclopedia Britannica online, which adhered to a similar model. 74 to the traditional model of encyclopedia in terms of the general structure, writing style, and attitude towards knowledge representation. In terms of the general composition and structure of the encyclopedia, Nupedia was supposed to be arranged in a topical category structure similar to the ones found in indexes of traditional encyclopedia89. Sanger openly admitted that in order to compose it he "consulted several English language desktop encyclopedias," and that he omitted subcategories when he felt he "lack[ed] the competence to say authoritatively what the subcategories should be". Nupedia's initial category scheme (Nupedia, 2001b) was thus arranged along manifestly conservative90 disciplinary lines. In terms of the writing style, Nupedia's guidelines do not exceed what one would generally expect from an encyclopedia, including referring to such resources as the Chicago Manual of Style or Strunk and White's Elements of Style (Nupedia, 2000d). Nupedia's founders' statements regarding Wikipedia's encyclopedic model usually stressed the equivalence between Nupedia and paper encyclopedias in terms of article content, and the need to stick to their standards. A significant manifestation of this tendency can be found in a discussion started when a participant suggested on the Nupedia-l mailing list to supplement traditional articles with manifold versions suited to a particular audience: "Short, long, complex, simple, technical, practical articles" (Banerjee, 2000). This suggestion raised a long discussion between the project's founders and contributors, and forced the later to present argument in favor of their conservative attitude. The most comprehensive response was Jimmy Wales', in a post later supported by Sanger. In it, Wales writes as following: I think this discussion is extremely important. Here are some thoughts. 89 Like almost all paper encyclopedias, the preliminary organization of Nupedia's contents was based on a version of the traditional division of the branches of knowledge, its top-level categories being foundational disciplines, natural sciences, social sciences, applied arts and sciences, and culture, each branched into disciplinary categories (Nupedia, 2001b). 90 As Sanger explained, when he defended his inclusion of the classics category, "there is no reason why we should… [deny] the importance or usefulness of well-accepted categories of scholarship such as, indeed, Classics" (Nupedia, 2001b). 75 My dream for Nupedia is to have a tool that is very widely (universally!) accessible. The open content license will enable people to take the content and redistribute it in a great many forms, including paper books to be sent to countries who cannot yet afford an internet infrastructure. As such, Nupedia ought not to be an extremely high level academic encyclopedia aimed only at specialists. Articles should be encyclopedic, as opposed to polemic or academic, if you see what I mean. At the same time, I don't envision this as an encyclopedia for kids -- at least not in the first pass. The idea that hypertext and computers can support a wide variety of levels is a very good insight, as is the insight that a high level article with hypertext links to a glossary for the "hard" words can render an article more widely available. However, my reservation is that many overly ambitious projects never manage to get off the ground. We intend to _succeed_ at building a fantastic resource, which means that we ought to *actually do* just what we *can do*, rather than simply dreaming of utopia. -----------If we make one solid pass through -- aiming for breadth, i.e. more-or-less complete coverage of everything, then at that point it will make sense to seek volunteer "translators" who can translate the articles to a different reading level, to put together a "Kid's Nupedia". This will be in many ways similar to translation efforts which seek to create a "German Nupedia" or "French Nupedia" or "Chinese Nupedia". But we can't do everything at once, not even with 1500 volunteers. If we can demonstrate to the world our _success_ at creating a first-pass high-quality encyclopedia, at the reading level of Britannica, let's say, then we will create a _movement_, an avalanche of support like the avalanche that is carrying Linux forward. We'll have thousands and thousands of volunteers streaming in, making even the most impossible tasks possible. But -- one step at a time. :-) --Jimbo (Wales, 2000a, Underscore in the original) 76 As this quote shows, the vision of the project is articulated as one that surpasses the realm of the internet, aspiring to reach, in book form, "countries who cannot yet afford an internet infrastructure." This vision, central to the project construction as a moral project, is mobilized as justification for not making Nupedia "an extremely high level academic encyclopedia aimed only at specialists," or alternatively "an encyclopedia for kids." After he acknowledges the virtues of proposed changes, Wales explains that Nupedia has limited resources, and trying to go beyond the traditional format might consume them and fail the project before it gets off the ground. The last and most important part of the post outlines the founders' argument for their conservative attitude. The strategic outline justifying the founders' attitude is based on a two-step program: at first, the project has to succeed in creating a "high quality encyclopedia, at the reading level of Britannica" (ibid). This, in turn, will generate enormous support that will enable to surpass the vision of traditional encyclopedia. However, this will only be possible once the "first-pass" encyclopedia is finished – And that encyclopedia is a traditional, Britannica-like one. As this description illustrates, Nupedia's founders were very insistent in their attempt to keep Nupedia in line with the established traditions of modern encyclopedism, and took a generally conservative outlook towards it. Britannica thus functioned in it as an Obligatory Passage Point (Bijker and Law, 1992; Callon, 1986; Latour, 1988), funneling all the activity in the project though its standards. The authority of this standard was based on a strict traditional legitimacy (Weber, 1978, pp. 266-231), as neither rational-legal justifications nor charismatic claims were involved in its formation. A similar attitude, though somewhat more complicated, can be seen in Nupedia's mode of production. 3.3 The Mode of Production "I maintained from the start that something really could not be a credible encyclopedia without oversight by experts. I reasoned that, if the project is open to all, it would require both management by experts and an unusually rigorous process." (Sanger, 2005, italics in the original) 77 This quote sums up well the attitude held by Sanger towards Nupedia's mode of knowledge production. While the Nupedia was supposed to be an encyclopedia that anyone could edit, it was had an elaborate editorial review process overseen by experts. This process combined two models of knowledge production, one based on academic peer review and the other based on public debate in mailing lists. These two models embodied different epistemic assumptions, the first relying on an expert-based stance and the other more akin to public deliberation. They were similar, however, in that they embody culturally sanctioned models of knowledge giving and of representation which stand at the heart of the modern stance.91 They thus both confer legitimacy, and provide ready-made solutions for the problems of knowledge giving, production and verification. However, the inherent tension between divergent modes of representation meant that their relative standing in the project had to be explicitly addressed and practically resolved. Evidently, the expert based model held an undisputed superiority within the project, and was deeply engraved in Nupedia's structure and processes. This dominance, however, waned over time, as slow production pace became a serious problem in the project. Nupedia's organizational structure, proposed by Sanger and elaborated with the help of other contributors, was based on a horizontal separation between areas of expertise (topical categories) and a hierarchical division of labor between several types of volunteers. At the top of the hierarchy was the editor-in-chief (Sanger himself), who was the final ruler on matters of knowledge production. Under him were the subject editors, responsible for formulating topic-specific policies and assigning and approving articles. Peer reviewers were responsible for approving articles during the review process. The editors and peer reviewers were also the members of Nupedia's advisory board, officially responsible for discussing and setting Nupedia's policies. To each article was assigned a single writer. Moderators were supposed to moderate the discussion of articles by non-experts in the open review process (see below). Finally, there were two roles that did not require expertise and were not associated with topical categories: Copyeditors that were supposed to 91 Latour (1993, pp. 12-48) even goes so far as claiming that the distinction between political and scientific representation – the epistemic cultures related to these models - is the very cornerstone of modernity. 78 copy-edit all the articles after their content was approved. Finally, casual participants could comment on articles during the open review process92. The manning of positions was based primarily on formal credentials. In the top of the hierarchy was the editor-in-chief, an academically certified epistemologist. Under him, the subject editors were supposed to be experts in their field and (with few exceptions) posses a Ph.D. in it. The demands from Peer reviewers were smaller, as "the vast majority have Ph.D.'s or are a few months from getting their Ph.D.'s, or have equivalent publishing, teaching, and/or professional experience." Editors and peer reviewers were (with very minor exceptions) the only persons allowed access to Nupedia's advisory board, officially responsible for discussing and setting Nupedia's policies. Writers were supposed to be "often, and as appropriate, an expert on the topic". Lead-reviewers were "selected from those with the greatest expertise in the relevant area". Moderators were "expected to have at least a bachelor's degree or similar qualification in the subject area they're moderating."93 Finally, Copyeditors and casual participants are the only two roles that did not require any expertise (Nupedia, 2001c). Thus, while not all participants in the project were supposed to be experts, they were always implicated in a hierarchy that vested authority according to level of expertise. This organizational structure was also the basis for the recognition and acknowledgement scheme which were an integral part of Nupedia's attempts to create a rewarding gift economy for its potential contributors. This connection was highlighted in a section appearing in Nupedia's page devoted to attracting volunteers, titled "Join us!" This section outlined the different functions in the project, the qualifications needed to perform each function, and the credit which volunteers filling each role will receive in the text: Editors get their names on each page in their category, and have biographical pages; Writers names appear in a byline on any article they write (indicating, by the way, that they are supposed to write them without collaboration). They may also receive biographical pages if they wish to; no credit is mentioned for Peer reviewers, Copyeditors and Observers. Thus, the main point of 92 Two other roles were those of programmers and translators. However, their authority was not extended over the experts on the subject of the debate, as comments by the subject editor; the subject peer reviewers and the author were not moderated. 93 79 this section (besides explaining the possible roles of potential participants) is the coupling of credentials, responsibility and renown. The production process in its mature form94 included seven steps: (1) the editor-in-chief or one of the subject editors assigns a topic to a writer. (2) The subject editor finds a lead reviewer. (3) The lead reviewer and writer engage in a blind review process, revising the article until the reviewer decides it is "nearly satisfactory for inclusion in Nupedia" (4) The article is published on a dedicated webpage, where peer-reviewers of the category as well as "the general public" comment and suggest improvements until the editor, lead reviewer and another peer reviewer approve it. (5) After the article is approved, the author selects two copyeditors to go over it. (6) The article is published on a dedicated webpage, where "the general public" can comment and suggest copyediting improvements until the two lead copyeditors approve it. (7) The category editor approves the article before it is published. As is evident from its review process, Nupedia relied on the participation of certified experts for its operation. The project's general structure was based on a hierarchy in which one's position and role was determined by his education and credentials95, and decisions and products in each level were subject to oversight and control of the levels above it. The manning of positions was executed in a top down fashion, and the writing of articles could not begin without a subject editor to assign them, and two additional peer-reviewers to form an "active" review group. However, the reliance on official, academically recognized expertise and identity was complemented by a second conception of knowledge production and authorization, based on egalitarian public deliberation. The contrast between these two models is evident in the existence of two review systems for articles, the first modeled after that of academic peer review, and the second after that of public discussions in mailing lists. Each reviewing and editing 94 This description is derived from version 3.2 of Nupedia's editorial policy guidelines (Nupedia, 2000d). Initial versions were less elaborate and did not include a separate lead review process, but very similar otherwise. 95 This is evident in the very first practical step Jimmy Wales made towards creating Nupedia: recruiting Larry Sanger to lead the project from the planning stages. According to Sanger (2005), Wales told him upon his recruitment that he was specifically interested in a philosopher to spearhead the project. He also conditioned his employment upon his finishing his Ph.D. quickly, and promised him a raise when it was done. 80 process embodied different epistemic cultures, with different notions of consensus; different origins; different goals; different technologies; and different principles of authority and responsibility. In general, the one relying on the competence of experts was deemed superior to the one relying on public discussion. In what follow, I describe the two review systems and the way they evolved to answer the pressing problem of slow production pace. 3.3.1 The Peer Review Process The source of Nupedia's peer review process is, in all probability, the academic model of article production96. Given Nupedia's leaning towards respectability and its founder's academic history, it is not surprising that they chose such a model. Within the academic community, the peer review model is widely considered as an integral part of the scientific process, to the point that it was considered "crucial for the effective development of science" (Zuckerman and Merton, 1971). The peer review model of legitimation suggests that the content of the article should be scrutinized by experts on the subject. These should be with a similar or superior educational background to that of the writer ("peer"). However, once the reviewer is chosen he is supposed to put aside his and the author's personal identity, their relations, academic status, etc. To ensure this, the review process is usually "blind" (as was the case in Nupedia), meaning that the authors and reviewers do not know each other's identity. This practice is considered of utmost importance, as it reduces the risk of contamination by personal or irrelevant aspects. Nupedia's lead review process followed the same practices, and could thus seek to appropriate the legitimacy that peer review process was accorded within scientific community. The participants in the peer review process are thus supposed to be endowed with an epistemic virtue (Daston and Galison, 2007) which allows them to set aside personal considerations and engage the process of evaluation with fairness97 (Mallard, 96 Sanger indicated this connection on several occasions, as when he stated that "So far we've been following the academic/peer review model of article writing" (Sanger, 2000c). 97 The meaning of fairness is usually constructed as entailing the transcending of particularistic considerations and the use of universalistic criteria of evaluation (Zuckerman and Merton, 1971). 81 Lamont and Guetzkow, 2009). In the deliberations between the writer and reviewer both are expected to set aside their personal opinions and assess the value of the written work objectively and impartially. This distancing of the act of judgment from the subjectivity of the judge is inherent to the peer review process. However, it is still embedded in the subjectivity of the peer reviewer as the carrier of the adequate virtue. In a similar vein, the lead reviewers in Nupedia were supposed to use as a rule of thumb the question "If this draft of this article were to be published under my name, would I be embarrassed by it?" the use of the notion of embarrassment here is very appropriate, as it targets the reviewer’s sense of professional competence while not forcing him to agree with the particular contents of the article. It also brings to the front the moral aspect of knowledge production: epistemic virtues, which dictate and qualifies standards, are mediated – at least in part – through the attribution or avowal of shame on those breaching them98. 3.3.2 The Open Review Process The open review process was modeled after the public discussion that took place on internet mailing list since the seventies. At the time, mailing lists were the most popular mode of extended online communication, and the model for the other widespread mode, the online forum99. As the World Wide Web was still composed mainly of static pages, this was the obvious choice for a method of deliberation. Originally based on the email format (though transcending it in later web manifestations), mailing lists allow multiple participants to post messages, see them and comment on them, and to group different discussions in "threads". Participants enter the discussions at will, and leave it at will. Participants can – and often do - post their comment through a virtual identity, and remain anonymous with respect to their real life identity. Discussion groups can also have a moderator approving or deleting posts, and are usually archived.100 Sanger and Wales both founded and moderated Though different articulations of fairness exist (for an overview, see Mallard, Lamont and Guetzkow, 2009) they all agree that fairness involves the reviewer putting aside his personal views to assess the article through some sort of impartial criteria. 98 See Mills (1940) on the attribution and avowal of motivation. 99 For a historical overview of the development of e-mails and mailing lists, see Hafner and Lyon, (1996, pp. 187-218). 100 Their most influential manifestation was in UseNet, the main venue for internet communication before the creation of the World Wide Web. For an overview, see Lueg and fisher (2003). 82 discussion groups in the past, and came to know each other when they participated on each other's lists. In contrast with the "one-to-one" communication of letters, and "one-to-many" communication of print, mailing lists are also participatory and communal, enabling communication of many-to-many (Wilson and Leighton, 2002). They are thus often considered as generating a kind of public sphere where issues can be discussed freely and without fear of repercussions (c.f. Poster, 1997). Mailing lists also often generate virtual communities which sustain participation-based, meritocratic gift economy which encourages their participants to share their knowledge with each other (Bergquist and Ljungberg, 2001; Rheingold, 2000). In these features and others, they embody the essence of the "new media" (Logan, 2010, see especially chapter 5). In a same vein, the open review process in Nupedia involved public discussions of specific articles, in which participants debated about the merits of the article, and raised suggestions to improve it101. Following the prevalent customs, and in contrast to the identity-based peer review system, Nupedia allowed anyone interested to participate in open review groups, even anonymously. The peer review and the open review models embody very different assumptions about the connection between subjectivity and knowledge. The public discussion model is based on the idea of contrasting different ideas, opinions and identities. It assumes that a discussion needs to be conducted between proponents representing different views, and that every view may be relevant. These ideas stood at the heart of the justification of the open review process in Nupedia. For example, in the FAQ page, when explaining how Nupedia can avoid bias, it is stated that in the open review process, "if the proponent of a controversial view slants an article in such a way as to make his view look better than the others, representatives of the other views will be on hand to ask him to tone it down, to include arguments in favor of other views, etc" (Nupedia, 2000g). In stark contrast with the assumption of objectivity and expertise implicit in the peer review model, this quote asserts an antagonistic view, in which opposed parties balance each other's biases until the truth is achieved. In this process, the personal views and positions of the participants are of utmost importance, as they do not serve as impartial judges but as adversaries. The 101 This process was initially supposed to take place on mailing lists, and was later moved to special web page. 83 consensus reached is a result of negotiation between opposing views that balance each other out, and the impartiality is the result of the process, not its precondition. This transgression from the epistemic culture of expertise, however, was not allowed to go very far. As arenas for public deliberation mailing list discussions they have no final ruling mechanisms and notoriously persistent, even in those very rare cases where some consensus is reached. As Nupedia's open review process was supposed to generate a final decision, an adequate mechanism had to be formulated. Interestingly, the mechanism chosen for that task was not that of a consensus between the reviewers, or a majority vote: In order to pass this stage and be approved, the article had to gain the approval of the editor, the lead reviewer and at least another peer reviewer of the subject category, while the opinion of other participants had no say in it 102. As this procedure indicates, the weight of the overall editing process was placed on peer review, rather than on the open review, since whatever effects the open review had on the article was mediated by the editors and peer reviewers and the public itself had no power to intervene in their decisions. Thus, the basic tenet of Nupedia's knowledge production process relied, in the last instance, on the authority of experts. To summarize, Nupedia dual review system tried to combine different mechanisms of mediated interaction, reconfiguring them according to the specific technological, environment and cultural assumptions of the project. While it strived to venture beyond conventional modes of knowledge production, it did not venture far from them. Its only innovation – the attempt to supplement the peer review process with an open review – was highly controlled and restricted, and was ultimately subsumed under the supremacy of experts. This, Nupedia's dual process actually relied on only one Obligatory Passage Points (Bijker and Law, 1992; Callon, 1986; Latour, 1988): that of expertise. In that, it is very closely implicated in what Boltanski and Thévenot's (2006) termed "the industrial polity," a regime of justification based on efficiency and productivity, 102 Peer reviewers, as opposed to casual participants, may, at this point, "wish to make it clear that that particular article version is not acceptable." However, if they are outnumbered, the article is still approved. 84 whose "men of worth" are technical experts. However, in a deeper sense it also relies on traditional legitimacy, as expertise is constructed in ceremonial terms (exemplified in the emphasis of credentials), rather than in functional one. 3.4 A Controversy: Modifying the Mode of Production Over its relatively short history, Nupedia's mode of production saw a number of changes, most of them occurring after Wikipedia's establishment (see above in the brief historical overview section). However, at least one major change predated it, and it proved a major step in the rethinking of the project that would ultimately give birth to Wikipedia. This change demonstrates the project's leaning towards the egalitarian public deliberation model and away from expertise/peer-review, as well as the ambivalence and hesitation towards this direction. Moreover, it illustrates the process of translation which objects and practices undergo as social changes displace them. I will thus describe this social drama at some length. While publicly adhering to an ethos of long term commitment and thorough work, Nupedia was soon confronted with the major problem that would eventually facilitate its failure: the slow pace of article production. By August 2000, it had only produced four articles. The project's founders decided they had to act103. In a post on Nupedia-l mailing list Sanger (2000c) introduced major changes to the production process. The opening and the ending indicate Sanger's decisive position: he begins the post by stating that "We in the office are concerned about the slow pace of article production, and we've decided on two ways to increase the pace of production"; and ends it by stating that "comments on these proposals are requested, but I want to act quickly, so please respond quickly if you want some input here". This wording frames the post in a way that asserts his right to change the article production system, while requesting "comments" from participants. His wish to act "quickly," as well as the very concrete articulation of his suggestions implies that he is determined on making them, and feels it is his prerogative. 103 It seems that the main driving force was Wales, and that Sanger was more skeptical towards the proposed changes (cf. Sanger, 2005). 85 The "two big changes" suggested address perceived problems in the article assignment and the review process that stems from the project’s reliance on expertise. The first change was to "actively encourage articles from nonexperts on topics that do not need experts" (Sanger, 2000c). In a later post summarizing and further explaining these same changes, Sanger (2000d) explains that the idea is that " we will accept all (or nearly all) article proposals for *brief, introductory* articles. […] You do not have to be an expert; you need only be a competent writer with decent research skills." This change, thus explained, implies a significant shift away from the ethos of expertise. It is further emphasized in Sanger's explanation for the change, where he contrasts the new, hopefully productive policy to the old "academic/peer review model of article writing" that is blamed for the slow pace of article production. The second change is even more substantive in terms of the production process: rather than peer-reviewing all the articles in the topical review groups, it is suggested that a new general review group, headed by Sanger himself will be set up to review articles on topics that don't yet have active review groups. This move both relies on Sanger's prestige as the editor-in-chief, and augments his role in the project. The rationale behind this move is that there are cases in which there are enough competent editors within the project, even if there's not yet a relevant review group. Moreover, it is suggested that all participants, regardless of credentials and expertise, should join the list and offer their input. This seemingly erodes even more the ethos of expertise. While the suggested moves appear to increase the weight of public deliberation model in the project, Sanger insisted on the supremacy of expert oversight. First of all, he limited the articles to be written by non-experts to short introductory articles on "prosaic topics of the sort that experts are not usually writing," suggesting that this move is in accordance with the routine practices of traditional encyclopedias and setting higher standards for more substantive or less "prosaic" articles. Secondly, he suggests using the general review process as an excuse to solicit the help of professionals outside Nupedia and thus to attract them to the project, indicating that experts are still the project's main sought-after resource. Finally, he asserts that articles approved by the general review list "aren't really peer reviewed" and should be properly reviewed once the relevant groups are formed 86 (Sanger, 2000d). Thus, he translated the seemingly anti-expert move in a way that can conform to the ethos of expertise. A more radical position can be found in Wales' replies to those opposing the move. In a post (Wales, 2000b) answering the objections of one of the most prominent contributors (Ruth Ifcher, nicknamed Rose Parks in the mailing list, the chief copyeditor of the project), he makes a much stronger case in favor of public participation as a substitute for experts review, at least with regard to short articles. In it, he offers several claims in favor of the proposed changes. Wales does not deny the importance of expertise in general, or in the project. However, he claims that in the process of writing and reviewing short articles the demand for expertise is redundant. As he states it, "for a short article, it is fairly easy for us to review basic quality without resorting to specialists. This will free up time for the specialists to do what they do best." Instead of expertise, he suggests another criterion for writer and reviewers, that of "competence". When he tries to illustrate the concept, he gives this example: "Milton Friedman is a famous economist. Britannica has a 2 paragraph article about him here […] Right now, we do not have an economics review group. But we do not need one in order to competently review a short 2 paragraph article about Milton Friedman. What is necessary is fairly simple... a few moments research on the web reveals a great deal of information about Friedman […] Taken together, any competent person could piece together an article as good *or better than* the article in Nupedia104, and we can easily quality control articles of this type. Competence is thus constructed as a quality shared by any "reasonably well educated" person who can access available online resources, compare versions of text, and compile his own account. This description radically changes the epistemic virtues related to writing articles: rather than knowledge and expertise, writers should posses the ability to consult and review texts – primarily online – and summarize them. This, as I show in the next chapter, is a major move towards the articulation of epistemic virtue in Wikipedia. The project's reliance on the model of the Encyclopaedia Britannica plays a critical role in Wales' reasoning. As Britannica was the model Nupedia sought to 104 Mistake in the original, should read Britannica instead of Nupedia. 87 emulate, it could serve as a standard against which particular articles can be assessed by the abovementioned competent persons. When required to make an important and controversial change in the project, Wales used Britannica as a pivot against which the project can change practices, all the while making sure that its standards remain adequate. From a more practical perspective, setting the Britannica article as standard allows Sanger and Wales to change the practices in Nupedia while guaranteeing that its process will still be legitimate in the eyes of the expert-participants. This move is a good example for the pragmatist principle that innovation usually arises from practical problems, as it is only due to such problems that the conservative stance typical to Nupedia was undermined. It also brings to light importance of external objects that can be used as equivalences and thus ground and justify changes in knowledge production. Britannica's articles serve Besides for the practical points noted above, Wales also makes a broader and more fundamental claim, asserting that that the project should be more open to nonexperts. In the end of his post, Wales elaborates on the broader point of view behind the suggested changes as he states: "I have a major concern that we are currently intimidating and scaring off many very good people. We want this project to be open, inclusive, rewarding to volunteers. One way to do that is to recognize that not every article needs an expert, and that not all reviewing needs to be handled by experts. Let's leave the experts time to do what they do best, but let's allow *everyone* to participate in their own way." Thus, in contrast to Sanger's more ambivalent position towards public participation, Wales emphasized its importance, and for the first time questioned the primacy of expertise in the project. However, He did not reject this ethos altogether, as he suggested that the changes will allow experts to limit their efforts to "advanced" articles. And while both Wales and Sanger conceded that the project needs new volunteers, Wales emphasized the need to use and attract non-experts while Sanger, in his post, addressed the need to attract more experts, Finally, in the last paragraph Wales explained the need to generate a large amount of general articles. As the project's backer, he asserted that instead of him "personally spending hundreds of thousands of dollars," the production of such 88 articles will help it "become self-supporting [...] by becoming useful to the general public, so that I can promote it." While I'm not sure what he means by the word promoting, the general idea seems to be increasing traffic to the website, and probably to sell advertisements in it. This explanation both asserts Wales' authority in the project, and at the same time moves the focus from the long term ambitions to the immediate need to generate productive content and make the encyclopedia usable as soon as possible, even at the expanse of the production process' rigor. To summarize, this social drama involved a shift of balance between Nupedia's two models of knowledge production, as it moved from an ethos of strict scientific expertise to a more public and inclusive model. This model, however, was now subtly changed: the public participation was now more than a matter of deliberation and contrasting views, as it relied on non experts constructing knowledge from available resources rather than rely on their own knowledge. While this shift did not seriously undermine the dominance of expertise, it certainly paved the way for more far-reaching attempts in the same direction. Following these discussions, both offers were implemented, and the general-l mailing list soon started operating (Sanger, 2000g). However, these changes weren't enough. The slow production process – and complaints about it – remained a constant throughout the project's history. As the year drew to a close, and the dot com markets kept declining, Wales kept trying to make Nupedia a more open project, based on a communitarian, meritocratic gift economy. At the same time, he tried to advance his epistemic stance based on the use of available materials rather than personal knowledge and expertise. Thus, in November, he pushed for the use of open content materials produced outside Nupedia (Sanger, 2000e). However, as far as I can tell, the actual use of external materials never got off the ground. In January, another attempt was made: creating a Wiki for the project, and allowing anyone who wished to edit it into an encyclopedia. The results of seemingly desperate move exceeded by far the expectations of its founders and changed the way we look for knowledge - and perhaps even the way we look at knowledge. 89 3.5 Conclusion As this chapter shows, the assembling of new epistemic projects is an act of bricloage (Hebdige, 1991, pp.102-106). They are built of older building blocks, collected from divergent and at time conflicting repertoires and models. In the case of Nupedia, many of these models – such as the peer review model, the ethos of expertise, the knowledge gift economy or the Britannica – exhibited a common, perhaps surprising feature: they were inherently modern, connected to traditions of representation that are at the heart of modernity, according to several interpretations (Giddens, 1991; Lyotard, 1984; Latour, 1993. And see above, chapter 1). These models, which helped generate a coherent vision for the project, were so widely familiar to the participants, that they warranted little, if any, translation. Nupedia's vision was modern through and through. In accordance with the attitude of its editor in chief, its actual manifestation was subject to a logic of cultural legitimacy, and based on academic expertise. In retrospect, this stance seems more limiting than enabling, as Nupedia failed to attract and mobilize the experts needed for its operation. I would like to suggest that this failure should be attributed, at least partially, to Nupedia's founders' reliance on practices derived from the academic world. This severely limited the repertoire of gift practices (Silber, 2007, 2013) usable by their potential contributors, and especially those prevalent in online content production at the time. Imitating the scientific gift economy (Hagstrom, 1982), Nupedia's production scheme was based on soliciting voluntary work from experts, in exchange for acknowledgement and recognition. Eventually, however, Nupedia did not succeed in creating an attractive gift economy attractive for credentialed experts, and blocked types of giving that could potentially undermine this economy but which were nevertheless potentially beneficial (for a review of design elements useful for institutions based on voluntary giving and cooperation, see Benkler, 2010). In spite of its attempts, Nupedia failed in creating and sustaining a contribution-based reputation system that offers sufficient symbolic rewards for its participants. As noted above, Nupedia's production process relied on the participation and contribution of credentialed experts, and the participants' standing in the project 90 was determined, to a large extent, by their academic credentials. Credentials determined their level of participation, their influence on individual articles and on general project policies – and the level of acknowledgement and recognition awarded to them. The coupling of expertise, responsibility and renown posited contribution as a mere intermediary between credentials and acknowledgement, naturally undermined enthusiasm for socially-embedded giving. This was even truer for non-experts, whose participation was restricted and considered a-priori less valuable. Even when attempts were made to actively solicit lay participation, their status in the project remained inferior. Moreover, there was a clear imbalance between Nupedia's demands of its academic participants and the rewards it offered. Academic experts were asked to engage in tasks – such as writing, editing or reviewing papers – very similar to their everyday activities, while being offered considerably less prestige and symbolic capital. And in contrast with the scientific gift economy they were routinely engaged in, the conversion rates of this capital were highly uncertain, as they could only be determined when and if Nupedia succeeds. In any case, the notion that participation in a private venture for an online encyclopedia could yield symbolic gains in the academic field was – and still is – very doubtful. Finally, Nupedia failed to create the communitarian context necessary to sustain interest in inter-community reputation. As noted above, its social structure was based on a formal hierarchy and topical differentiation, both of which undermined attempts to construct it as a community. Moreover, it operated through many topical mailing lists dedicated only to professional discussions (almost all of which had relatively few participants), a general mailing list that seldom included important discussions and was thus not very popular, and one used solely by the advisory board for policy discussions. Thus, lacking a mutual space for interaction, Nupedia failed to form a global microstructure105 (Knorr-Cetina, 2005) that could transcend its nature as a formal organization, and generate the feeling of communal participation and engagement needed to encourage acts of generalized exchange. It didn't even have a space where members could see each other's actions and contributions, which could 105 Global microstructures are based on features characteristic to face to face interactions such as trust and acquaintance, rather than on clear structure and hierarchy. 91 have encouraged participation either as a form of reciprocity for other participants' actions or as a means of out-giving them. Regarding the second, individual mode of giving, Nupedia attempts to create a design that would sustain a viable gift economy hampered its ability to attract such gifts. Its design failed to create a mechanism for free, unmediated giving, and also did not give its participants a feeling of worth and fulfillment in their contributions. First, giving knowledge to Nupedia demanded of the giver much more than the giving itself: Nupedia's production process demanded participants to undergo a rather lengthy identification process before beginning to contribute to it, and many positions in the project required contributors to divulge their real life identity. These demands posed a barrier between the willingness to give and its fulfillment, which may have deterred potential contributors interested solely in the free, immediate and unconstrained giving of knowledge. Moreover, Nupedia's processes demanded that contributors be assigned tasks and article topics by higher-level participants, thus conditioning their gift upon their willingness to partake in a hierarchical social structure. Such conditions for giving virtually prevented the possibility of an individualized gift that is detached from a specific social context. They posed obligatory passage points (Bijker and Law, 1992; Callon, 1986; Latour, 1988) which seriously constrained the possibility of a free, disinterested giving. Finally, Nupedia failed to abide by one of the most essential aspects of the gift process: the obligation to receive (Mauss's (1990 [1925]). Knowledge given to Nupedia was not accepted as is, but rather had to undergo a review process by experts. This method of operation may very well have damaged individual givers' enthusiasm towards giving knowledge – especially since it made their act of giving extremely similar to the unpleasant chores of schools and academic institutions. As Jimmy Wales later noted: "I knew [Nupedia] wasn‘t going to work when I personally sat down to write an article about Robert Merton, who[…] won a Nobel Prize for option pricing theory. So I said, oh well, I have a published paper in the area, I know something about this, and I sat down to write the article and I felt like I was back in graduate school because they were going to give my paper to professors to review and I was 92 going to get comments and, you know, I might get a C grade or a B grade or something" (Lamb, 2005) Bricolage is always a gamble. And in the case of Nupedia, this gamble failed. Ambitious and serious as it was, Nupedia never got off the ground. It did not succeed in attracting sufficient productive volunteers, it failed to make them produce articles, and its review process was excruciatingly slow. The result of Nupedia's failure to receive voluntary gifts of knowledge was a dialectic movement: an attempt to make the project less serious, less ambitious, and less strict. The two most notable Obligatory Passage points in Nupedia, the model of the Encyclopaedia Britannica and the expertise-based rigorous production process, were the first to be dismantled. Thus, it may be argued that Nupedia's most important contribution to Wikipedia was a negative one: the rejection of traditionally legitimized models and practices forced Wikipedia's founders to devise a new epistemic culture, based on innovative codes, standards and practices. As Pragmatism tells us, it is only when problems occur that unthinkable epistemic practices emerge. To save his project, Wales suggested a new epistemic virtue, based not on personal knowledge and education, but rather on the ability to competently judge and import resources available on the web. The first signs of this move are visible already in the processes described in this chapter, but its true repercussions only became evident when the founders' worry about production pace resulted in a new, self-proclaimed "very silly project" (Sanger, 2005) with the silly name Wikipedia. 93 Chapter 4: Wikipedia's Founding Period As the previous chapter shows, Nupedia was infused with a tension between conflicting attitudes, with Wales usually pulling towards more openness and faster content production, and Sanger pulling towards legitimacy and authority. While Sanger's attitude originally had the upper hand, the tension grew as the time passed. The project's failure to produce large quantities of articles resulted in Wales' attitude slowly becoming more manifest. Moves such as soliciting articles from non-experts, the founding of General-l and the suggestion to use open content materials produced outside Nupedia were all indications of a shift of balance away from an emphasis on legitimacy and quality and towards an emphasis on openness and quantity. In a sense, Wikipedia was a continuation and radicalization of shift away from the pursuit of legitimacy. It was self-consciously constructed as unrespectable – as Sanger (2005) later stated, "I came up with the name 'Wikipedia,' a silly name for what was at first a very silly project." At the same time, it wasn't constructed as an integral part of Nupedia but first as a complementary project and very soon as an independent project – with the explicit intent to preserve Nupedia's legitimacy. Wikipedia began as an experiment, and it exceeded its creators' wildest dreams for better and (in some cases) for worse. It replaced Nupedia seriousness and rigor with a very liberal outlook, its formal organizational structure and hierarchy with a loose network structure, and its emphasis on the moral commitment and potential gains of its participants with an explicit emphasis on "fun". In what follows, I'll describe how Wikipedia grew from an empty website to a robust and dynamic community, dedicated to the creation of an encyclopedia using new models of knowledge, its producers and its production. Wiki, short for WikiWikiWeb, was a Perl-based application developed by Ward Cunningham and introduced on his website on 1995 as means to enable collaborative editing and idea sharing between extreme programmers. Extremely simple, it was nothing but a very simple hypertext system designed for the World Wide Web. It allowed users to easily edit text pages and connect those pages using hyperlinks. Hyperlinks were generated by writing of a word with two bolded letters 95 (e.g. WikI or WikiPedia). If a page by that name did not exist, it was automatically generated as a blank page. Over the last years of the nineteen-nineties, a small network of Wiki based websites began developing around Ward Cunningham website, some using his application and others based on newly developed ones based on his design (see Leuf and Cunningham, 2001). The question who was the first to introduce the idea of using a Wiki to create an online encyclopedia was, for a time, the subject of an intense quarrel. In 2005, as Wikipedia rose to fame Wales began claiming that he was its sole founder, and that the idea of using a Wiki for Nupedia came from another Bomis employee. Sanger, in response, published a memoir of that time (Sanger, 2005), a large part of which was dedicated to rebuke these claims, as well as other descriptions of the early history of Nupedia and Wikipedia he found misleading. Its publication sparked a long, heated and fascinating debate of the memoir and the issue of how to attribute a "founder" status.106 As Wales noted in this discussion that he had "only very very minor quibbles with Larry's history" (Wales, 2005b), I use it as a generally credible source for the early history of Wikipedia. Other important resources for this section are the "Wikipedia announcements" page (Wikipedia, 2001a, 2002a) and Wikipedia-l mailing lists. These are the sources of the information in this overview, unless noted otherwise. The decisive moment in the usual story of Wikipedia's origins is known as "The conversation at the taco stand". On January 2nd 2001, Sanger met Ben Kovitz, a friend whom he met when they both participated on philosophy-oriented mailing lists (including the one Sanger created). Kovitz was a programmer employing the method of extreme programming (Wikipedia, 2012e) and had been, for some time, a participant in Ward Cunningham's Wiki, a central location for discussions and development of extreme programmers. Thus, when he and Sanger met over dinner, he enthusiastically told him about the concept of Wiki, and they discussed the possibility to use it in Nupedia107. Kovitz later noted the extreme programming standard 106 See the discussions in Wikipedia-l mailing list from April 2005 (Wikimedia, 2005). Most relevant comments' subject is "Sanger's Memoir" or "Re: Sanger's Memoir." 107 Kovitz and Sanger's report diverge on the details of the discussion: Kovitz remembered that he mentioned Wikis in response to Sanger's complaints about Nupedia and its slow rate of production, while Sanger recalls that he just told him about it, and that it made him think it could be a good solution to his problems. Also, Kovitz's remembered that Sanger was initially skeptical, but thought it 96 technique of "pair programming," the collaboration of multiple programmers on a single piece of code (cf. K. Beck, 2000), as an inspiration for the idea of using Wiki as a means to create an encyclopedia. Sanger contacted Wales that very night, and suggested they set a Wiki for Nupedia. For his part, Wales remembers hearing about the Wiki earlier from Bomis employee Jeremy Rosenfeld, but doesn't deny that it was Sanger's suggestion that made him set up a Wiki for Nupedia (Wales 2005b). In January 10th, Sanger posted on the Nupedia-l and the advisory board an email introducing the idea and asked if there were objections to start a Nupedia Wiki. The same day, the Nupedia Wiki was up. However, due to objections raised in the advisory board mailing list, it was decided to set it up as an entirely independent project, on an independent website. Wikipedia.com went live on January 15th, the date since then celebrated as "Wikipedia day". The Wiki was open to editing by everyone, and anyone could start a new page or edit an existing one. Editing was relatively simple and the Wiki contained only simple text and very basic markup. During the first months of its existence, Wikipedia enjoyed a period of relative isolation that was essential for its subsequent success. During this period, a small and relatively stable group of participants laid the ground for future development, as they formed a cohesive and productive virtual community. As they built the initial mass of articles required to attract web surfers, and formed a relatively stable community, with internal standards and established practices. Within this period, Wikipedia changed from a chaotic and unstructured experimentation with the Wiki concept to a real project – though one based on the notion of "fun" more than morality or self interest. Participants in the project knew each other personally, and had time to get acquainted and learn how to resolve issues in the project. Arrivals of new participants were relatively infrequent during that period, and they usually joined the emergent community on its terms or quickly left. Bomis personal – Sanger, Wales and Tim Shell – held a prominent position in the community, had a central role in discussing policy and contributed a lot of content. A few participants – such as Astronomer, AyeSpy and Ruth Ifcher, arrived from Nupedia, while others arrived following links was an interesting option. Sanger, on the other hand, remembers he "was nearly instantly excited by the idea." (Both versions appear in Wikipedia, 2012e). 97 from ward's Wiki. Many of the first Wikipedians, such as Wales, Shell, Manning Bartlet, Bryce Harington and Lee Daniel Crocker, were computer programmers, web developers and internet enthusiast – a fact that had a profound impact on the project. There were relatively few academics in the humanities and social sciences. Wikipedia soon began attracting new participants, most of which were not related to Nupedia. As it took considerable time for Wikipedia to generate traffic from search engines, most participants arrived through links placed in stories about the project featured on various websites. Some arrived following links from ward's Wiki, but many, if not most, of the new participants arrived from Slashdot.com, a highly popular website featuring "news for nerds" and excessive discussions about them. Many computer programmers and members of the technical elite frequented the website, and each time a link was given to Wikipedia some of them would join the project. Wikipedia was "slashdoted" (as these events came to be known) twice in its early months, on March 5th and March 29th (Wikipedia, 2001a). Slashdot brought to Wikipedia a very distinct group of participants: savvy and dedicated internet users immersed in its culture and ethos, technologically inclined and suspicious of any attempt to police and regulate it – or parts of it. This group, the regular audience of such websites, had a profound effect on Wikipedia108, as they brought with them ideas and practices that would shape many of Wikipedia's features and policies. Policies regarding legitimate content were very liberal during this period, as all sorts of contents were added to the encyclopedia including, for example, Larry Sanger's lecture notes from a philosophy course he taught (Wikipedia, 2001b), Tim Shell's extensive notes on Ayn Rand's Atlas Shrugged (Wikipedia, 2001c) and various materials in public domain copied from other websites. There was no clear distinction between articles and other types of content – such as policies, discussions, personal user pages and commentary on the project – apart for indications in the page names. Thus, the responsibility to distinguish between types of content, and to adhere to the project other content policies, was laid on the project's participants. During these first few months, many of Wikipedia's norms and policies were articulated and institutionalized. The initial policies of the project were very liberal, a 108 This influence was so apparent that Wikipedia was sometimes referred to as "the encyclopedia that Slashdot built" (Lih, 2009, p. 69). 98 spirit captured in one of Wikipedia's earliest policies, called "ignore all rules" (Wikipedia, 2012a). Formulated by Larry Sanger, it stated that "If rules make your nervous and depressed, and not desirous of participating in the wiki, then ignore them entirely and go about your business." However, this policy itself was a direct indication that Wikipedia started having some rules, as important Wikipedia policies and practices were being shaped during this period, moving it away from the cultural norms prevalent in Wiki culture. These included the creation of naming convention for articles; the deletion of vandalism and "patent nonsense"; the removal of discussions from article pages to special discussion subpages (called "talk pages"); the copying of public domain materials available on the web (such as the first volume of the Encyclopaedia Britannica's 11th edition and the CIA factbook's website); and the articulation of Wikipedia's main content policies, namely "neutral point of view" and "what Wikipedia is not" (Wikipedia, 2012f, 2012g). Such policies were usually implemented through the creation of an effective consensus (or rough consensus, in internet parlance): a state where standards and policies are usually not enforced by a designated authority but rather embraced by enough of the participants to become de facto rules. However, first attempts were made, especially by Larry Sanger, to enforce some policies even against opposition. These attempts brought about some resentment, but Sanger's positions and actions were usually respected, as he was considered by participants (alongside jimmy Wales) as having a measure of authority over the project109. The delicate balance between freedom and rules that was achieved in this stage did not hold for long, as a wave of new participants entered the project in late July. On July 24th, Sanger noted in a post on Wikipedia-l that Britannica was now beginning to charge readers for viewing full articles, and suggested using this move to promote Wikipedia. The next day, following on his own advice, Sanger posted an article about Wikipedia and Nupedia as alternatives to Britannica to Kuro5hin.com, a collaborative discussion website that presented itself as focused on "Technology and Culture, from the Trenches" (Sanger, 2001g) A day later, the article was mentioned in 109 Another important development that happened during that period was the development of Wikipedia's in languages other than English. As Nupedia already had foreign language versions, and some of their contributors were active in Wikipedia, it was only natural to start sister-projects in these languages. Thus, Catalan, German, and Japanese Wikipedia's were formed in March, joined by French, Chinese, Esperanto, Hebrew, Italian, Portuguese, Spanish and Russian in May. The links between international Wikipedias were rare, and each evolved independently. 99 a prominent location on Slashdot. This resulted in heavy traffic to the website, and the arrival of a relatively large body of new participants. These participants contribution to the project proved vital in following months, but they also brought unrest. Some of those new participants were reluctant to partake in the effective consensus gathered around established policies. As new decisions had to be made on important issues, they began challenging the leadership of Larry Sanger, who was the one actually initiating and practically making most of them. Tensions in the project began to rise, but the production of content went on in a rapid pace. On July 27th, Sanger announced that with the help of these new participants, Wikipedia passed the 7000 articles benchmark (Wikipedia, 2001a). At the same time, the surge of traffic to the website caused rose concerns about Wikipedia's UseModWiki software ability to handle the project's needs. Thus, discussion of a "phase II Wiki" began. On august 24, Magnus Manske, a contributor to both Wikipedia and Nupedia, announced that he wrote voluntarily a PHP code for a new version of Wikipedia (Manske, 2001). A day later it was uploaded to a test site, to be tested and debugged by the Wikis editors, who reported bugs and requested features on Wikipedia-l and on a special Wikipedia page. After a period of testing and improvements Wikipedia moved to the new software on January 25th, ten days after its first anniversary. It was further improved and debugged by Wikipedia editors in the following months (Wikipedia, 2002a). The surge of new editors in July was, in retrospect, only a precursor for what was to follow in September, a groundbreaking month in terms of notability and traffic. On September 4th, MIT's techreview.com website featured an interview with Sanger and Wales about Wikipedia. On September 10th, wired magazine gave a link to the techreview.com interview, generating more traffic. In September 20th, an article on the project appeared in the New York Times (Meyers, 2001), and republished again on the 24th by the International Herald Tribune. On the same day, Kuro5hin.com featured another article by Sanger on the project (Wikipedia, 2001a). This story was picked up by technically oriented blogs, which generated more traffic. While these reports had a major impact on Wikipedia, they were overshadowed by this month's dramatic events. On September 11th 2001 – now universally known as "nine eleven" – Wikipedia's participants (almost all of which 100 were Americans) followed with anxiety the news and reports of the attacks and their aftermath. A Wiki article named "September 11 attacks" (the earliest available version of this page can be found in Wikipedia, 2001d) was immediately created (even though Wikipedia did not cover news events up to that point), and was updated rapidly as events unfolded. An editor known as "The Cunctator," one of the new participants that arrived after July's slashdoting, orchestrated and contributed much of the work on the article. The article soon became the basis for an extensive project covering a plethora of subjects, ranging from regular articles about issues related to the event to a minute by minute timeline of the events, a list of casualties, survivors and missing persons, summary of political and economic effects, calls for blood donations, etc. This project was a tour de-force, giving a clear example of Wikipedia's potential as it became a highly useful resource for surfers looking for summaries of the event and background information. It was linked to from various websites, generating more traffic. In September and Octobers traffic to Wikipedia developed very quickly, both due to increased publicity, and to a sharp increase of traffic coming in from google.com. This surge brought with it an increase in article production rate, as Wikipedia reached the 10,000 articles benchmark in September 7th, 11,000 in September 19th, 12,000 in October 4th and 15,000 by October 25th (Wikipedia, 2001a). The surge in traffic and production caused some server problems, which led to hastening the work on the new software, but its main effect was a surge of new editors unfamiliar with Wikipedia's norms, and also of vandals who defaced and erased content. In response, Sanger formed in November 29th "the Wikipedia militia," a group of volunteers that would make sure that surges of traffic will not cause abuse to the project. This move stirred up serious controversies (Wikipedia, 2001e), which were rooted in deeper disagreements about authority and social relations in the project that had already caused conflicts between Sanger and some of the contributors. As noted above, Sanger had a directing role in the project, but he constantly stressed the important of community consensus about policies, and had no means to impose his stand against divergent views. As new editors that did not agree with the consensus started objecting, increasingly heated and unresolved debates about policy became commonplace. Sanger felt obliged to enforce his opinion against those of his 101 opponents, and began taking a more active role in policy setting and enforcement, further escalating conflicts with his opposition110. There were numerous points of friction, all connected to the same basic issues concerning Wikipedia's authority structure. Notable controversies broke around issues such as the "neutral point of view" policy (Wikipedia, 2001f), Sanger's deletion of articles containing dictionary style definitions (Wikipedia, 2001g); Wales' idea to equip a group of Wikipedia veterans with anti vandalism tools not available for ordinary users (Wales, 2001b); Sanger's deletion of content posted by users to their personal userpages (cf. Kissane, 2001; Meta-Wiki, 2001a, 2001b); the decision to move commentary and discussions about Wikipedia to a separate Wiki, called metaWiki (Wales, 2001c); the decision to exclude subpages from the phase II software (Cunctator, 2001), etc. quickly turning into personal battles abundant with mud smearing, they destabilized Sanger's position in the project, and forced Wales to take a more active role in policy setting. This resulted in Wales' posting, for the first time, a statement of principles outlining his vision of the project and his role as a final arbitrator in it (Wales, 2001d), but even this failed to put an end to skirmishes. This series of controversies came to an abrupt end when Larry Sanger was fired from Bomis. His salary was reduced in December and January and he was ultimately fired in February. In his announcement about his being fired, on February 13th, he stated that he will continue to be involved in the project – and that he hopes that a plan to place advertisements on Wikipedia will allow him to get his old job back (Sanger, 2002a). That statement stirred up such a controversy in a community largely opposing to advertisements on web pages, that the Spanish Wikipedia soon forked, copied all their content and started a mirror project under the title Encyclopedia Libre (Tkacz, 2011). Less than a month later, on March 1st, Sanger announced his total resignation from the project, leaving Wales as the sole authority in Wikipedia (Sanger, 2002b). Surprisingly, perhaps, Wikipedia's production went on virtually undisturbed, as it reached the 20,000 articles benchmark on January 9th, and 23,000 by February 4th and 27,000 by March 12th (Wikipedia, 2002a). Press coverage and links from websites accumulated, as well as traffic from Google. In retrospect, however, it is amazing how small and underdeveloped Wikipedia was at the time. 110 For a summary of Sanger position and consequent responses, see Meta-Wiki (2001a, 2001b). 102 Most of the articles were very far from the standard set by traditional encyclopedias, and basic topics were still very far from comprehensive coverage. Traffic was still very low, as it was only in August that Wikipedia reached an average of about 20,000 hits per day (Wikipedia, 2002b). The number of regular active editors was still below 200. The project had an uncertain future – but also the foundations that will enable it to grow. 103 Chapter 5: Encyclopedising the Wiki Technology As noted in Chapter 1, accounts of Wikipedia usually attribute its success to either the "wisdom of the crowds" (Surowiecki, 2004, Benkler, 2006, Tapscott and Williams, 2006), to "community norms" directing the production of knowledge and the relations between various participants in the community of practice producing it (e.g., Reinhardt, 2003; Bryant et al., 2005; Rafaeli, Hayat and Ariel, 2009; O'Sullivan, 2009; Reagle, 2010a), or both (Kittur et al., 2008; Niederer and van Dijck, 2010). All these explanations stress the centrality of Wikipedia's technology and the radically collaborative production process, allowing anyone to edit Wikipedia's articles at any given moment and encouraging multiple editors to work together on a same article. However, these explanations neglect to address the fact that this production process was not present upon Wikipedia's inception, and is rather an accomplishment achieved by its founders and early participants. The feasibility of the project was originally doubted by many, including the inventor of the Wiki software (Lih, 2009, p. 75) and Wikipedia's own founders (Sanger, 2005). The main concern was the radically collaborative production process of Wikis at the time111, which stood in stark contrast with traditional encyclopedia production, based on individual writers, postfactum editing and expert review. Wikipedia thus had to develop procedures governing radical collaboration and mechanisms of epistemic governance, which would enable the construction of both community of practice and underlying technological and managerial dynamics (Niederer and van Dijck, 2010) needed for mass participation. Both the Wiki technology and its production processes had to undergo considerable changes to accommodate the unique, evolving needs of the project. In the following chapter, I will use a symbolic-interactionist (Clarke and Star, 2003) and constructivist (KnorrCetina, 1981) framework to explain the development and institutionalization of policies and mechanisms governing collaborative content production in Wikipedia. This process, initiated as a response to concrete, local problems and conflicts, 111 The production process of earlier Wikis was different from the one subsequently developed in Wikipedia. 104 changed both the Wiki technology and the encyclopedic model, and created a new epistemic culture that would prove crucial to Wikipedia's success. Wikipedia in its early history was in what Thévenot termed a critical situation (Thévenot, 2001a, 2002a, 2002b): a situation that is not subject to a single mode of evaluation and requires social negotiation of its overarching principles of coordination. Such negotiations emerged in the attempt to establish policies that would regulate radical collaboration and solve problems stemming from its inconsistencies with the traditional encyclopedic model. Thévenot suggests that coordination in such situations is heavily dependent on a process of form investment (Thévenot, 1984), creating equivalences across time and space between various practices, objects, persons, etc. Its successful completion enables divergent parties to cooperate in such critical situations, and eventually surpass them. The importance of this analysis lies in its emphasis on the important of external cultural factors, incorporated to the situation through this process in the form of pre-existing cultural modes of coordination. However, in contrast to Thévenot I emphasize the importance of habitual activity even in such critical situations: following the pragmatist school (Joas, 1993,1996; Gross, 2009), I observe the ways in which habit and creativity alternate in processes of problem-solving and institutionalization Moreover, while Thévenot addresses general or universal modes of coordination or "orders of worth" (such as domestic, market or industrial modes), my focus here is on a much more localized cultural factors and modes of coordination, based mainly on the model of the encyclopedia and the Wiki process. In contrast with Thévenot universal modes of coordination, Laclau and Mouffe (1985) emphasize the highly contingent nature of the process of articulation, rooted in social antagonisms. Laclau and Mouffe (1985, p. 105) define articulation as "any practice establishing a relation among elements such that their identity is modified as a result of the articulatory practice," resulting in the "structured totality" of a new discourse. The importance of their approach for this work as indicated by the antagonistic undertones of this conceptualization, this process took place amidst conflicts between various individuals and groups with conflicting visions of the project; moreover the coordination of action was also accompanied by mechanisms of social control, enactment of authority, and exclusion of the opposing party from the 105 project. However, describing these processes and their result in terms of "discourse" would be misleading, as Wikipedia was not a structured totality, but rather a much more flexible structure never achieving the level of totality, but rather a dynamic meeting point of several discourses, or social worlds. To describe how mechanisms enabling such a dynamic tension between various positions were constructed, partially stabilizing their relations whilst creating a semi-hegemonic dynamic structure, I will utilize the framework developed around the concept of translation in science and technology studies. (Callon, 1986, Latour, 1988, p. 43; Bijker and Law, 1992, p. 31), In this process, various actors with divergent viewpoints and interests are brought together and enlisted in favor of the common construction of a epistemic projects 112. In the original Callon-Latour-Lawmodel of translation, the translation culminates with the creation of a definitive authority or Obligatory Passage Point (OPP): an actor, a problem, a procedure, etc. that funnels the actor-network and aligns the actions of all the actors connected through (or to) it. This model was later extended by Susan Leigh Star (Star and Griesemer, 1989, Bowker and Star, 1999, Star, 2010), as she moved the focus to situations where "several obligatory points of passage are negotiated with several kinds of allies"113. The primary results of these negotiations are methods standardization, the coordination of the activities of various allies to make their products compatible through a common standard, and the creation of boundary objects114. Boundary objects are "[scientific] objects which both inhabit several intersecting social worlds…and satisfy the informational requirements of each of them… [they] are both plastic enough to adapt to local needs and the constraints of several parties employing them, yet robust enough to maintain a common identity across sites" (Star and Griesemer, 1989, p. 393). Following Star, I emphasize the interplay between policies enabling method standardization and boundary objects related to them. However, while in Star's model the actors involved in their creation represent distinct social worlds, the participants in Wikipedia's case were more 112 Thévenot (2002b) noted that translations, (as well as boundary objects discussed below) are Parallel to his term "compromise," I preferred the concept of translation as it does not presupposes that its subjects are general orders of worth. In contrast with Laclau and Mouffe's conceptualization, they transcend the primacy of discourse, treating it as only one of several elements involved in this process. 113 Another important aspect of this move is the attribution of agency mainly to human actors who do the actual negotiations, rather than distributing it equally to humans and non-humans. 114 The term has since been applied in a wide range of topics, and became prevalent in the realm of technology studies (cf. Flichy, 2007a, 2007b; Fujimura, 1992; Henderson, 1991). 106 individualized, their relevant identities harder to define, and their constant influx demanded constant, recurring acts of translation. These challenges were compounded by Wikipedia's technological and spatial make-up. Wikipedia began as a bare network of text pages, each dedicated to an encyclopedic subject. The radical collaboration process meant that in principle, at least, alignment had to be achieved regarding each individual page, as each was a boundary object in itself, edited by various actors with potentially conflicting interests. As it lacked such elements as a registration form, terms of use document, or selection mechanism, Wikipedia did not have any obligatory passage points that could regulate selection, socialization and guidance. It had no built-in surveillance and enforcement mechanisms that could help deal with problematic edits. This situation created in Wikipedia an epistemic culture that is even more flexible and polysemic than the ones described by Star and her co-writers. Having no OPP's that can commit a large amount of allies to a common pattern of action or generate stable methods standardization, it had to generate softer measures of social coordination, less authoritative and strict than those found in scientific contexts, but still effective enough for the project to function. As I shall describe below, this type of coordination was achieved in one of two ways: in policy pages, special, highly visible sections of the network that were systematically and strategically linked to, and thus made accessible from various nodes in the network; and in the software code arranging the very form and attributes of the website and network of web pages it contained. None of these methods, it should be noted, was effective enough to be considered a genuine OPP (Obligatory Passage Point). While certain participants (namely, Sanger as Wales) and cultural schemes (mainly the encyclopedic model) did have privileged position in negotiating standards and boundary objects, authority in the project never came near the power it held in scientific projects of the kind described by Latour, Callon, Law or even Star. Thus, I would rather call those passage points that funneled Wikipedia's activity privileged (rather than obligatory) passage points115. 115 I will discuss this concept further in the last chapter of this dissertation. 107 In what follows, I first describe the initial actions and interactions which led to the formation of a mode of production based on radical collaboration and extremely liberal standards for inclusion of content. I then go on to describe the construction of two policies that were developed as means of make this form of production suitable for the task of creating an encyclopedia: The purification of article pages and Neutral Point of View. The first of them created a special category of Wiki pages referred to as article or subject pages, where writing is to be styled after traditional encyclopedia articles116. It also specified that discussion of articles should not take place in the article pages themselves. This policy was the most essential in domesticating the Wiki technology as it set a precedent that content policies should be based on traditional encyclopedic standards, even if those contrast with customs prevalent in other Wikis. The second major content policy formulated in Wikipedia enjoined a neutral point of view, still noted today as one of Wikipedia's hallmarks, and considered as a major reason for its success (Reagle, 2010a). Created originally as means to enable the joint work of participants with conflicting views, it dictated that that the opinions of all parties to the disputes should be presented in the article in a manner they can agree on. It was the first and only policy to be declared "non-negotiable" (Wikipedia, 2012f), and subsequently became an OPP (or nodal point117) for almost all later major content disputes. While other such policies were created at the time, these policies, and the privileged passage points constructed to sustain them, became the backbone for Wikipedia's epistemic governance Through the institutionalization of these policies, the following principles of governance rose to reside over the process of coordination: the primacy of the traditional encyclopedic model as a "standard for standards"; the legitimacy of the Wiki process as a resource for schemes and practices (where these do not contradict the encyclopedic model); radical collaboration and lack of hierarchy in the writing and editing process; the reliance on effective consensus between editors to establish policy; the dialogical nature of knowledge construction, and the reliance on references 116 The distinction had to be socially constructed and enforced, as Wikipedia's Wiki software did not differentiate between types of wikipages. The differentiation of namespaces did not take place until the launch of the "phase II Software" in January 2002 (See below). 117 Nodal point, a term originally coined by Lacan, is a master signifier – "a particular element assuming a 'universal' structuring function within a certain discursive field – actually, whatever organization that field has is only the result of this function – without the particularity of the element per se predetermining such a function" (Laclau and Mouffe, 1985, p. xi). 108 to other texts rather than on expertise and original research to establish knowledge; and finally, the preference for productivity and content quantity over control and quality. They were also heavily implicated in the construction of PPP's in the project, setting precedents to later attempts at translation of that sort. 5.1 Phase 1: Setting the Ground Rules Wikipedia was born as an experiment in collaborative content creation, with very little idea of what actually to do with such content once created. The text placed on its homepage during its first days first read This wiki is an experiment. But, for those who might be confused about this point, it is '''not Nupedia'''. Nupedia is a serious encyclopedia project found at http://www.nupedia.com. This wiki is a proposed "fun" supplement to Nupedia! (Wikipedia, 2001h) Wikipedia's first participants were a small but heterogeneous group, comprised of Nupedia editors, a few participants that arrived from other Wikis, and a group of students from a Human-Computer Interaction class at Lewis and Clark College that was sent by their teacher. As Wikipedia had no institutionalized structure or practices at the time, most of them simply played with it, creating and editing pages at random without giving much thought to the consequences of their actions. Larry Sanger, on the other hand, quickly began attempting to steer the project towards a productive path, by writing various pages explaining the tradition developed in Wiki website (often referred to as "Wiki culture"), its processes and its goals. His first move was to create conscription devices (Henderson, 1991): mechanisms enlisting group participation and coordinating production practices by invoking collective knowledge created and articulated through social interaction. A page titled Wikipedia Process (Wikipedia, 2001i) was started in January 17 was to discussions between participants regarding the Wikipedia's functioning, spanning all sorts of topics.118. Another page, titled how does one edit a page? was erected on the same day as a 118 While Sanger was the most frequent writer in it, stating his opinions regarding many of the questions raised, he was not alone in this: some other editors, especially those arriving from other Wikis, took the project seriously enough to raise initial policy issues and attempt to offer solutions. The topics discussed were highly varied, including naming conventions for articles, adding images to articles, Wikipedia's relations with the FOSS movement, and how to prevent disruptive or problematic participants. 109 tutorial to Wiki-culture, explaining the established practices prevalent in other Wiki websites at the time, and linking to pages concerning more specific issues. While most of the instructions in how does one edit a page? and the pages it linked to were rather technical, one such page named be bold in updating articles became a major policy addressing the issue of collaboration in Wikipedia, and arguably the most important of them. It was initially formulated by Larry Sanger in January 17, mainly as an explanation to how Wikis work in general. Initially, it simply stated that "Wikis don't work if people aren't bold. You've got to get out there and make those changes, correct that grammar, add those facts, precisify that language, etc., etc. It's OK. It's what everyone expects. Amazingly, it all works out. It does require some amount of politeness, but it works. You'll see" (Wikipedia, 2001l, bold in the original). However, as two days later Sanger added another section to the page, this time much more prescriptive in its tone, stating: "OK, you folks still don't understand, so let me make this clear: if someone writes an inferior or merely humorous article or article stub, don't worry about their feelings. Correct it, add to it, and if it's a total waste of time, outright replace it! That's the nature of a WikiWiki, and long live Wikipedia!" on the same day, Sanger posted a link to the policy from Wikipedia's homepage (the only policy to receive such an honor), making it highly visible and giving it special significance in the project. Being bold in updating articles is, in some respects, a translation of practices already institutionalized in Wiki-culture. However, it took on new meaning in Wikipedia: in former Wikis, it was customary to alternate between discussions and summaries of them (see below in the next section), and everyone was welcome to edit summaries they feel did not represent their view. In Wikipedia, on the other hand, editing each other's articles became a tool for managing disagreements without discussions. It thus became the basis of a new mode of production, based on radical collaboration. This mode was subsequently inscribed in each and every page in the project: instead of the standard "edit" link appearing at the bottom of pages in Wikis at the time, Wikipedia's founders placed at the bottom of every page in the project a standard text reading "you can edit this page right now! It's a free, community project" (with a link leading to the edit interface). 110 This radical mode of production stood in stark opposition with Nupedia's rigorous editing and deliberating mechanisms. In a sense, this was a rejection of the open-source gift economy as a form of peer review (Bergquist and Ljungberg, 2001), as direct editing were to take the place of comments, criticism and dialogue. It also denied the relevance of hierarchy and expertise, as it gave participants a free hand in their involvement with each other's contributions. Everyone could participate in this collaborative authoring experiment. It didn't even demand a stable online identity, let alone the divulgence of real-life identity. This radical form of collaboration, which now seems so essential to Wikipedia's process, was not self-evident upon its inception. In order for it to become self evident if first had to be introduced and taught. This is evident not only from the creation of conscription devices to promote it, but from Sanger's need to reiterate it and elaborate it two days after it was first posted. This seems to indicate that editing each other's texts was far from self evident even after this policy was first declared. Moreover, the events of the following days proved that even this reiteration was not sufficient to encourage radical collaboration. On January 21st, after various nonsense and obscenities continued to be posted on the Wiki, some Nupedia users started posting septic remarks questioning the use of the anarchic and chaotic Wiki and the wisdom of radical collaboration on the Wikipedia Process page. In response, Sanger attempted to supplement the be bold in updating pages policy and put things into order. His thus posted on the Wikipedia Process page the suggestion "I propose that we delete Patent nonsense when we run across it". Fifteen minutes later, Sanger started the page patent nonsense (Wikipedia, 2001j), where he distinguished between legitimate content and content that had no place in Wikipedia: Patent nonsense (to be carefully distinguished from brilliant prose as well as private nonsense), particularly on a wiki, falls into a just few general categories: * Literal nonsense, i.e., text that, usually intentionally, has no assignable meaning at all. This tends to be created after the consumption of too much adult beverage, for example, or under the influence of immaturity and/or stupidity. (The writings of James Joyce being the obvious exception). 111 * Stuff that, while apparently meaningful after a fashion, is so completely and irremediably confused that no intelligent person can be expected to try to make heads or tails of it. (The writings of James Joyce being the obvious exception.) The following, while often regrettable, is not patent nonsense, and therefore according to a proposed rule on the Wikipedia process page, should not for ''that'' reason be deleted (but some suggestions for improvement are included): * Really badly-written stuff. (So correct it.) * Partisan screed. (Make it hedged or add qualifications in order to make it less biased.) * Religious excogitations. (Make it factual. Yes, there are ways of doing this; e.g., add phrases like "Followers of X hold that...") * Mere opinion, masquerading as fact. (Remove it to an appropriately-named page, or reword in order to make it factstating.) * Incompetent and/or immature stuff. (Well, it's not nonsense, but perhaps it does deserve to be greatly improved. So improve it.) As the quote above shows, this move towards creating an authority structure in the project was done very carefully: the suggestion was made with humorous and light-headed wording, deletion was to take place only in the most extreme cases, and personal jokes (Private Nonsense) were explicitly allowed119. When Sanger actually began implementing his suggestion, he even started a page called "bad jokes and other deleted nonsense" (Wikipedia, 2001k), and offered to archive the deleted texts in it. While in principle, this crux of this policy regarded the deletion of content, more than half of the text dealt with content that should not be deleted. This part of the policy was especially important, as it supplemented and completed the be bold in updating articles policy by explaining what should be done with presumably inappropriate content. The stance it reflected, which had a deep and lasting impact on Wikipedia's epistemic culture, was quite simple: recycle and reuse any content that is not completely worthless. This stance meant that Wikipedia, as a rule, should accept almost any edit, whether it is well written or not, biased or not, substantiated or not. 119 Though this was a later addition by an anonymous editor, Sanger implicitly supported it as he left it on the page after editing it multiple times. 112 In contrast with the be bold in updating articles policy, which was articulated as a description of practices derived from prevalent Wiki-culture120, in the patent nonsense case Sanger avoided any kind of justification (Boltanski and Thévenot, 2006) or form investment based on existing cultural models. This occasion marks an initial attempt to shape Wikipedia as an autonomous project, independent of Wikiculture. However, the lack of articulated justifications for this policy – even in narrow terms such as "Wiki tradition" – was a significant precedent. This lack of theorization may be seen as an initial step on a path that led Wikipedia to its current position as a project that "only works in practice" (Wikipedia, 2006). Sanger's actions soon proved themselves as effective, as the Wiki began settling down. By the end of the month, his and other editors' attempts at removing "Patent nonsense" and creating actual articles were successful enough to demonstrate that Wikipedia did have a potential to be a productive and promising endeavor. While a large variety of text formats were submitted to Wikipedia's pages, many of them not resembling encyclopedic articles by any measure121, "patent nonsense" became rare phenomena, and almost all the texts submitted to Wikipedia pages were related, in some way or another, to the effort of creating an encyclopedia. This situation conformed to the initial vision of the project's founders, who explicitly rooted for experimentation and preferred quantity over quality, at least in these early days. As there was no real opposition to Sanger's proposals, they became an "effective consensus" – not necessarily practiced by everyone, but considered legitimate and practiced by at least a considerable part of the participants. By affirming Wikipedia's aim (creating "fact stating" encyclopedia articles) and methods (collaborative editing and deletion of "nonsense" texts), Sanger set the terms which all later policy developments in the project have to accommodate. This move constructed Wikipedia process and how does one edit a page? (along with their subpages) as scopic media, "centering and mediating device[s] through which things become assembled and from which they are projected" (Knorr-Cetina, 2005). They functioned as conscription devices (Henderson, 1991), enlisting group participation and 120 And was thus an instance of form investment (Thévenot, 1984). Wikipedia included, for example, a large number of lecture notes from a philosophy course Larry Sanger taught (Wikipedia, 2001b), Tim Shell's extensive description and analysis of Ayn Rand's Atlas Shrugged (Wikipedia, 2001c), copies of various states' constitutions, parts of the bible, personal commentary pages and personal user pages. 121 113 coordinating production practices by invoking collective knowledge created and articulated through social interaction. This bundle of pages created a seed for a core network inside the larger network of Wikipedia pages which veteran participants knew they should check into if they were interested in policy and policy debates, and which could be used to educate newcomers. Shortly after, however, the Wikipedia process page lost its prominent position as a conscription device. The abundance of discussions, the wide diversity of topics (which included important policy issues, technical discussions and trivial comments), and the lack of an established format for posting and interactions made it ineffective. Thus, when a series of problems and controversies – which will be described below – emerged which demanded discussion and decision, Sanger opened a new page specifically for policy discussions, called Rules to consider. Its first paragraph read: Since this is a wiki, there are no editors. We must rely on developing our own good habits and occasionally taking a bit of time to correct the results of someone else's bad habits. But it might help to specifically enunciate particularly rules that some of us wish we'd make an effort to follow. So here's a page containing such rules. Two suggested features of this page are: add your name to a list of the rule's "supporters" to get an idea of how strongly Wikipedians support a rule and "[nameofrule] Debate" pages where we can talk about the merits of the proposed rule. (The latter will help keep this page nice and clean for those people who are mainly interested in the rules themselves). (Wikipedia, 2001ad, italics in the original) This page, in contrast with the ones described above, was a thoughtfully designed mechanism for coordination and policy making. It soon became Wikipedia's most important venue for policy making, and remained so in the following year. Moreover, it set the tone for policy discussions and formation procedures: the procedures suggested above, as well as the articulation of the page's goal, are strictly egalitarian, as they preclude the option of hierarchy, both in editing Wikipedia and in policy setting. The articulation precludes any sort of coercion in the implementation of these policies, making it a voluntary effort dependent on community participation. Thus, Wikipedia's main policy page at the time maintained a productive tension between freedom and obligation. This tension was intensified by Sanger's decision to add at the top of the page, above all the other rules added to it, a rule titled ignore all rules, which stated that "if 114 rules make your nervous and depressed, and not desirous of participating in the wiki, then ignore them entirely and go about your business." Sanger later noted that it reflected on his decision to add this rule, stating that "the reason was that I thought we needed experience with how wikis should work, and even more importantly at that point we needed participants more than we needed rules" (Sanger, 2005). Indeed, this rule, even though it was more symbolic than practical, served to remind that in contrast with Nupedia, Wikipedia's founders favored productivity over control and oversight, and were willing to forgo any claim for legitimacy in exchange for the encyclopedia to actually being produced. This liberal stance played a major role in Wikipedia's initial growth, as it allowed it to accumulate potentially problematic resources at a time when in needed quantity more than quality. The combination of radical collaboration and a liberal and inclusive attitude towards contributed content, set it apart from Nupedia, and allowed it to succeed in amassing content whereas the latter failed. As noted above, Nupedia's insistence on cultural legitimacy and control mechanisms, which were meant to attract experts to partake in its gift economy "crowded out" (Benkler, 2010) spontaneous, individual giving. Wikipedia's liberal stance, an antithesis of Nupedia's policies, catered exactly such individual giving, as participants were encouraged to contribute immediately, and anyway they saw fit. Moreover, Nupedia was undermined by the considerable delays in the process needed to complete the act of giving: the registration process, the articulation of a complete, good enough article, the length of its copyediting and peer-review processes, etc. in contrast with these delays, as well as to the similarity of the submission process to that of school chores (Lamb, 2005), Wikipedia's automatic acceptance of contributions allowed givers to complete their act immediately. The swift appearance on the website accorded the giver actions with instantaneous worth, and embedded his content – if not his persona – in the project. Thus, This immediacy did not only encourage individual giving, it was also crucial in soliciting socially embedded giving, as such spontaneous edits are usually the first steps towards becoming a socially embedded Wikipedian (Bryant et al, 2005). At the same time, these steps did not crowd out socially embedded contributors, as they provided a context through which an environment sustaining a 115 viable gift economy can emerge. Though extremely liberal, be bold in updating articles and patent nonsense represented initial attempts at creating orders of worth (Boltanski and Thévenot, 2006, Thévenot, 2001b), standards for evaluation of actions, and epistemic virtues (Wood, 1998; Daston and Galison, 2007) – all of which lay the basis for locally embedded social recognition and reciprocity. The following sections will outline the intensifications of such efforts, focusing on problems and controversies related to the radical production method constructed in the events outlined above. 5.2 Phase 2: Purifying Article Pages: the “Refactoring Debate” Around the end of January 2001, the actions of Sanger and a few other editors made the creation and editing of pages on encyclopedic topics the prominent activity in Wikipedia. Sanger's suggestion that editors should delete patent nonsense (Wikipedia, 2001j) and be bold in updating articles (Wikipedia, 2001l) were well received, and participants began treating Wikipedia as an experiment in encyclopedia production rather than just a "'fun' supplement to Nupedia". Before long, however, disputes began over the content of articles, conducted in first-person remarks added above or below the disputed content – and in some case completely replacing the relevant article. This development is not surprising, even though it may seem as conflicting with prevalent portrayals of Wikipedia as a harmonious and cooperative community. While Nupedia's mechanisms for discussing and deciding disagreements were detached from the actual articles, Wikipedia's founders' insistence on immediate publication precluded such options, and forced them to chose alternatives that will enable post-factum discussion while preserving the integrity of article content. Initially, Sanger and Wales decided to use the model prevalent in Wikis at the time. Wikis (and especially the ones that inspired Wikipedia) usually emulated the form of threaded discussions in mailing lists and internet forums, and much of their content was written in the first person and signed by their author. Like many communities created in such environments, Wikis were communitarian projects based on a gift-economy which bestowed participants with recognition and reputation (Rheingold, 2000; Lih, 2009). However, as discussions grew longer, respected 116 participants often summarized the discussions to create "documents" which would be better organized and readable. Thus, the use of the first person discussion was essential to the Wiki process. At the same time, however, the use of first person and dialogue stands in contrast with the established model of encyclopedia articles. The question of how to deal with this conflict was the heart of the conflict later known as "the refactoring debate". Through this conflict, Wikipedia's participants made the first crucial steps towards articulating the Wiki process and the model of the encyclopedia, at the same time changing their use and meaning. Moreover, this conflict set a crucial precedent for the development and institutionalization of policies in Wikipedia. On January 26th, AyeSpy posted on the Christianity (Wikipedia, 2001m) article a few paragraphs of musing about religion, politics and the distinction between eastern and western religions, giving it the title "Commentary." In response, Sanger deleted these paragraphs, opened a new page titled Christianity/Talk (Wikipedia, 2001n) and moved them there, leaving a note on the original Christianity page stating: Commentary moved to Christianity Talk. Let's avoid stating idiosyncratic views on the main pages. Put those for *Talk, *Discuss, or whatever. While this incident, in itself, was unrelated to the problem of discussions in article pages (as there was no discussion here, but rather an addition of content by a single user), and was not accompanied by a form investment or justification, it produced a standard new practice that was later transported and translated into a new context, where it was entangled with other practices and justifications to become a new content policy. The existence of disputes in article pages first came under scrutiny in a Wikipedia-l post on January 27th (one of the first sent to it since its opening on January 21st) by Tim Shell, a partner in Bomis, where he commented that: If Wikipedia is to be an encyclopedia, then it probably is not appropriate to have threaded discussions on a subject page. See for example Altruism, where one person gives a flame bait description of the concept, and numerous people then argue back about that description. If a discussion is appropriate, perhaps there should be a standard discussion page, as Altruism Discussion or Altruism Debate that is linked to from the subject page. (Shell, 2001) 117 As its opening sentence indicates, Shell's attempted to de-legitimize a prevalent (or at least emerging) practice by making an explicit form investment (Thévenot, 1984): pointing out its inconsistency with the encyclopedic model. He then proposed a practical alternative, based on the technical features of the Wiki (namely the ease of opening new pages and moving content from page to page), that could enable Wikipedia to stay consistent with that model. While Shell's suggestion may seem obvious in retrospect, it was not at all obvious when it was presented. In fact, in the debate that followed, both on the mailing list and on a Wikipedia page named Refactoring policy (Wikipedia, 2001o) opened by Sanger to further debate the subject, the majority of participants had reservations about the suggestion, or rejected it outright. However, before going on to describe the debate's content I would like to describe Shell's actions and their consequences on the issue of policy creation in Wikipedia. The strategy Shell used to tackle what he thought of as a problem in Wikipedia was the first attempt at creating a mechanism to debate and rectify such problems. Its focus was the Wikipedia-l mailing list, which he used as means for propagating his articulation of the issue. Through his reference to the altruism (Wikipedia, 2001v) article, Shell built a connection from a peripheral node in the Wikipedia network (i.e. the article) whose exposure to other participants was limited to a medium which reached the project’s founders and major participants in their private email inboxes. This move constructed the mailing list as a conscription device122 (Henderson, 1991). While Wikipedia already had the beginning of such a system – the Wikipedia process (Wikipedia, 2001i) page and its various subpages – Shell was either unaware of its existence or simply preferred the mailing list format with its convenient threaded discussion feature and established conventions. Initially, this move seemed successful as participants started replying to Shell's comment and debating the subject. However, a day later Wales posted a message stating that "I supported the creation of this list, but it occurs to me that quite possibly discussions which take place here ought to be taking place on Wikipedia itself" (Wales, 2001f). As a result, Sanger established the primacy of the former scopic 122 As noted earlier, conscription devices are boundary objects that enlist and organize group participation. 118 medium by opening the Refactoring policy page to continue the debate, and linked to it from the Wikipedia process page. The discussion of Shell's suggestion was carried on mainly in the Refactoring policy page, and new pages dealing with other policy issues soon began appearing, interlinked through the Wikipedia process page and later through a new page substituting it, titled Wikipedia policy (Wikipedia, 2001p). This method had some advantages over the mailing list as a conscription device, as the creation of connections (links) in it was much simpler, and the movement through them much faster. Its existence prevented the need to re-articulate policies in each new dispute, as participants versed in these policies could just link to them when conflict appeared in one of the Wiki pages rather than opening a discussion on the mailing list. Moving now from the mechanisms for policy debate to discuss its substance, the reasons for rejecting Shell's suggestion to move discussion to special pages were rooted in the Wiki process: conventions developed in existing Wikis regarding the right way to use them. Participants that were acquainted with such conventions, such as Jimmy Wales123, Bryce Harrington and Clifford Adams124, rejected Shell's suggestion claiming that mixing discussion and content was a crucial element in the "Wiki process"125. As Wales stated, There is already a cultural tradition in the wiki world called "refactoring". The idea is that there is a "discussion mode" and a "document mode". In discussion mode, people have a threaded discussion, with many different issues being raised. Then, some WikiMaster comes through after the discussion has died down, and "refactors" the page. This involves editing/rewriting/rearranging, so that all points of view are presented fairly. (Wales, 2001g) Thus, the refactoring debate was a conflict between two contrasting modes of coordination, or investments of form. While Shell justified his position through a contrast between prevalent practice and the encyclopedic model, his opponents 123 While Wales originally objected Shell's position, he later took a more moderate position claiming that the decision to keep the discussion on subject page or move it to a Talk page should be contextual, depending on factors such as its length, inflammatory potential etc. (Wales, 2001e). 124 Author of the UseModWiki which was used by Wikipedia at the time. 125 This process, integral to the concept of the Wiki at the time, but problematic from the point of view of an encyclopedia process, was one of the reasons for Cunningham's original skepticism towards the project (Lih, 2009, p. 75). 119 justified theirs through basing it on the institutionalized Wiki process. The debate was essentially over what form should have primacy in Wikipedia. What tipped the balance in the discussion was Larry Sanger's support for Shell's position. While he was not an "editor-in-chief" in Wikipedia (as he was in Nupedia), Sanger did have a measure of authority in the project – mostly due to the fact that he was the most active participant and the most enthusiastic about policy making. Sanger stated in several places that discussions in general are unproductive, and that the best way to work on Wikipedia is simply edit articles. However, he agreed with Shell that if discussions were to take place, it was better to put them in separate pages. Thus, in response to Wales' position on the mailing list (Sanger, 2001c) he explicitly rejected the relevance of Wikis' history, claiming that "I'm not sure that the history of wiki is relevant here, because we are, after all, making an attempt at creating an encyclopedia…" He later reiterated his position on the Refactoring policy page, where he explained that "There's an excellent reason for Wikipedia in particular to encourage the practice of moving discussions to separate pages: the pages are supposed to be encyclopedia articles, not discussions." In another post on the same page, he explained that: […] mainly what I'm concerned about is that people will start treating Wikipedia as yet another discussion forum, rather than an encyclopedia. As I see it […] all that emotional energy that goes into hashing through partisan and other issues could much more easily be channeled toward improving articles. […] The purpose of the wiki, as I see it, is--well, it's whatever people make it--but what I'd like it to be is the creation of good encyclopedia articles. Underlining this concrete discussion lay a more philosophical disagreement, concerning the relations between technology and culture. The supporters of refactoring claimed that Wikis, as sociotechnical ensembles126 that include social aspects as well as technological ones, comprised both the code supporting the website and standard methods for using it. If these methods were overlooked, it wouldn’t function properly. The encyclopedic model, on the other hand, was constructed as flexible enough to include the option of discussions – at least temporarily. In contrast, Shell and Sanger supported a position which saw the Wiki as a flexible tool that can be adapted to various forms of practice according to the needs of its users, 126 On the notion of sociotechnical ensemble and sociotechnical systems see Law, (1991), Bijker and Law (1992), Bijker (1993, 1995a, 1995b). 120 emphasizing human agency and cultural constructs over technology. This is evident not only in their rejection of their opponents' position, but also in Sanger's invocation of the fear that Wikipedia could turn into "yet another discussion forum" – thus stressing the polysemic nature of technology. On the other hand, Shell and Sanger constructed the model of the encyclopedic article as immutable, justifying their position through the claim that "encyclopedia articles don't have discussion right there on the page itself". While there was no mechanism in place to decide between these opposing views, Shell and Sanger were the ones whose opinion ultimately won. While Sanger had the last word on the Refactoring policy page as no one continued to argue with him after a while, the most important reasons for his victory were not coercion or persuasion127, but rather practice and momentum. While the debate was going on, Sanger (and some of his supporters) kept deleting discussions from articles and copying their content to subpages labeled /Talk128. After starting Christianity/Talk on January 26th, he went on to create Bush/Talk on January 27th (Wikipedia, 2001q), History of United States/Talk on January 30th129 (Wikipedia, 2001r), etc130. This practice met no opposition, and was immediately imitated by other participants in the project who began moving comments and discussions to talk pages, making them a standard in the project. While not all users avoided commenting in article pages, the vast majority of users did, creating an "effective consensus" that became taken for granted in the project. When new participants arrived, they often saw talk pages while going through the website, making it a taken-for-granted part of the website from their 127 Some measure of authority was evidently involved in this process, as his opponents ultimately accepted the legitimacy of Talk pages. 128 The use of the word "talk" for the subpage is a reminder for Sanger's inspiration for this practice: the Nupedia subject specific mailing list (such as sociology-l) were always accompanied by a second list called talk (such as sociology-talk) dedicated to general discussions of the subject unrelated to the concrete editorial work. As Sanger feared that endless discussions of the type common on Usenet newsgroups will draw the projects focus if allowed in the editing process, he decided to give them a place of their own. In Wikipedia, he drew upon his original idea, but in a way that changed its meaning from a relief valve for editors seeking discussion to a central part of the production process. 129 At the same time Joshua Grosse, a participant who supported Sanger's position started another page called History of United States Discussion (Wikipedia, 2001t). The debate continued on both pages for some time. 130 Interestingly, the Altruism article that helped sparked the refactoring debate only got a Talk page on March 2nd (Wikipedia, 2001u), long after it became a standard. 121 perspective. And if they began discussing issues on article pages, they soon learned that it was inappropriate, as their comments were moved to talk pages131. Even though the refactoring debate set a clear precedence that the encyclopedic model should have primacy in Wikipedia when making policy decisions, its actual results were a compromise between the Wiki and encyclopedic models132: Article pages were written only in the third person and left unsigned133 and controversial or conflicting opinions were presented in the articles itself from a neutral point of view and not as assertions of personal opinion (see below). However, unlike conventional encyclopedia articles, they contained a direct link to the discussions that helped create their content134. The traditional practices related to the Wiki process were delimited to talk pages, where the central mode of writing was a threaded discussion in the first person. In these pages, refactoring remained a prevalent practice, as long and outdated discussions were summarized and deleted. Moreover, a third type emerged, in the form of commentary pages. On March 26, after he felt that too much of the writing in article involved a significant idiosyncratic note, Sanger started the Wikipedia Commentary (Wikipedia, 2001db) page, where participants can publish their opinions and thoughts that were unsuitable for both article pages and discussion pages. Subsequently, he moved texts that he felt were inappropriate to that section, and began publishing his thoughts there himself. By May 6th Sanger felt confident enough to suggest the automatic creation of talk subpages by the software (Sanger, 2001d). This suggestion was picked up in the specifications talks preceding the construction of the "phase two software" by Magnus Manske in August (e.g. Manske, 2001), and was indeed incorporated in it upon its release in January 2002, making Sanger and Shell's position irreversibly inscribed in the technological infrastructure of the project. For every Wiki page, a corresponding talk page was automatically generated, and a link to it appeared automatically whenever the article was viewed or edited. Articles and talk pages were assigned 131 Another important practical move towards institutionalization was made when WojPob came up with the idea of creating Talk pages as subpages of articles, a move that made their creation much easier and automatically linked them to the article page. This practice was later publicized by Wales, and became standard in the project (Wikipedia, 2001s) 132 With some additional input from the mechanism developed in Nupedia, see above note 125. 133 This practice was not self evident, as some users did sign their names on article pages. 134 In addition, they contain links to previous edits of the page – an automatic feature in the Wiki software installed for the project. 122 different namespaces, making the talk pages an integral part of the websites code and structure135. This move transformed the website's design into a privileged passage point: as it automated the creation of talk pages, Wikipedia's structure made them omnipresent and more institutionalized. The use of computer software enabled Wikipedia's participants to deploy a manifestation of this privileged passage point at any node in the network, without having to set one up by manual labor, translating an institutionalized practice to technological object. However, this technological object did not amount to the level of an OPP, because participants could still write in first person on article pages. As there was no approval mechanism for edits before their publication, dealing with such edits was confined to efforts after the fact. To conclude, the refactoring debate laid the ground for the way standards are created and enforced in Wikipedia: through debate of a concrete and immediate problem, a practice that was developed elsewhere in the Wiki was adopted and translated to become a socially accepted solution. It then underwent a process of form investment, aligning itself with the model of the encyclopedia, and institutionalized through discussions and policy pages to the point that it became the subject of an effective consensus. This effective consensus was maintained through routinization and through technological codification, as it was reprogrammed into the technological infrastructure of the website. This process should not be confused with the building of a consensus in functional or cultural terms, as such did not exist. The supporters of the original Wiki process approach were not all convinced, nor were they made to comply with the decision. What took place in this episode was the construction of pragmatic tools that succeeded against rival options. It resulted in the creation and enactment of authority, albeit one which had to be periodically reiterated to avoid atrophy. Its outcome was thus a semi-hegemonic dynamic structure within which opposition had to function. As we shall see, this route became typical to Wikipedia's policy development. 135 It also meant that a simple search for a term would only lead to the article pages. 123 5.3 Phase 3: The Neutral Point of View The creation of effective consensus on refactoring in February was accompanied by further attempts to articulate a new, radical form of collaboration, now partially divorced from the original Wiki-process, and the encyclopedic model. The most notable and influential attempt was the formation of the neutral point of view policy (Wikipedia, 2012f). The elaboration of this policy was much longer and more complex than the one described above, as it involved the de-legitimation of subtler and more divergent practices, and the tying together of a substantial number of conflicts. As noted in the previous chapter, the emphasis on "lack of bias" was central in Nupedia's discussions and guidelines, and was one of Sanger's "pet peeves" as an editor (Sanger, 2005). In Wikipedia, the issue of bias surfaced as a practical reality and problem almost as soon as it was launched. In its first month, it already saw a considerable number of discussions and conflicts fueled by accusations of bias, in articles such as Altruism (Wikipedia, 2001v), History of United States (Wikipedia, 2001w), Libertarianism (Wikipedia, 2001x), Anarchy (Wikipedia, 2001y), Traditional Anarchism (Wikipedia, 2001z), Democracy (Wikipedia, 2001aa), and George W Bush (Wikipedia, 2001ab). These discussions, which initially took place in the topic pages, were one of the main reasons for the refactoring debate, and were gradually moved to the adjacent talk pages after its resolution. However, this did not solve the essential problem, namely the need to enable participants with opposing views to collaborate in building content pages. The dissenting and conflicting opinions of participants in these debates offer a glimpse into the plethora of radical positions prevalent in the social milieu that fostered Wikipedia on its early days: Wales was an avowed objectivist; Tim Shell was a libertarian; Richard Kulisz was an anarcho-syndicalist; Bryce Harrington, an editor in the radical political site indymedia.com, was an American dissident; Joshua Gross was a leftist liberal; WojPob was a polish social democrat. This diversity, inherent in its nature as a global microstructure (Knorr-Cetina, 2005; Knorr-Cetina and Brugger, 2000), presented a serious challenge to its attempt to rely on their collaboration in producing articles. The cultural and political differences made attempts to reach consensus on the content of articles exceedingly problematic, and led the project's founders to try and find general ways of dealing with such views. The solution found, 124 as we shall see, was a combination of several practices developed in specific controversies, and a mode of coordination and justification rooted in a specific encyclopedic form. As we shall see, This path exemplified the way pragmatic patterns of action involving local problem solving are translated into higher level transposable schemes (cf. Sewell, 1992; Wenger, 1998), and the importance of external cultural factors, appearing in the form of pre-existing cultural forms, models and modes of coordination – such as the traditional encyclopedic form – which did not function as overreaching principles determining the outcome of situations, but were rather incorporated in them through antagonistic processes of form investment. The properties of such preexisting forms and models have an important role in selecting and transposing certain solutions rather than other options. 5.3.1 The First Controversial Articles The first attempt to solve a major content dispute took place on January 30th 2001, amidst the refactoring debate. On January 29th, Bryce Harrington started a page called History of United States (Wikipedia, 2001w) and posted a rather lengthy and critical article, which he himself admitted to be "depressing." On the very same day, Wales posted on the top of the page a comment characterizing the article as "very biased and anti-American." A few hours later, Joshua Gross posted a response saying that he thinks the article is "remarkably fair […] I think this is fairly close to what a real, objective summary (maybe a little bit of negative propaganda but nothing compared to the positive stuff you normally get) would look like. Leaves out good things, but also bad things, being a summary." This discussion was the type of dialogue that Sanger feared when he warned that Wikipedia is in danger of becoming "yet another discussion forum" and suggested proactively editing the content of articles rather than discussing them. Within the next hour, Tim Shell began to extensively edit the article in order to paint a more positive picture of American history. After reading Shell's edits, Gross added another comment noting that "instead of going ahead and adding good and deleting bad and deleting good and adding bad, and so forth, I suggest we have a separate discussion on what is an appropriate take here on History of United States Discussion. 125 Leave this as is for now, and know that it is controversial." He then started a page called History of United States Discussion (Wikipedia, 2001t), where he suggested to "carefully discuss each controversial point that comes up before writing about it" and he and the other participants started debating specific points in the article. A few minutes after Gross stated the History of United States Discussion page, Sanger started another page called History of United States Talk (Wikipedia, 2001r) and copied to it the comments from the original article page. Under Joshua Gross's comment quoted above, he added the following segment: Joshua, here is my view: I think we shouldn't aim for objectivity; we should aim for lack of bias. The way to achieve lack of bias is to explicitly acknowledge within the article itself points on which there is disagreement, and make sure that the body of the article does not betray any particular position on these disagreements. Thus, lack of bias is not achieved by striking a middle position between the views that the U.S. had a glorious wonderful history and that it was dark and evil. It is achieved by relating what objective facts everyone can agree on, and then making explicit what points people disagree about. By the way, anyone is free to change this article at any time, you know. :-) Sanger's solution to the problem was an attempt to suggest a general standardized method for the project. However, As Sanger's suggestion came after the issue was already being resolved through direct interaction between the parties to the controversy, and as it did not conflict with that process, no opposition was voiced, and participants soon returned to discussing how to actually write the article. In fact, Sanger's view had no real bearing on the case, as the original article was abandoned rather than fixed. The debate started on the History of United States Discussion page soon sprawled over many articles as participants began breaking up the article to smaller topics and historical periods, and writing them in some details rather than engaging in broad characterizations of the kind Harrington's original article included. The original article was deleted from the page a day later, replaced with a list of links to the various articles and discussions started as a result of the discussion (Wikipedia, 2001ac), and work on an overview article was discontinued136. 136 It was resumed after a few months up by new users who weren't involved in this episode. 126 The outcome of this controversy was a practical policy, or a sequence of actions for dealing with conflicting views: moving discussions and controversies to separate talk pages, and splitting a big controversial subject into multiple smaller ones. What was still missing was a solution for the text in the original article page, which in this case was completely erased. Both the erasure of the article text and the fracturing of the discussion to multiple pages (mostly talk pages) seem to represent an attempt to avoid having to reach a final decision on the text of the page. This strategy was understandable, as the article page was supposed to be a boundary object that all sides of the conflict could utilize. As compromise seemed far, and any decision would exclude some participants (thus denying the page its status as boundary object), the strategy formed was one of diffusion, replacing a single boundary object, whose position was both highly important and highly unstable, with a host of boundary objects. While Sanger's actions had no immediate effects, it did have important consequences. Sanger outlined a sequence of action designed as a mediation and representation strategy for dealing with conflicts, and gave it a name: "lack of bias". That name “blackboxed” (Latour, 1999) this sequence of actions and gave them a theoretical and justifying context. It became one of the first policies to be suggested as a general rule in the project. When Sanger started the first actual policy page in Wikipedia, titled rules to consider (Wikipedia, 2001ad), on February 4th, it one of the first two rules he posted on it was "avoid bias": "Since this is an encyclopedia, after a fashion, it would be best if you represented your mere opinions either (1) not at all, (2) on *Debate, *Talk, or *Discussion pages linked from the bottom of the page that you're tempted to grace, or (3) represented in a fact-stating fashion, i.e., which attributes a particular opinion to a particular person or group, rather than asserting the opinion as fact." Incorporating the lessons of this controversy, Sanger forced the practical solution to undergo a process of form investment, aligning itself with the model of the encyclopedia, though specifics of that model itself were almost never invoked. In a sense, Sanger reintroduced in Wikipedia Nupedia's "lack of bias" policy (see previous chapter). This act of translation subtly changed the meaning of the original policy: while the Nupedia lack of bias policy was concerned with the correct representation 127 of disagreements between experts137, its corollary in Wikipedia policy was concerned with the problem of editors expressing their "mere opinions" and resolution of disagreements between editors. As former examples demonstrated, the creation of conscription devices (Henderson, 1991) or PPP's demands careful coordination between various processes. When the processes involved – the generalization of the solution to a particular problem, its effective translation through form investment (Thévenot, 1984) in prevalent modes of coordination, and its technological or social inscription – occur simultaneously and support each other (as was the case in the refactoring debate), a general policy may emerge and achieve the status of a privileged passage point. In this particular case, however, Sanger's attempt failed, as he did not succeed in achieving an effective consensus. The inclusion of "lack of bias" in the rules to consider page did not have the desired effect, as other controversies and accusations of bias soon flared up on other articles. There are several possible explanations for this failure. The first is Sanger's failure to use Wikipedia's conscription devices effectively. As the policy was located inside a list of "rules to consider" and written as a suggestion ("it would be best") rather than a directive, it was a problematic resource to rely on for socialization or in cases of conflict. Moreover, the form investment relating to the encyclopedic model was done offhandedly, sufficing with the somewhat demeaning assertion that "this is an encyclopedia, after a fashion." Finally, the policy was not really translated from the practical solution developed to answer the local problem in the History of United States, but rather from Nupedia's policies, thus failing to address such issues as the creation of talk pages as a response to bias or controversy, the forking of article pages, what to write in the article page itself in cases of controversy, etc. Thus, the production of an effective policy had to wait for other conflicts to present examples of successful resolutions. The ineffectiveness of the lack of policy bias was demonstrated merely four days after its suggestion, when a major controversy similar to the previous one and 137 As the Nupedia guidelines state "This question is a good (albeit not infallible) test of a lack of bias: 'On every issue about which there might be even minor dispute among experts on this subject, is it very difficult or impossible for the reader to determine what the view is to which the author adheres?' 128 involving many of the same participants erupted on the article capitalism (Wikipedia, 2001af). On February 2nd, Tim Shell posted a brief article on capitalism, defining it in terms embedded in libertarian ideology as "A system based on the absence of political intervention in the economy, which rests on the assumption that politics and economics are distinct from one another. Often capitalism is part of a wider, Libertarian social theory that advocates the absence of political intervention from most aspects of society." This definition, as well as the list of relevant theorists that included only liberal and neo-liberal ones,138 caused unrest amongst some participants such as Joshua Gross and the user nicknamed Astronomer, who tried to improve it through editing rather than discussion, by expanding it and hedging the more controversial claims. However, the bulk of the article remained the same. Conflict erupted on February 8th, when WojPob posted in the bottom of the page the page a polemic response stating, amongst other objections, that "This is a very biased, right-wing article, written in a style that reminds me of Goebbels' propaganda." Less than two minutes later, Astronomer started a talk page and moved WojPob's comment there, and Tim Shell and WojPob started arguing over the content of the article in it. A few hours later, Sanger stepped in and suggested – following his preference of editing over discussion – that WojPob should just change the article "so that it's more accurate by your lights". However, before WojPob had a chance to do so, Jimmy Wales did. In an attempt to solve the controversy, he added the names of anti-capitalist theorists to the article, and quickly added to the talk page a note "In an attempt to make the article appear to be less one-sided, I added some anti-capitalist theorists to the list of theorists. I think that the article is mostly good as it stands right now because it doesn't advocate for or against capitalism, but merely defines the term in a way that a broad spectrum of people can agree is accurate. I say broad spectrum because I don't think that everyone will agree. 138 The list included Adam Smith, David Ricardo, Ludwig von Mises, Freidrich Hayek and Milton Friedman. 129 Perhaps a revised version might acknowledge some other definitions of the term?" Wales' move offered a new solution to the problem of disagreement: instead of the fighting parties editing each other's work, or merely debating it on a discussion page, he suggested a reformulation of the article content by a third party. This defined the general policy Wales would advocate for with regards to controversial articles: restating them in a way all parties participating in the production conflict can live with. In this particular case, it proved successful. Wales' authority was strong enough as to prevent further discussion, and editing on the article wasn't resumed until mid-April. However, the results of this conflict were confined to the article and talk page, and only those aware of them knew of it. It was only later, when this sequence of actions was repacked, justified and presented in a central location on the network through a conscription device that a larger effective consensus could be achieved and it could become a general policy. The lesson of this episode, from a pragmatist point of view, is the importance of the visibility of practices: a solution to a problem will only surpass the local context of its creation if it is visible enough to be considered relevant for other contexts. Meanwhile, the building blocks of this policy were still being gathered. To the discussion on a talk page and rewriting of the original article in a way both sides could live with another, an additional feature was soon to be added – the forking of article pages. 5.3.2 Football and the Forking of Article Pages A third debate that helped shape Wikipedia's policy took place around the seemingly innocent football article (Wikipedia, 2001ae). While its contents were uncontroversial, its name posed a problem: on January 24th, Linus Tolke, a European participant, posted on the article a short description of Association Football, commonly known in the U.S.A. as soccer139, in a conscious attempt to "stir something up" as he put it – i.e., to test how Wikipedia deals with issues arising from its global and multicultural nature. This was a clear challenge against the united-states-centered 139 "Association football" is a somewhat artificial term for the known simply as football in most parts of the world. In U.S.A. and a few other countries, such as Canada and Australia, it is known as soccer. 130 outlook that was already emerging in Wikipedia, due to the fact that its founders and many of its participants were Americans. This move triggered a debate, the first of its kind, on the question of how to practically confront such dilemmas. The general consensus was that the article should be split in two, but the question remained how these articles should be named. The European participants rejected the idea of naming the European sport soccer, as it was an American name rarely used in Europe. The debate, which involved many of Wikipedia's contributors at the time, was calm, civil and at times humorous, as the issue was one that most participants had no strong feeling about. However, in contrast with the heated but very short conflicts presented above, it moved along very slowly, without resolution in sight for more than two weeks. The final resolution was reached almost accidently, through uncoordinated action. On February 8th, WojPob moved the debate from the article page to a talk page he opened (following the newly created policy to place discussions in such pages), leaving in it only the original definition referring to Football in the European sense – as it was the only part of the text that wasn't a debate. Within the hour, Astronomer extensively rewrote the article so it will refer to both games, opening it with the statement: "Two teams play against each other to score with a ball in the opponent's area. The agreements end here." Below it, he added a headline "If you are outside the United States, football means […]", under which he described the general concept of Association football. After that, he drew a break line, and under the headline "However, if you are in the United States, football is something different: […]" went on to describe American football. A couple of hours later, immediately after his edit solved the conflict on the capitalism debate, Wales used his authority to conclude this one as well, as he added to the beginning of the article the sentence "Here at Wikipedia, in the interest of peace among all the peoples of the Earth, we shall call this sport American Football". He then went on to start the American Football (Wikipedia, 2001ag) article and started filling it with content. As none of the other participants opposed this move, and some of them began to productively edit the related articles, the suggested compromise became an effective consensus. This episode set an important precedent, as participants jointly succeeded in turning a conflict into consensus, without the aid of a mediator. In addition, it resulted 131 in a pattern that could sometimes – though not always - be used to solve conflicts, namely the "forking" (splitting) of an article to allow for the representation of various meanings or outlooks. However this pattern was not explicitly acknowledged as a precedent, as it was a rather unique case that seemingly had little bearing on many of the ensuing conflicts in the project. Moreover, there was no explicit principle behind the decision, but rather ad-hoc actions by several editors, and a final arbitrary gesture by Wales towards un-American editors. Still, by this time all the building blocks that will be used to formulate the neutral point of view policy were in place. The only thing missing was a trigger that would force the translation of the practical knowledge accumulated into a clear and effective policy, disseminated throughout the network from a privileged passage point. This trigger arrived in the figure of Richard Kulisz, the first real opposition to Wikipedia's founders. 5.3.3 Anarchy, Conflict and Policy Formation On February 13th, Wales began to actively partake in a conflict sparked in the article anarchy (Wikipedia, 2001y) between Tim Shell and Richard Kulisz, a self proclaimed anarchist that frequented Ward Cunningham's Portland Pattern Repository Wiki. On February 2nd, Shell wrote the initial content of the anarchy article, again advocating his libertarian views and claiming that all other types of anarchism oppose any kind of social order. Kulisz, a professed anarchist, opposed what he saw as a libertarian cooptation of the term by Shell, as it was traditionally related to the radical left anarchism to which he affiliated himself. Kulisz' claimed on the talk page that Shell was reversing his attempts and filling the article with "propaganda". Subsequently, Sanger and Wales both intervened and attempted to achieve a consensus between participants, but to no avail. Finally, on February 13th, following the pattern he devised on the Capitalism article, Wales deleted the article and posted on it a message stating "I am about to attempt to find a happy middle ground here. Give me a little while to write..." less than half an hour later he posted a new article, noting on the edit summary rubric that "We found peace about the term 'football', and so with 'anarchy'". 132 In his version, he emulated and enhanced Astronomer's tactic in his editing of the football article and his own attempt at creating a consensus in the capitalism article. The new article was comprised of four parts: a general preface stating that "Anarchy is a term used in two different ways, to describe two very different political views […]"; a short description of the "traditional" use of the term summarizing Kulisz' position, with a link to a new article called Traditional Anarchism (Wikipedia, 2001z); a short description of Shell's position with a link to a new article called Anarchocapitalism (Wikipedia, 2001ah); and a final section describing Shell's and Kulisz's criticism of the opposing view. Minutes later, Sanger hailed that attempt on the article's talk page (Wikipedia, 2001ai) and attempted to translate it into a general policy, stating that: It seems to me Jimbo has shown a way around this dispute, although Richard and Tim may disagree. At least Jimbo has illustrated the general direction that we should move in when we have these sorts of disputes. The aim is to describe the theories in such a way as cannot be objected to by their proponents or their opponents--by attributing the very descriptions of the theories to their proponents and then explaining what the differences and conflicts between competing theories are (rather than ignoring the fact that there is a conflict, or engaging the conflicts on the page itself). It seems to me this is something that anyone here can do with a little practice. Might be good for you, too. :-) – Larry Sanger As this description stayed in the relatively marginal part of the network, it had no larger effects. Moreover, it soon proved false, as Wales did not "describe the theories in such a way as cannot be objected to by their proponents or their opponents". In contrast with the capitalism article discussion, Wales' move did not solve the problem, as Kulisz insisted that his views are still misrepresented. Over the following days, Kulisz, Sanger and Wales conversed and argued on various pages, attempting to reach some sort of consensus on the content of the various anarchismrelated articles. These attempts, which were not very successful, exploded on February 16th when, in the course of editing the traditional anarchism article, Sanger added to the article an open challenge, in the form of the sentence "A simple and common objection to anarchism is that any notion of communal control implies something akin to a State. Anarchists reply..." He copied this sentence to the talk page, and explained that he was "…hoping someone would supply the reply." This was, of course, an 133 implementation of his "lack of bias" policy, attempting to allow Kulisz to present his views in a way that would be beneficial to the article. The result, however, was different from what he hoped for. Kulisz, the implicit addressee, posted his response on the article page, stating that "Anarchists reply that the similarity between the power hierarchies of corporations and the layered division of labor in cooperatives is superficial. […] If you compare both organizations using the items on the Psychopathy Checklist as a guide then it becomes clear that cooperatives act like normal humans, humane, warm, feeling and stable while the corporation can only be described as a psychopathic entity that cold-bloodedly uses workers for its own ends and acts on one impulse after another. As an example, one of the items on the PCL-R is "criminal versatility". Corporations have it, Microsoft demonstrates it when it buys companies expert in areas MS knows nothing about. While the attribution of this position to "anarchists" was supposedly enough to guard the text from accusations of bias, that was not the case. For Wales, the portrayal of corporations as psychopathic entities engaging in acts of "criminal versatility" was the last straw. He promptly deleted the part of the answer equating between corporations and psychopaths, and posted on the traditional anarchism/Talk page (Wikipedia, 2001aj) a message stating that "[…] I've gone to the mat trying to help remove what Richard perceived as bias from Tim's description of anarchy, and I'm repaid with propaganda? I figure I'm too annoyed right now to make all the changes that this article now needs." At this point, the argument rose above the concrete question of how to represent anarchism and moved to the more general level of content policy. In response to Wales' post, Kulisz stated that "This convenient definition of propaganda as 'an argument or viewpoint I strongly disagree with that may be widely convincing to others' destroys the value of the Wikipedia. Is the purpose of an encyclopedia to establish facts or is it to promulgate some dominant point of view?" To Kulisz' distinction between "facts" and "dominant point of view" Wales answered with a double distinction, conflating representation and morality. Addressing the issue of representation, he asserted that "We can present facts _about_ bizarre points of view such as yours, but we must not present your bizarre point of view _as fact_. I'm sure you can see the difference'' (Underscore in the original). He then translated this distinction to moral terms, noting that "I helped you make changes to things that you 134 disagreed with, in an effort to achieve consensus. I thought, and argued on your behalf, that you only wanted to present certain ideas fairly and in a neutral manner. But I now believe that you only want to put forward your own political agenda.'' In his answers to Kulisz, Wales drew a distinct boundary between good and bad representation which was intimately linked to the challenges of collaborative production: good representation both presents disputed opinions as such (rather than stating them as facts), fairly and in a neutral manner, and is part of an effort to achieve consensus between participants. Good representation and cooperation are thus constructed and entangled as elated aspects of a single epistemic virtue (Daston and Galison, 2007), whereas epistemic vice is constructed as an insistence on promoting one's own agenda over those of others and resisting effective compromises. This move set the ground for the actual formation of the new policy. On February 16th 21:27, less than quarter of an hour after his argument with Kulisz, Wales started a new page titled neutral point of view (Wikipedia, 2001ak) where he presented a general policy for preventing and dealing with disputes. To blackbox it, he used the formulations presented in the dispute with Kulisz. He incorporated Kulisz’s formulation that Wikipedia should not promote a "Dominant point of view", his statement that he expected Kulisz to write "fairly and in a neutral manner" ad Sanger’s attempt to create a general policy out of his refactoring in the Anarchy/Talk page, and stated that: A general purpose encyclopedia is a collection of synthesized knowledge presented from a neutral point of view. To whatever extent possible, encyclopedic writing should steer clear of taking any particular stance '''other than''' the stance of the neutral point of view. The neutral point of view attempts to present ideas and facts in such a fashion that both supporters and opponents can agree. Of course, 100% agreement is not possible; there are ideologues in the world who will not concede to any presentation other than a forceful statement of their own point of view. We can only seek a type of writing that is agreeable to essentially rational people who may differ on particular points. Some examples may help to drive home the point I am trying to make. An encyclopedic article should not argue that corporations are criminals, even if the author believes it to be so. It should instead 135 present the fact that _some people_ believe it, and what their reasons are, and then as well it should present what the other side says. An encyclopedia article should not argue that laissez-faire capitalism is the best social system. (I happen to believe this, by the way.) It should instead present the arguments of the advocates of that point of view, and the arguments of the people who disagree with that point of view. Perhaps the easiest way to make your writing more encyclopedic, is to write about _what people believe_, rather than _what is so_. If this strikes you as somehow subjectivist or collectivist or imperialist, then ask me about it, because I think that you are just mistaken. What people believe is a matter of objective fact, and we can present _that_ quite easily from the neutral point of view. (Underscore in the original). Building on the cases of the debates described above, Sanger's formulations and his experience in Nupedia, Wales rewrote the ineffective "lack of bias" to make it a cornerstone for the project. In contrast with the "lack of bias" policy, it incorporated a concrete model of the encyclopedia and a clear justification in terms of knowledge representation as it portrayed the encyclopedia as "a collection of synthesized knowledge presented from a neutral point of view". It was worded in a clear, authoritative and direct manner, and in the first person. It was not a "rule to consider", with its hesitant undertone and negotiable standpoint, but rather a clear and commanding directive, leaving room for explanations but not for changes. It built on concrete examples of debates that Wikipedia's editors were familiar with, and gave clear directions as to how to follow this policy. In short, Wales built on his involvement in several attempts to solve concrete problems in the project and on his authority as the founder of the project in order to portray an ideal image of the encyclopedia that would justify the standards he promoted. Moreover, Wales incorporated the possibility of dissent into the policy itself, he anticipated "that there are ideologues in the world who will not concede to any presentation other than a forceful statement of their own point of view." By pointing to this option, and limiting the pretentions of the project to "essentially rational people" he effectively excluded from its imagined community anyone who could not be made to agree on a representation of his views. As anticipated by Wales, Kulisz was not convinced. He answered Wales' post on the traditional anarchism/Talk page, insisting that his portrayal of corporations as 136 psychopathic was a fact. Wales refused to discuss the issue further, claiming that "I'm not the least bit interested in debating with you the truth or falsehood of these claims. I am only interested in seeing that everything gets stated in a neutral, encyclopedic manner." This refusal kept focusing the discussion on a rather abstract level, slowly uncovering the rift in the epistemological positions between the sides: Kulisz presented an inherently political view of knowledge, and saw Wikipedia as a means to distance knowledge from the academic ethos and to radicalize it. Building on views of the sixties’ radical left-wing counter culture140, he claimed that "'neutrality' is just a cover for pushing the mainstream ideology," and that in terms of content it results in "bland apolitical tripe". As he stated in a reply to Sanger, "if you tell me that encyclopedias are inherently academic and middle-class artifacts then I am prepared to accept your judgment on the matter. You would be confirming my suspicions. In that case, I would limit my participation appropriately." Sanger promptly replied that this discussion was "more evidence that we should just write articles instead of engaging in this sort of pointless, sophomoric mental masturbation. If you don't want to do so, Richard, and you prefer to engage in the above sort of ranting on this wiki, I would suggest you stop wasting our time and 'limit your participation appropriately.' Indeed, Kulisz abruptly stopped editing in Wikipedia, and the discussion – as well as the editing conflicts on the various anarchism related articles – stopped. He was not missed by other participants, who did not support his position or request him to stay. To conclude, the result of this episode was the tying together of a number of prior efforts into one clear and generally accepted policy, and the casting of its major opponent out of the project and the alternative he sought to advance. The solution reached in this case was determined, to some extent, by Kulisz' position: his standpoint was so radical and so unrelated to the actual production practices performed by most Wikipedians that he found no support whatsoever. This reaction exemplifies the importance of external cultural factors, as pre-existing models of representation prevalent among participants in the project failed Kulisz' attempt at form investment (Thévenot, 1984). These models, characterized by a relative disassociation between knowledge production and political advocacy had an important role in selecting the specific solution achieved in this case. 140 For an exploration of the influence of the 60's counter-culture on the internet, see for example (Turner, 2006). 137 Kulisz' pattern of action also played an important part in determining the outcome: his combative and generally uncooperative attitude made him a problematic ally, obstructing him from enlisting other participants to his goals. What seems to me as the most important factor leading to Kulisz' defeat, was the fact that he only posted his responses in the anarchy related pages, thus leaving the relevant policy pages as statements that showed no opposition, and enabled them to be viewed as relatively unproblematic and consensual. When he quit the project he left virtually no signs of his opposition in central locations such as the rules to consider page, the neutral point of view page or the project's mailing list. Thus, the creation of knowledge-related policy involved both a change in the production process, and a change in the project's social structure – that now remained less pluralistic with regard to positions about knowledge production, but more cohesive and uniform. Kulisz's departure uncovers some implicit social results of policy formation: pragmatic problem solving is not only about enabling cooperation and coordination (as usually emphasized by members of the pragmatist school and social interactions such as Joas, Star and Thévenot), but also a mechanism of exclusion, as it creates boundaries between those participants who support it and those who don't. As conflicting attempts of translations tried and tested, the losing parties will often find no place in the transmuted project. Thus, translation changed not only patterns of social action, but also the identity of the actors themselves, as new actornetwork are shaped and mutated through this process. 5.3.4 "Larry's Big Reply" In the following months, the neutral point of view policy proved successful as editors of conflicting views proved capable of discussing their views and achieving a measure of consensus that allowed them to jointly produce articles. This calm period was interrupted when a new debate broke on the newly created creationism (Wikipedia, 2001al) article on April 4th involving Sanger, Lee Daniel Crocker, John M. Lynch, Aye Spy and Joshua Gross, regarding the question whether the article should state that the creationist theory is false. This time, the conflict was of a different nature, as it was not between opposing views; participants were unanimous 138 in their rejection of the theory. However, while Sanger and AyeSpy tried to avoid an explicit endorsement of evolutionary theory on the article, the other participants stated that it should be presented as a scientific fact, and that creationism should thus be presented as false. Soon the debate – continued on the creationism/Talk (Wikipedia, 2001am) page – sprawled to the issue of historical revisionism and holocaust denial, as supporters of the later position claimed that the result of a strict Neutral point of view policy would necessitate in presenting such views as legitimate. Finally, Sanger posted a lengthy reply he titled "Larry's Big Reply" (Wikipedia, 2001an). In it he elaborated his views on the subject, invoking the neutral point of view and the model and purpose of the encyclopedia in support of his view. In conclusion, he summarizes his position My conclusion: We should not impose our values on other thinking people. You are all liberal-minded people, I trust--not liberal politically, necessarily, but liberal in the sense that you want to free minds. I enjoin you to think carefully about the best way to achieve this. By failing to take stands on controversial issues, we ''aren't'' demonstrating weakness--in fact, we are demonstrating the strength of our faith in the minds of our fellow human beings. We should let them arrive at their own conclusions. We should trust them to use their own minds--just as ''you'' want to be trusted. More benighted souls than our enlightened selves will appreciate our stance and be more apt to listen when we hand down the truth. The epistemological stance advocated in this post, while embedded in Wales' neutral point of view policy, goes far beyond it as it translates it from the realm of political disputes to that of scientific knowledge. The position adopted by Sanger gave a wide definition of the "essentially rational people" – much wider than his opponents wished for – as he opposed the identification of rationality with science and withheld from scientists and experts their authority over knowledge. It suggested that as this authority is in itself controversial, adhering to it will only alienate potential readers. Thus, the role of the encyclopedia itself is changed: it is no longer a way to educate people with the truth, as found by the various branches of scholarship, but to give them information about the various views prevalent in society. The nature of truth itself is transformed here: it stops being a relationship with the world, and becomes a relationship with culture. The authority of science becomes an empirical question: is there, on a specific topic an uncontroversial scientific statement available to the 139 project's participants? If so, then it is presumed as the authoritative and legitimate statement on the subject. If not, statements may be presented as the dominant position, but not as true. As the other participants in the discussion began reacting to it, it turned out that this time Sanger's post succeeded magnificently in convincing them, and the debate was quickly and clearly settled. This episode widened the extant of the neutral point of view, now being used to determine how to act in cases of controversies involving scientific or public conflict, and not just in cases of political conflicts between editors. "Larry's big reply" also became a major explanation and rationale to support the policy, and at the same time further distilled the form of the encyclopedia that Wikipedia strove to achieve. This episode marked Sanger's prominence and privileged position in the realm of content policies formation. Even though the original policy in this case was written by Wales (while Sanger's original attempt failed), it was Sanger who sustained and promoted it in the following months. During this period, Sanger strongly advocated for the neutral point of view policy, referring and providing a link to it in various locations: in policy pages such as rules to consider, Wikipedia policy, and give the author a chance debate (Wikipedia, 2001df); introductory pages such as how to start a page (Wikipedia, 2001dg) and welcome, newcomers (Wikipedia, 2001dh); and many talk pages of articles he felt that failed to meet its standard, or where conflicts were taking place, such as Abortion/Talk (Wikipedia, 2001di), scientology/Talk (Wikipedia, 2001dj), racism/Talk (Wikipedia, 2001dk), John Stuart Mill/Talk (Wikipedia, 2001dl), Unamerican activities/Talk (Wikipedia, 2001dm). This pattern of action created a network of editors and pages that eventually supported and promoted the policy independently. Soon, the long and cumbersome "neutral point of view" was further blackboxed and replaced by many users with the acronym NPOV, and POV became the parlance for bias. The blackboxing helped solidify the policy as it changed the everyday language of the original formulation into a technical term that newcomers encountered as an integral component of the project. The institutionalization of the neutral point of view policy does not mean that it became an OPP in the project. It was a very strong PPP, but it was repeatedly and heatedly debated, as the project's reliance on effective consensus as means of creating policies meant that it could also, in principle, be changed or abolished. In addition, it was 140 often challenged by infringements, and had no universal enforcement mechanism to prevent the recurrence of such actions. 5.3.5 Crisis and Resolution – Wales' Statement of Principles In October, with the influx of unprecedented amounts of new editors, extensive debates began about the meaning, limits and weight of the policy, centering on the representation of scientific knowledge, and the treatment of controversial religious issues such as abortion and scientology. These debates were part of the larger controversies that traversed the community apart at the time, and a sign of the waning of Sanger's influence over the project, even though most participants in these discussions supported his positions. As part of the attempts to overcome these controversies, Jimmy Wales made an unprecedented move by making the policy "non-negotiable". The neutral point of view policy was distinguished from other content related policies in the project. It was extracted from the realm of consensus and became as close to an OPP as possible, made into a nodal point that cannot be removed. This was done in a landmark post by Jimmy Wales on his personal userpage on October 27th (Wales, 2001d). Addressing a plethora of controversies that shook the community at the time, Wales declared some principles to be final and absolute in Wikipedia, and himself as the final arbitrator regarding them, as he stated: "these are my principles, such that I am the final judge of them. This does not mean that I will not listen to you, but it does mean that at some ultimate, fundamental level, this is how Wikipedia will be run." The first of these principles read: "Wikipedia's success to date is entirely a function of our open community. This community will continue to live and breathe and grow only so long as those of us who participate in it continue to Do The Right Thing. Doing The Right Thing takes many forms, but perhaps most central is the preservation of our shared vision for the NPOV and for a culture of thoughtful, diplomatic honesty." Thus, the neutral point of view was declared as "a shared vision" exactly at the moment when it ceased to be one, and the open community aspect of Wikipedia, underlying the previous effective-consensus based policy formation process, was hailed at the moment it was undermined by the personalization of authority. This is 141 precisely the function of a nodal point, or OPP: it is a stopping point beyond which translations and changes of meaning have minimal or no effect, even if they continue. Discussions from this point on had to take into account that only the interpretation of the policy was now open for debate, while the policy itself became "non-negotiable". A second move, made by Sanger, was a reformulation of the policy. After long months of existence, the neutral point of view page consisted mainly of comments and discussion appearing below the original formulation. Following the increase in conflicts around the policy in October, Sanger began summarizing the main points in the debate, and began formalizing a draft for a new version of the policy. Editors, and especially those who had reservations with the policy, were asked to contribute and Sanger presented the various issues that were raised and attempted to explain them away. After discussions and a second draft, he posted the new formulation on Wikipedia in the beginning of January, leaving Wales' original formulation at the top of the page, and linking to a wealth of examples, debates and articles at the bottom. While this move did not solve all controversies regarding the policy, it did diminish them considerably. This owes at least in part to Sanger stamina in long and contentious debates, and to the fact that the policy page and adjacent pages were so long, that engaging with them became a daunting task. The neutral point of view policy page thus became the center of a vast network of conscription devices, including various policy, commentary and talk pages, links to them from various locations in the network, both from peripheral article pages and central policy pages, tutorial and FAQ pages. 5.4 Conclusion The shaping and institutionalization of the neutral point of view policy and the separation between article and talk pages clearly demonstrate the connection between social structure, creative practices, forms of knowledge production and technology. The construction of these policies governing knowledge representation was just as importantly the construction of policies for governing social interaction in cases of conflicts. Moreover, it shaped Wikipedia's, both in terms of its authority structure and in terms of the composition of the networks comprising it. Its complex history, its 142 failed precedents and alternatives and the plurality of types of external cultural resources drawn together to create it (including previous practical experience, proven technical solutions, general modes of coordination and forms established of representation) point to the highly creative work involved in its translations. Its subsequent expansion to cover issues such as cultural bias and controversies regarding science and expertise points to the inherently polysemic nature of policies: Never completely fixed, the continuous process of blackboxing that bundled new meanings and practices with older ones extended it and made it ever more central as new parts of the network were linked to it. At the same time, it made it an object of contention, a resource and prize in social conflicts between rivaling actor-networks. In both cases, the policies were the result of an effective consensus achieved through the enactment of power and authority. However, lacking the technological aid that the separation of article and talk pages enjoyed, the maintenance of effective consensus on the neutral point of view policy demanded much more elaborate translations and continued negotiations to sustain. At the same time, it became institutionalized enough to support a common ground for participants with highly divergent and often conflicting views, and develop into an assortment of mechanisms to enable their constructive cooperation. This fostered the creation of a "culture of critical discourse" (Gouldner, 1979), where justifications and arguments are openly discussed and judged based on their merits, while preventing social fragmentation of the project. As implied above, this culture was not an amorphous idea: it was carved in the project through the institutionalization of standards and practices that had crucial effects on Wikipedia's production, content and consumption. A notable example is the creation of talk pages, which allowed participants to engage in productive discussion and all readers to view these discussions. This open attitude, constructing every reader as a potential editor, had important implications on the project's reception, as talk pages gave its users effective tools to assess the value of the article. Likewise, the institutionalized practice of leaving comments on the article page pointing to biased, missing or un-encyclopedic content, both enabled editors to guide one another through the work necessary to improve the encyclopedia and helped even completely casual readers to assess the article. such comments, first articulated by editors 143 individually, slowly underwent a process of standardization until the creation of templates – prefabricated standard comments and warnings that can be conveniently pasted to the article through a link – made them an integral part of Wikipedia's technological infrastructure. Such mechanisms thus served as conscription devices (Henderson, 1991) which enabled Wikipedia's editors to enlist reader to their network, encouraging them to share the vision of the project as an unfinished one; one that requires constant care and a critical outlook in engaging it. The vision of Wikipedia articles as always-in-the-making created a common denominator between various kinds of contributors, and allowed it to continually attract gifts of knowledge. While the policies described in this chapter were involved in the creation of a gift economy which preferred and rewarded certain types of contributions, this vision left room for other types of giving – most importantly individual giving (as opposed to group-oriented, see ch.1) – as it was translated in practice to an absolute insistence on the immediacy of contribution. This insistence on immediacy was ever-present in Wikipedia, as it took the form of the note under each article which stated that "you can edit this page right now! It's a free, community project." It forced Wikipedia's founders and participants to relinquish the possibility of strict control over content (as was the case in Nupedia). The price they paid for that strategy was the inability to claim legitimacy for its content. The reward, much more important at the time, was the ability to cater both individual and group-oriented contributors at the same time – and the chance of transforming the former into latter. The decisions described in this chapter precluded the possibility of gaining legitimacy for Wikipedia based on its production method, in the manner attempted by Nupedia. In the next chapter, I describe how Wikipedia's founders and participants attempted to accord their project with some measure of legitimacy immersed in the form and content of its articles, rather than its production. While the policies presented in this chapter stabilized and institutionalized many aspects of Wikipedia's collaborative production process they hardly addressed the content of the encyclopedia itself. Due to its extremely liberal stance towards content in its early months, the general issue of which types of texts should be considered encyclopedic and thus to be included in Wikipedia was still to be determined. Just as the Wiki 144 process had to undergo local translation, so did the encyclopedic model. The next chapter will thus focus on the image of the encyclopedia as it developed in Wikipedia. 145 Chapter 6: Wikifying the Encyclopedic Form This chapter picks up the theme of the previous one: how the construction of content policies in Wikipedia operated as a site for articulation of the production process and the model of the encyclopedia. While the previous chapter focused on policies dealing specifically with collaborative production, this chapter describes the institutionalization of policies establishing standards for the inclusion or exclusion of content. The translation of the encyclopedic form to Wikipedia, the interpretation and selection of relevant attributes, was a contentious process, intimately related to struggle over power and authority in the project. It culminated in the construction of effective mechanisms for creating, modifying and enforcing content policies and the articulation of an effective and relevant encyclopedic model. This model, relying on attributes of the traditional encyclopedic article form but supplementing them with innovative elements, played a crucial part in enabling Wikipedia's growth and success. As noted earlier, there was no clear image of what encyclopedia Wikipedia is intended to produce in its first months, and almost no conventions and standards regarding desired content. A reader randomly surfing through Wikipedia in that period, would scarcely identify it as an encyclopedia141. Its hybrid mixture of content forms was sustained by a permissive attitude that stemmed from Wikipedia's founders' wish to accumulate large quantities of content as quickly as possible. Sanger and Wales' frustration about not being able to promote Nupedia (see above, chapter 3) and make it viable due to its lack of useful content drove them to construct Wikipedia content policies as liberally as possible, preferring wide participation over legitimacy 141 It contained, for example, a large number of lecture notes from a philosophy course Larry Sanger taught (Wikipedia, 2001b); Tim Shell's extensive description and analysis of Ayn Rand's Atlas Shrugged (Wikipedia, 2001c) - a total of 346 articles comprising around 15% of all Wikipedia pages in its first months; copies of various states' constitutions, parts of the bible. Personal commentary pages on various subjects were also widespread, as well as policy pages, technical pages, and pages dedicated to discussing them. In addition, following a prevalent Wiki custom, participants could – and many did – start a personal page on the Wiki. These pages and all their subpages were considered for quite some time a personal area where users can post whatever content they choose to. Tim shell, for example used the subpages of his user page to keep a personal diary for a while, as well as records of places he's been (Wikipedia, 2001ao). Sanger regularly posted columns and general thoughts on its subpages (Wikipedia, 2001ap). 146 and quantity over quality. While this lack of selectiveness fulfilled its founders' aims, it soon became obvious that if Wikipedia doesn’t have some measure of selectiveness regarding content, it would run the risk of atrophying, or becoming just another venue for endless discussions and unsorted content (see above, Chapter 5). This problem was intensified by the diversity of Wikipedia's participants: they lived all over the world, came from very different social and cultural backgrounds, usually did not know each other beforehand and held very different opinions regarding the project's content. However, they were generally united in the vision of creating an encyclopedia. This meant that in order for them to coordinate action effectively in their evolving global microstructure (Knorr-Cetina, 2005; Knorr-Cetina and Brugger, 2000) they had to decide what an encyclopedia is, in practical terms, for a project that was clearly distinct both in its technological infrastructure and its production method. Wikipedia's first year thus saw several attempts to formulate such policies, usually initiated when new networks of participant emerged in the project and promoted content that founders and veteran participants found problematic. The pattern of action that emerged to deal with such events had a relatively regular structure: in the first stage, an editing practice performed was identified as problematic by one of the veteran users, and an alternative was sought. In the second stage, the legitimacy of the content was determined in conscription devices such as policy pages or mailing lists. This stage usually involved the project's founders as prominent articulators, and the model of the encyclopedia constructed that serves as a legitimate standard. It also involved blackboxing: using a name, or a small semantic array, that substituted the description and explanation of policies. In the third stage, this bundle – the policy – was disseminated and enforced using conscription devices (Henderson, 1991), mostly in the form of the core network of policy pages. The attempts of newcomers to reconstruct the form of Wikipedia's articles were routinely countered by a dominant ideology promoted by the project's founders (especially by Sanger, who took the leading role in policy formation at the time) stating that "Wikipedia is an encyclopedia." This formulation represents Sanger's rejection, and efforts to discourage, competing ideologies in Wikipedia. Such ideologies, promoted by various users including LDC (Lee Daniel Crocker), LA2 147 (Lars Aronsson) and the Cunctator (who did not disclose his real name), emphasized the innovative aspects of the project – being digital, hypertextual and collaborative – and the need to modify the encyclopedia's model to utilize them. The result of these conflicts was usually a compromise, though one that tended to support the traditional model of the encyclopedia over more innovative and permissive options. However, it was permissive enough to prevent dissenters from leaving the project. Practically, Wikipedia's founders held that content should generally be complacent with the traditional form of the encyclopedic article as the articulator constructed it in relevant terms of its style and structure. They held that the concept of an encyclopedia was clear and unproblematic: As Jimmy Wales stated, "One reason for the success of the project so far is that it is instantly accessible. We have a concept which people grasp intuitively: 'encyclopedia'. And we have software that is usable with virtually no learning of any kind" (Wales, 2001j). Wikipedia's history undoubtedly supports this statement to some extent, but not entirely, as the attempt to translate the "intuitive" concept of encyclopedia to concrete policies often involved disagreements and conflicts. The outcome of these conflicts was, at the time, far from self evident and their settling demanded an exertion of symbolic power and the enlistment of allies. Symbolic power was usually manifested in the ability to create and manipulate conscription devices (Henderson, 1991) and other types of mediators – texts, links, software code etc. – to effectively coordinate participants to perform similar chains of actions. Through such mediators, problematic content went through a process of form investment (Thévenot, 2002a, 2002b): creating equivalences across time and space between various practices, objects, persons, in order to stabilize and anchor the participants' interpretation of this allegedly intuitive concept. This process progressed whenever conflicts regarding eligible content emerged, and was usually successfully carried out, mainly by the project's founders. Thus, the construction of content policy in reaction to concrete practices and problems was simultaneously a construction of an encyclopedic form. This form had peculiar properties: it was articulated through a series of binary oppositions between Wikipedia and other forms of content, mainly ones prevalent in its online environment. It did not rise to the level of a positive definition, but rather placed Wikipedia in the center of a locally articulated system of 148 differences (Laclau and Mouffe, 1985, p. 132) that allowed it to include unorthodox elements while retaining the fundamental justifying principle that "Wikipedia is an encyclopedia." In what follows, I focus on the construction of two such policies; Wikipedia is not a Dictionary and What Wikipedia is not. To a large extent, they still function as the backbone of Wikipedia's content selection policy. 6.1 Phase 1: Wiki is not Paper The first attempt to programmatically rearticulate the form of the encyclopedia to accommodate Wikipedia's innovative production model was made by Lee Daniel Crocker on March 19th, 2001. After the policy formation initiative in February that resulted in the policies discussed in the previous chapter, the Wikipedia community stabilized as a set standard practices became an effective consensus within a core of regular participants that could receive and socialize newcomers. The attempt to translate the encyclopedic model to this new environment was initiated as a fundamental discussion of Wikipedia's unique properties as a medium, and established the notion that Wikipedia should remain first and foremost an encyclopedia in terms of article format, but could allow itself to be very permissive in terms of legitimate article topics. The event the initiated this discussion was the creation of a new page named Wiki is not an encyclopedia (Wikipedia, 2001aq) Lee Daniel Crocker, one of the participants most involved in policy discussions during Wikipedia's early years142. In March 19, he started this page as a place to discuss "[…] how Wiki differs from older media, and what that implies about how we should use or modify traditions from those other media." As its title suggests, it attempted to steer it away from the traditional encyclopedic model. Shell asserted that this model had to be reinterpreted and adopted to fit the model of a hypertext Wiki. 142 Crocker, who joined the project in February (after being previously involved in Nupedia), was a computer programmer and free software enthusiast, involved in several significant software and internet projects and an avowed techno-utopist. 149 The content of this page was comprised of several sections, each tackling a different characteristic of the Wiki and illustrating its hypothetical consequences on the form of the encyclopedia constructed in the project. Among these was a section titled no size limit, in which Crocker suggested in contrast with traditional paper encyclopedias' parsimony, Wikipedia could include an unlimited number of pages on various topics143, as "hard disks are cheap." In another section, titled opinions, he suggested that while articles should be "as factual as possible", there should be a subpage titled /opinions for each articles, where participants could voice their opinions on the subject144. The first of these suggestions was already discussed in Nupedia, who cited the lack of size limits as a feature that could make it the "broadest, deepest, and most up-to-date encyclopedia in history." However, Crocker's version was far more radical, as it suggested writing extensively on subjects whose very place in the encyclopedia was doubtful, thus expanding its traditional boundaries by blurring the distinction between high and low culture. However, it was the second suggestion that generated controversy, as it reopened some of the issues that seemed closed by the institutionalization of the neutral point of view policy. An hour after the posting of the page, Wales edited it, adding on the top of the page the remark "The title of this page is entirely wrong. Wikipedia is an encyclopedia. If it were to become yet another discussion forum, I would just shut it down. The vision of yet another discussion forum doesn't motivate me in the least." This was not a very unusual statement in terms of content: Wales simply reiterated a primary concern of the projects founders, described in the previous chapter. In terms of asserting his authority, however, this statement was unprecedented, as Wales positioned himself for the first time as a final arbitrator regarding the project's future. On this subject, at least, he posited himself as a veritable OPP145, as his statement sharply outlined the boundaries of permissible practices and discourse in Wikipedia. 143 Crocker cited types of poker games and the Simpson's TV series characters as examples. Two other section dealt with style differences, suggesting to replace explanation of abbreviations and jargon with links to articles about them; and ease of editing, suggesting that "There should be less need of weasel words like 'at the time of this writing', 'generally recognized as', 'commonly believed that', etc. Just say it the way you think it is, and if you're proven wrong, come back and edit it later." 145 Wales' prerogative in this case – similar to the one he would later fill when he would declare that the Neutral Point of View is non-negotiable – was as close as he got to being and OPP. However, as later discussions indicated, the time honored "right to fork" in FOSS projects (dictating that participants 144 150 Wales then went on to post a response to each of the page's sections. In contrast with the statement above, his specific comments were generally positive in tone and in agreements with Croker's suggestions, except on the subject of opinions, where he stated his "strong disagreement," explaining again that "Wikipedia IS an encyclopedia. The Wikipedia should write neutrally about opinions, but the Wikipedia should not put forward opinions."146 Responding to Wales' comments, Crocker promptly moved the page's content to a new page titled Wiki is not paper Wikipedia, 2001ar). In it, he stated that he agrees with Wales that Wikipedia is an encyclopedia, and that he shares "his desire that it not become yet another discussion forum." This move placed the page on the legitimate side of Wikipedia's discursive boundaries, and allowed the discussion to continue, as veteran participants such as Larry Sanger and Tim Shell soon posted responses and suggestions to it. Crocker also added new sections, dealing with the consequences of timeliness and collaborative authorship, and urged other participants to add their own sections or comments. In the discussion which followed, two important suggestions emerged that would have significant effects on Wikipedia's image. The first of those, discussed above, is the highly permissive threshold for inclusion in Wikipedia, especially with regards to popular culture. A second and related suggestion was to focus efforts on articles related to current events, or news stories, in an attempt to generate traffic to the website. These decisions broke the traditions of both print and computer encyclopedias, shaping Wikipedia as it is known today – permissive, writing serious articles about the most minute and esoteric subjects, and so up to date that it often function as a news source. In contrast with this permissive attitude regarding subject selection, the desired form of specific articles remained subordinated to a traditional form. This compromise illustrates the co-construction of Wikipedia and the model of the encyclopedia it generated. Wishing to construct their project to as a new kind of encyclopedia adapted to its innovative medium and technology, Wikipedia's founders should be allowed to copy all existing materials to a new project if the owner of the old one falls from favor) prevented him that position, as he could not totally control the project's future. 146 The use of the term "the Wikipedia" in Wales' comment is reminiscent of the fact that it was quite some time before a consensual use of the term was achieved. 151 and participants had to select which elements of the encyclopedic model are essential, and which can be discarded. This decision was not self-evident, as Nupedia, for example, adhered to the form of the print encyclopedia both in terms of style and of scope. However, as it forwent Nupedia's emphasis on legitimacy and respectability, Wikipedia could suffice with a more skeletal image of the encyclopedia, stretching its form to include such low-culture or ephemeral subjects as popular cultured and current events. This episode marked the boundaries limiting practices and justification in the project, and demonstrated that a large degree of innovation can exist within them. Wiki is not paper ultimately failed to become a significant site for the articulation of Wikipedia's form of the encyclopedia. While it played a significant role in the initial construction of Wikipedia's form, and was temporarily incorporated to Wikipedia's set of conscription devices (Henderson, 1991)147, it was not declared or constructed as a policy page, and it never achieved that status. In the following months, it was moved to the commentary section and later to the meta-wiki (MetaWiki, 2001c) constructed to house the commentary pages when they were moved out of the Wikipedia Wiki (see below). In contrast with actual policy pages, which stayed on the main Wiki, it remained outside the core network of policy pages, the assortment of privileged passage points that was used to standardize methods and translate the various visions of Wikipedia's participants into an effective consensus. The actual articulation of Wikipedia's encyclopedic form took place primarily in other pages and in the context of conflicts over actual practices rather than over hypothetical possibilities, as was the case here. 6.2 Phase 2: Wikipedia is not a Dictionary The relatively stable state of Wikipedia's community lasted from February until May, when it was disrupted by the joining in of new participants who promoted a new approach to the project. This approach, emphasizing breadth and connectivity between articles at the expanse of depth, was met with firm opposition by Larry Sanger, who took it upon himself to establish and enforce production policies in the 147 For example, it was temporarily the first article linked to from the What Wikipedia is not policy page, even though it was not a policy page itself (cf. Wikipedia 2003). 152 project. In order to formulate these policies, Sanger attempted a process of form investment (Thévenot, 1984) as he aggregated practices, justifications and articulations from earlier discussions to create a general policy which was summarized under the title Wikipedia is not a dictionary (Wikipedia, 2012as). In contrast with previous efforts, this one was met with considerable opposition and had a relatively lengthy process of institutionalization, as achieving an effective consensus took considerable time and effort. The conflict, articulated mainly in terms derived of the traditional encyclopedic form, started a second major policy initiative. While the former one, in February, produced standard methods for collaborative production, this one produced standard criteria defining what content is considered legitimate in Wikipedia. 6.2.1 Encyclopedias vs. Dictionaries Many of the articles written during Wikipedia's first months were very short, containing only one or a few sentences, usually comprised of a general description of the subject. Those so-called "stub" articles were usually considered as a legitimate way to increase the breadth of the encyclopedia, even at the expanse of its depth. The reasoning behind it was that such articles would eventually be expanded to become full length articles. It was only in April 2001, about a month and a half after the creation of the NPOV policy that concerned were raised that such articles may bring diminish Wikipedia's legitimacy. It was then that the first attempt was made to distinguish legitimate encyclopedic articles from illegitimate ones. The initial trigger to discussing the distinction between encyclopedia and dictionary was a question raised in Public domain resources (Wikipedia, 2001at), a page dedicated to listing usable internet resources that could help editors create articles. On march 28th, an unidentified editor posted on it a suggestion to use the public domain 1913 Webster's Unabridged Dictionary (which was available online) as a "basic definition reference." About an hour later, Sanger deleted the suggestion, noting in the edit summary that "1913 is too old." In response, Wales started a Public Domain Resources/Talk (Wikipedia, 2001au) page, and objected to Sanger's position, claiming that "using the public domain 1913 Webster's Unabridged as a basic 153 reference for definitions of words can be very helpful. Of course, it is important for us to remember that Wikipedia is an encyclopedia, not a dictionary, so the random copying of entries from that work can be discouraged. But using those entries as a jumping off point seems like a great idea." In response, Sanger yielded his position, and stated that "I have no objection whatsoever to putting a link to a 1913 public domain dictionary on the Public domain resources page. My reservations are about what people will most likely ''do'' with it […] As you said, we aren't really after dictionary definitions here; encyclopedias are different from dictionaries. We can ''use'' text ''from'' a public domain dictionary, sure. In most cases, though, I don't think it would make much sense simply to ''cut and paste'' a definition from an dictionary into Wikipedia." Consequently, Wales reposted a link to the dictionary on the Public Domain Resources page, followed by a warning reading "please use freely but sparingly, keeping in mind that Wikipedia is an encyclopedia, not a dictionary." A few weeks later, on April 17th, an unknown editor started a very short article titled drunk (Wikipedia, 2012av), reading: "Drunk in its most common usage, is being intoxicated with alcohol," and noting various slang terms for it. Upon reading it, Sanger started Drunk/Talk (Wikipedia, 2001aw) page, where he stated, building on Wales' formulation in the former discussion that: Wikipedia is an encyclopedia, not a dictionary. The only reason, as far as I can tell, for enumerating various senses that a word can have is to direct people to appropriate articles; I don't see how Wikipedia has an interest qua encyclopedia in enumerating senses of a word in the way that a dictionary does. I'm certainly open to debate on this point; perhaps I'm not seeing things quite right. This invitation invoked three different responses, each defending "dictionary articles" from a different point of view. The first, posted by Tim Shell Less than an hour later, was articulated in terms of cost/benefit, wondering "What is the cost of a dictionary-like entry? Is Wikipedia worse off for having them?" the second, posted by the author of the original drunk page, advocated a process oriented approach, seeing such articles as a stage in the article building process. It stated that "For wiki to be most useful, a dictionary like stub article giving a short definition of a topic should be at the top of longer articles. 154 New topics may initially manifest as dictionary sized entries and eventually grow into full articles." The third and most radical response was posted by Lee Daniel Crocker. In accordance with his far-reaching stance towards innovation in Wikipedia, Crocker attacked the relevance of the very distinction between encyclopedia and dictionary. Crocker claimed that The distinction between "dictionary" and "encyclopedia" is imposed by the limitations of paper. There is no reason we should obey such an arbitrary distinction when it is not useful. Wikipedia is "A place to look up things you may have read but don't understand." Certainly we want descriptive text in much greater detail than a dictionary because we don't suffer the size limitations, but there's no reason at all not to have the dictionary-like information as well, and in fact I think we should have more of it, and a standard way for presenting it. An hour later, Sanger moved the whole conversation to a new page he titled Encyclopedia and or versus dictionary (Wikipedia, 2012ax), where he posted his answer, dealing with all three point of view. In reaction to Tim Shell's rhetorical question, he claimed that "the cost of having many dictionary-like entries--which list several senses of a word or phrase and give no further information than the meaning of the word or phrase--is to give the impression that Wikipedia has no ambitions to being any more than a dictionary." In reaction to the original author's claim that such articles are a natural result of Wikipedia's article production process, Sanger agreed, claiming that" […] there's nothing wrong with 'stub' articles per se, and in fact I think it's great to have them. I just wish that when you make them, they wouldn't consist just of lists of definitions of senses of the title word or phrase." Finally, in reaction to Lee Daniel Crocker's argument, he insisted that there is an essential difference between dictionaries and encyclopedias: The purpose of a dictionary is to give the meanings of words and short, common phrases; the purpose of an encyclopedia is to impart knowledge. Dictionaries help us to understand language; encyclopedias go far beyond that […] Why doesn't someone make http://www.wiktionary.com to make a wiki-based dictionary? That would be interesting, and it could be useful, too. […] encyclopedias provide more information than is typically needed when one consults a dictionary. It would be silly to come to Wikipedia if all you wanted to know is the meaning of the word-- 155 and usually, when we (as some of us often do) consult a dictionary, that is all we want to know. […] So I think we should say that our sole purpose is to build an encyclopedia, and our habits should be consistent with this purpose. […] we on Wikipedia should focus not on perfecting our lexicography but on increasing and imparting our (a posteriori) knowledge. I do think that we "perfect our lexicography" at the overall expense of our central purpose, of building an encyclopedia. The discussion continued until April 19th, in both the Encyclopedia and or versus dictionary/Talk page and on the Wikipedia-l mailing list, using these same arguments or ones very similar to them, without real consequences or decisions. The move that set the tone for the discussion, and finally determined its outcome, was Sanger's suggestion to create a "Wikitionary" – a proposal he simultaneously posted on the page and to the mailing list (Sanger, 2001e). Whether a conscious attempt of deflecting dissent or not, it was highly effective in eroding the opposition to Sanger and Wales' stance. Sanger's suggestion got considerable support on the mailing list (for example, Wales, 2001h and Rybo, 2001), alongside loud objections (such as Gérard, 2001, and Parks, 2001) and a lively debate began as to its merits and technical implementation. While plans of creating it were not carried out at the time148, these discussions reinforced the distinction between encyclopedia and dictionary without having to actually articulate it. Soon, the discussion about Wikitionary replaced the original debate on the form of the Wikipedia, and it quickly died out. Wales' original statement that Wikipedia is not a dictionary, and the subsequent discussion, didn't have much effect, as hardly anyone made an effort to disseminate or enforce it - or for that matter, to breach it 149. The Encyclopedia and or versus dictionary page soon became inactive, and was not incorporated into the core of content policy pages. The importance of this conflict is mainly in its establishing the relevance of the distinction between encyclopedia and dictionary, and in blackboxing the relevant position under the catchy title "Wikipedia is not a dictionary." 148 It was implemented much later, in December 2002, long after Sanger left the project. The only such incident occurred on May 14th when ran into the article traitor, including only a short definition of the term, and started traitor/Talk page where he remarked that "Wikipedia isn't a dictionary" and asked that the article be broadened. However, this action did not produce any reactions or repercussions. 149 156 6.2.2 Conflicting Strategies of Encyclopedia production The occasion that brought about the invocation of "Wikipedia is not a dictionary" as a practical policy occurred in May, as part of a larger confrontation over Wikipedia's nature and goals. This confrontation happened after a small but influential wave of (mostly European) new users joined the project in late April and may, including, among others, KoyaanisQatsi (also known as KQ, joined April 21st), Paul Robinson (joined May 1st), LA2, Pinkunicorn and Wathiik (all joined May 10th), Hornlo (joined May 13th) and Sjc (joined May 24th). These new editors were extremely prolific and played an important role in expanding the project. However, some of them introduced new, unorthodox and controversial editing practices. Though they differed in the details of their editing practices, most of them put a prime on the breadth of Wikipedia and on interconnectivity between articles, often at the expanse of depth and seriousness. This was exemplified in a tendency to write short articles containing only a few sentences with very basic information and interlink them to other articles, and in the creation of pages whose sole purpose was to help navigation, by massing links connected to specific topics, dates in history, etc. The user that led these trends and stood at the focus of the conflict was LA2. A programmer and internet entrepreneur150, he was connected to a group of Swedish contributors who were involved in the University of Linköping's Lysator computer club, including also LinusTolke, Mjausson and Pinkunicorn. LA2 assumed a role Similar to those of the protagonists portrayed by Star and Griesemer (1989), Functioning as an OPP for a network of allies, and more importantly, for the editing practices prevalent in it. In this capacity, he attempted to translate these practices into standard methods (or at least acceptable methods) and redefine Wikipedia's characteristics as a boundary object in a way compatible with his aims. While the other members of the network performed similar practices, he was the most prominent and prolific, and the only one that articulated and actively advocated these practices using Wikipedia's conscription devices (Henderson, 1991). LA2 began editing pages in Wikipedia under that nickname on May 10th, and immediately proved to be extremely productive, contributing an unusual amount of 150 Even though he used a nickname, he didn't conceal his real name, Lars Aronsson. He was the founder of Project Runeberg the Scandinavian version of Project Gutenberg (http://runeberg.org/) 157 144 edits by the end of the day – more than the monthly yield of most veteran editors151. He wrote about a large variety of subjects he often knew little about, such as communication, mathematics, history, philosophy, comics, Sweden, Higher education institutes and German politics. His editing style was unique and easily recognizable: almost all his edits were very short, containing usually no more than a sentence or two with very basic information, and links to other articles, one of which he usually went on to edit next. Some of his articles gave only a short literal definition152; some were comprised almost completely of links to other articles153; many others contained mainly links to external websites154. An unusual percentage of his edits resulted in the creation of new articles155. This pattern was an intentional strategy, as LA2 later commented that: "My opinion was that different people could contribute a skeleton of new article headings and hypertext links, while others could contribute longer texts to each article." LA2's practices and his vision that emphasized the network aspect of Wikipedia over the content of its articles strongly contradicted Sanger's view. Sanger saw the heart of the project as the production of mature or "good enough" encyclopedia articles, generally devaluating all other kinds of practices, especially if he saw them as interfering with this goal. This emphasis may have been related to his vision of using Wikipedia's content in Nupedia, which he still saw as the more important project. Another cause for friction, at least as important, was LA2's disregard for established practices and seniority in the project, and the obvious lack of seriousness expressed in the humorous nature of some of his edits. The outcome of these contrasting attitudes was a lengthy conflict between LA2 and Sanger, conducted all over Wikipedia, aided by various scopic media (Knorr-Cetina, 2005) and conscription devices (Henderson, 1991). Its result was the exclusion of the former from the project and the formation of a new policy by the later. 151 In his first 106 edits he was not logged in, but the identical IP address in the archive, as well as the similar editing style, indicate that his first edit was made on 9:03 (Wikipedia, 2001ay). 152 For example, the Ship article read: "A big boat. See Sea Transport or transport" (Wikipedia, 2001az). 153 For example, Philosopher (Wikipedia, 2001ba) and Comic-Strip (Wikipedia, 2001bb) were comprised completely of such links. 154 For example, the WIPO article read: "The World Intellectual Property Organization. See intellectual property and http://www.wipo.org/" (Wikipedia, 2001bc). The University of Oxford article read: "The famous university in Oxford, England. *Official website, http://www.oxford.ac.uk/" (Wikipedia, 2001bd). 155 41% percents of his edits resulted in the creation of new articles, in contrast with less than 30% for other editors at the time. 158 On 12:19, May 16th Sanger edited Wikipedia after being away for a couple of hours156. As a routine practice, he probably scanned the "recent changes" page –a central scopic medium in Wikipedia, enabling editors to see all the edits made in the system. Going over it, he could not overlook the new Wikipedian who literally took over it. LA2 was responsible for almost all the edits made while Sanger was away157. Even more conspicuous was an edit he made to the Wikipedia article (Wikipedia, 2001be). For a new editor, this was highly unusual, as the Wikipedia page, describing the project, was rarely edited, and exclusively by the most veteran of editors – Sanger, Wales, Gross and Shell. Even more unusual was the edit itself – a complaint about Wikipedia's slowness, linking to a new page LA2 started on the topic. Sanger quickly moved the edit to the Wikipedia bugs (Wikipedia, 2001bf) page, the usual venue for such complaints, and responded to it in a practical manner, suggesting that maybe the issue should also be raised on Wikipedia-l. After this incident, Sanger naturally checked LA2's user page, and discovered the following paragraph: A brave fighter can do with a short sword, for he is not afraid to go close to his enemy. The way to acquire knowledge is by close and intense study of the subject matter. My purpose with contributing to Wikipedia is not to make a lasting and useful encyclopedia, because I think that requires a whole lot more organization than this, but I think that I can learn something from using and contributing to this collaboratively edited hypertext. (Wikipedia, 2001bg) In this statement, LA2 excluded himself from the Wikipedia community, doubting that it could achieve it manifested goal. His goals are different: to learn about collaborative editing of hypertext. Upon reading this, immediately after dealing with the Wikipedia page edit, Sanger posted a response on LA2's page. Perhaps in an attempt to diffuse the tension, he wrote: "As an encyclopedia editor, I am now of the opinion that a lasting and useful encyclopedia must first exist--and the best way to put it into existence if it's a wholly volunteer project is to get rid of the official organization wherever self-organization can do. Please see http://www.nupedia.com. :-)". 156 The following is a reconstruction based primarily on the edit logs of these dates, and to some extent on Sanger's memoir and Wikipedia-l mailing list archive. I took some literary freedoms in writing the narrative – mostly in places where I could not determine the exact chronological order of specific actions - but all assumptions are plausible, and none are critical to the unfolding of the events. 157 44 edits out of a total of 61. 159 A few minutes later, while going over LA2's recent edits, Sanger found another disturbing edit: LA2 started a new page called Open Source Initiative (Wikipedia, 2001bh). The text on the page read: "See the website http://www.osi.org/. (not that it is related, but quite funny)." The article was a joke, made in somewhat poor taste: While the open source initiative was an organization dedicated to promoting open source software, the link led to the homepage of a Canadian pig breeding company that shared the initiative's acronym, called "Ontario Swine Improvement." Sanger promptly edited removed LA2's content and replaced it with the text "The Open Source Initiative is... (sorry, I don't know...this is my way of asking what it is!)." In the edit summary, he wrote "Removed link. Wikipedia is neither a links collection nor a joke collection." Sanger's initial reaction towards LA2's edits seemed to work. A day after Sanger commented on LA2's userpage that "a useful encyclopedia must first exist", he answered that "existence is a necessary but perhaps not sufficient requirement. But the only way we can find out is to create it." Subsequently, LA2 began engaging the project somewhat more seriously – an attitude exemplified both in a sharp decrease in the number of humorous edits, and in posting policy suggestions on the rules to consider. While rules to consider, like all Wikipedia pages, was open to editing by anyone, it was the first time a non-veteran editor added a new rule to it – and in LA2's case, it was only a week after his joining the project. The main rule he added, titled "build the web", described what he considered good practices for linking between articles – a subject not addressed at the time. The rule read: "Articles in an encyclopedia are nodes in a hypertext system. Don't just write the article, but also consider its place in the link web. Make upward links to categories and contexts (Charles Darwin was a biologist, Sahara is a desert in Africa, enlightenment happened in the 18th century). Make sideways links to neighboring articles (for proton see also electron, Oregon borders on California). Don't build category trees too deep and narrow, or too flat. Writing category directories first (top-down) will help ensure that subcategory articles get useful names (church names are not good now)" (Wikipedia, 2001bi) LA2's practices and his vision that emphasized the network aspect of Wikipedia over the content of its articles contradicted Sanger's view, to an extent. Sanger saw the heart of the project as the production of mature or "good enough" encyclopedia articles, generally devaluating all other kinds of practices, especially if 160 he saw them as interfering with this goal. This emphasis may have been related to his vision of using Wikipedia's content in Nupedia, which he still saw as the more important project. However, there was nothing in this proposed policy, or in most of LA2's actual edits, that directly conflicted with Sanger's vision and the project's prevalent practices and policies. During the following days Sanger monitored LA2's edits, making on topic comments and minor edits where he felt it was appropriate. During this period, however, Sanger's comments remained courteous and polite158: LA2 productivity made him a valuable asset for the project. For a while, at least, it seemed that Wikipedia's status as a boundary object possessing the ability to accommodate the tendencies of various networks would prevail. This relative calm, however, did not last long. 6.2.3 Conflict and Policy Formation The conflict between Sanger and LA2 erupted after Sanger criticized a number of LA2's edits in a very short interval on May 20th. Most of his comments were directed at two editing practices characteristic of LA2's pattern of editing (though not limited to him): the creation of glossary-type articles explaining the meaning of computer-related terms and of articles comprised mainly of links to external websites. As noted above, Sanger feared the prospect of Wikipedia incorporating large amount of dictionary type articles and the new focus on computer related terms threatened to do just that. As for the second type of articles, Sanger feared of Wikipedia becoming a "list of link" should be interpreted, at least in part, as a manifestation of its position in the field of internet knowledge resources. DMOZ, One of its main sources of inspiration, was exactly that – an index of links. Thus, in order to establish its place 158 One example was the Trojan horse article (Wikipedia, 2001bj), which initially described only the computer related meaning of the term. Though veteran editors soon added a description of its mythological origins, the computer related definition remained first on the page. Thus, on May 19th Sanger moved this definition to the end of the article, noting on the edit summary that "This isn't an encyclopedia of computerese-and-oh-also-a-little-Western-civ. :-) So, I put the computerese sense second, hope no one minds." Another example was on the hacking article (Wikipedia, 2001bk), where Sanger commented "What are some methods that hackers use? What have (some) hackers done to earn their reputation (unauthorized access)? Have there been some famous hacks? I ask these questions because I'm concerned that the above is just like an entry in a dictionary of usage, and Wikipedia isn't a dictionary of usage, it's an encyclopedia". 161 and identity – and especially considering the ease and practical sensibility of loading articles with links instead of actual content – Wikipedia had to be steered away from that option. The trigger for the conflict was an edit on the Boston article LA2 created, comprised mainly of links to external websites (Wikipedia, 2001bl). Sanger's edits to the article itself were few and subtle: he replaced one external link (to the Boston Globe's website) with one directing to a new Wikipedia article he opened on the subject, and added the headline "links" above the links to the cities websites, thus signaling that it was not in itself article content. The important part of this edit, however, was the edit summary which read "Wikipedia is a not a list of links, it's an encyclopedia. :-)". The smiley sign at the end of the comment is supposed to create a feeling of lightness and diffuse any kind of tension it may generate159. Some twenty minutes later, LA2 posted a comment on Sanger's user page: Comment on "Wikipedia is a not a list of links, it's an encyclopedia. :-)": There is no way we can input as many facts (about a city such as Boston) into Wikipedia as there are on the city's official website. If Wikipedia is "a way to find information", then links to external websites should be an excellent means to this end. (Wikipedia, 2001bm): Though LA2 was probably unaware of the meaning of his assertion, he actually destabilized Wikipedia, sending it into a critical situation (Thévenot, 2002a, 2002b) where it was no longer clear what it is and what regime is governing the practices in it. While Sanger attempted to model the form of Wikipedia's articles on that of traditional encyclopedias, LA2's editing practices and justifications suggested that Wikipedia articles should be viewed as "a way to find information", rather than an encyclopedia articles per se. The suggestion that Wikipedia should be made to fit the informational needs of readers as soon and as efficiently as possible was implicitly an attempt to replace the vision of the project based on the encyclopedic model with one based on practical usability and relative disregard for its traditional form. 159 In the interval, Sanger made another gesture towards LA2. On 6:36, he added a post on the Wikipedia Announcements page (Wikipedia, 2001a), Reporting the rapid increase in the number of pages in Wikipedia – jumping from 3969 pages containing at least one comma to 4985 in just ten days. Sanger attributed this increase to "a few relatively new hands like KoyaanisQatsi, LA2, Wathiik, Mjausson and others," thus directly and publicly acknowledging LA2's contribution. 162 It took Sanger considerable time to react to LA2's comment, as he did not make any edits to Wikipedia for more than a day160. When he returned, he started an all-out attack on LA2. He first answered LA2's comment on his personal userpage (Larry Sanger), stating that: I agree with you, but I meant exactly what I said. Wikipedia is not a list of links. I think it is a very, very bad habit to get into, to start just listing links in place of a real encyclopedia entry! This laconic response, devoid of real justifications other than the invocation of the encyclopedic form, acknowledged the validity of LA2's suggestion and his reasoning, but refused to accept it. It was an insistence on the supremacy of the original vision for the project, based on the model of the encyclopedia, even at the expanse of efficiency. Sanger felt confident enough in his authority to assert it without a minimal justification or explanation. After posting his response, Sanger continued to fix other edits made by LA2. His pattern of editing uncovers the intentional inspection of LA2's edits, made possible through Wikipedia's scopic media. The first such edit was on the scheme (Wikipedia, 2001bn) article. The article originally referred to a programming language by that name, and LA2 added a general definition of the term. While doing so, he edited it to resemble a dictionary article, enumerating the divergent meanings and adding a part of speech in the beginning of each meaning, so it looked thus: n Plan or catalog over objects in a group. The same objects can be categorized by more than one scheme. See also Wikipedia category schemes. npr Short for Schemer, a variant of the programming language Lisp, where the available commands have been considerably reduced to make things much easier to work with. Sanger deleted LA2's contribution, removing the enumeration and the description of part of speech, and replacing the first definition with the sentence "see conceptual scheme". He then started a scheme/Talk (Wikipedia, 2001bo) page, and explained his action, stating: To create an entry, "scheme," just in order to give a brief definition seems unproductive to me. On my view, articles that simply give a brief, mediocre definition of a word are best left 160 Sanger explained his absence by blaming internet connectivity problems. During this interval, LA2 made over 180 edits. 163 uncreated; the notion of a conceptual scheme is much-studied by philosophy. ''That'' deserves an encyclopedia entry--not a definition, but a discussion of what has been said about conceptual schemes. Sanger's move invoked again the traditional model o the encyclopedia to attack LA2's edits. However, this time the problem was not the lack of content, but rather its type (definition) and quality (mediocre). Moreover, his comment this time was much less polite, pointing out substantive shortcomings in LA2's edit, not just to its inconsistency with the encyclopedic form. LA2 answered Sanger's comment on the same page eight minutes later: Larry, what are you trying to tell me here? Try to be constructive. An entry for "Scheme" already existed, describing the programming language. I wanted to point out that the word Scheme also has a more general meaning. Anybody is free to add pointers to conceptual schemes or whatever they find relevant. Anybody arriving at the entry for "scheme" (by search, link, or otherwise) should be interested in finding a roadmap of different meanings of the word, or at least I would. -- Should we be silent about the non-programming language meaning of the word? Should I go on to write about conceptual schemes? (Sorry, I hardly have the knowledge.) Or should I simply go away? In contrast with his previous response to Sanger, LA2's response in this case was much more defensive. Refusing to accept Sanger's construction of the project as a traditional encyclopedia in the strict sense, he justified his actions once more in practical terms: the general meaning of the term was something he as a reader would be interested in, and hence it merits inclusion. This move transformed his article into a qualified object within a specific order of worth, based on practical usability. However, LA2's comment was first and foremost an expression of frustration. Unable to follow Sanger's instructions, and reluctant to except his articulation of the project, he first raised the option of leaving the project. Sanger probably didn't have a chance to read LA2's reaction, as he was already reacting to his edit on another article page, which was even more problematic from his point of view. This chronological discrepancy, born from the medium temporal and spatial properties, prevented the option of reaching an agreement through negotiation, and resulted in unilateral actions which would have significant results for both participants and for the project in general. 164 While LA2 replied to Sanger on the scheme article, Sanger edited the article titled Am (Wikipedia, 2001bp), started by LA2 a few minutes earlier. In it, LA2 listed 5 different meanings of the word, again with enumeration and a description of parts of speech, so it closely resembled a dictionary article. In contrast with his edit in the scheme article, Sanger did not alter the content itself, and neither did he start a talk page. Rather, he added in the bottom of the page a link to a new page titled Wikipedia is not a dictionary (Wikipedia, 2001as). He then immediately added the same link to LA2's user page (Wikipedia, 2001bq), in a comment stating "LA2, in my opinion, Wikipedia is not a dictionary, and it shouldn't be treated as one". Five minutes later, he created the actual content of Wikipedia is not a dictionary. It read: Wikipedia is not a dictionary. Some Wikipedians have had a habit of starting an article, listing several senses of a term, and then stopping working on that article, as though they were interested only in giving the meanings of various terms. I don't like this habit. It encourages in new Wikipedians the idea that we are, in fact, just developing a dictionary. While on the one hand I am delighted that Wikipedia is growing in ''breadth,'' I am coming around to the view that breadth ''at the expense'' of the very notion of what we are working on is a bad idea. As I see it, brief, "stub" articles are to be encouraged--as long as they aren't just definitions. They should contain encyclopedic information of some sort. Not just the meanings of terms. Moreover, there are plenty of senses of terms that ''aren't'' of interest ''in an encyclopedia.'' They would be ''in a dictionary,'' but Wikipedia isn't a dictionary. --Larry Sanger Sanger strategy followed the pattern outlined by former occasions of policy formation, and especially the one employed by Wales during the creation of the neutral point of view page: he transmuted a practical rule designed to de-legitimize a specific practice into a policy, moving from specific cases to a higher level of abstraction through form investment and blackboxing. It built on the previous formulation that "Wikipedia is not a dictionary" to establish a new standard for editing practices which went far beyond the original, literal sense of the statement. In a move similar to Wales', he gave the policy a page of its own and blackboxed it with 165 a catchy name. He then advanced the page to the core network of Wikipedia's policy pages by adding it to the rules to consider page, where he gave a shorter formulation of the policy, leaving out the first person parts and the reference to preexisting cases. The process of form investment that Sanger performed in this instance placed Wikipedia articles within one of two categories: dictionary articles comprised of definitions, or encyclopedic articles containing "encyclopedic information of some sort." This distinction was constructed as essential to determine the legitimacy of articles and edits. The issue of practical usability, promoted by LA2, could not be articulated within that frame, as it was part of a different regime incommensurable with the one suggested. This move transformed LA2's articles into qualified objects that should be tested (Boltanski and Thévenot, 2006) against the demands of a regime he himself did not acknowledge as relevant. However, this transformation was only a byproduct: LA2's articles were not explicitly mentioned in the page, and were related to it only through links. They functioned as scaffolding that could be removed once the policy was in place, and moved from the center of the drama to its margins, demoted from critical moments to mere examples. The importance of the policy lay mainly in the contrast between typified cases of unwanted editing and an ideal form of the encyclopedia,, and the specific guidelines it gave as to which types of edits are considered as legitimate and which do not. However, the policy's aim went beyond its title: For he did not only target dictionary-like articles, but the whole strategy of "building the web," which was based on focusing on Wikipedia's breadth at the expanse of depth. Further collaboration for this interpretation can be found in Sanger's reply to LA2's insinuation that he might leave the project. Immediately after posting Wikipedia is not a dictionary, Sanger found the reply left by LA2 on scheme/Talk and answered it, urging him to stay and explaining his position: […] No, I don't want you to go away! I'd just like you to concentrate your energies on your areas of expertise. What irritated me (I admit it, I was irritated) was that you created a link to an article, called "scheme," on a topic on which there is expert "knowledge" (i.e., what philosophers theorize about conceptual schemes), and yet what you wrote did little more than give a rather mediocre definition of a term that any reasonably well-educated person understands. I don't see what value there is in doing that, other than to create a node in a web of mediocre dictionary 166 definitions. This inspired me to write Wikipedia is not a dictionary. […] This comment gives a deeper insight into Sanger's actions, rooted in a vision that demanded that every effort to expand Wikipedia meets a minimal standard of quality. In contrast with Wikipedia's institutionalized production processes, Sanger suggested that the way realize this standard is to "concentrate your energies on your areas of expertise." However, the actual strategy Sanger employed was not based on insisting that people should write on their areas of expertise, but rather that they should not write about subject when their knowledge does not exceed that of "any reasonably well-educated person." In light of this interpretation, the distinction between encyclopedia and dictionary thus takes a double meaning: it is both a distinction between different cultural forms and between different types of producers endowed with different epistemic virtues. The dictionary is identified with the layman, who may know the general meaning of a term, while the encyclopedia is identified with the expert, who should know the "expert knowledge" – that is, what academic experts have to say about the subject. This distinction does not break away completely from the immediate context of the discussion: the "mediocre" knowledge of the dictionary is explicitly linked to the vision of "building the web", which Sanger refers to as an attempt to create a "web of mediocre dictionary definitions." By the time Sanger finished formulating the policy and answering LA2 the latter had already made his move and posted his final say in the matter: on May 21st, 12:54, LA2 added a comment to the Wikipedia is not a dictionary page saying "There is a very simple solution to your problems. I'm out of here--LA2". Two minutes later he posted a similar comment on his userpage, and stopped participating in the project161. In contrast with Richard Kulisz' departure (outlined in the previous chapter), LA2's disappearance generated considerable uproar and undermined Sanger's policy initiative. In contrast with Kulisz, LA2 was a highly productive and integrated member in the community. Moreover, while Kulisz' limited his posts to article and talk pages, LA2 left his final message on the very policy page that was the 161 While he eventually returned to the project five months later he did not do so as long as Sanger headed it. 167 pretext for his departure. Following his announcement, Sanger and various other editors posted messages urging him not to leave the project, but to no avail. He never responded to them, or acknowledged that he read them. Those messages, while having no effect on LA2, undermined Sanger's new policy initiative.162 Their presence on the Wikipedia is not a dictionary page meant that it could by no means be constructed as an effective consensus. While some of the editors sided with Sanger's position or offered again to build "Wikitionary" (or a special space in Wikipedia), others directly contradicted Sanger position, as they presented it as his personal opinion rather than an established policy. For example, STG (Stefan Gilbert), a veteran editor, posted on the Wikipedia is not a dictionary a comment urging La2 to "stick around" and adding that "Larry has some opinions (which weren't directed at anyone personally) that aren't necessarily shared by everyone. Personally, I don't see a problem with dictionary-style entries". As a result of this and other comments, Sanger soon conceded his attempt to set a general enforceable policy such as neutral point of view. He stated that "for the record, I entirely respect STGs opinion and that smart, honest people can disagree about this," and that "I am just giving my opinion. Remember, this is a Wiki." Sanger's withdrawal closed the subject with a reasonable compromise that all parties could live with. In the days following LA2's departure, Sanger dropped the subject of dictionary articles. Thus, the result of the conflict between Sanger and LA were indecisive: while Sanger articulated Wikipedia is not a dictionary as an important policy and drove LA2 out of the project, LA2 departure and the support he subsequently got from various participants forced Sanger to give up any hope of making it into a binding policy, at least for the time being. However, LA2's departure severely damaged the "building the web" approach, as it left it without its main promoter and its most avid implementer. Finally, it is noteworthy that throughout this controversy Jimmy Wales, Sanger's boss and co founder, remained silent, leaving the actual fighting to Sanger and LA2 and refusing to take a position on either side. One anonymous editor even started a page called Wikipedia is a dictionary (Wikipedia, 2001br), which Sanger promptly edited to read "No. Wikipedia is not a dictionary. Stop treating it like one. Wikipedia is an encyclopedia." 162 168 6.2.4 Larry's Policy Columns While the social drama around LA2 ended rather quickly, it echoed through the project for some time later, as Sanger felt compelled to reiterate his position on the subject. On June 10th, almost three weeks after the end of the discussion on Wikipedia is not a dictionary, Sanger started a new subpage in his userpage, titled Larry Sanger/Columns (Wikipedia, 2001ap) as a location for presenting his thoughts on various Wikipedia related subjects. The first of these columns, Why a list of the senses of a word is not an encyclopedia article (Wikipedia, 2001bs) explained his reasons for opposing articles such as scheme or Am. His main argument against these articles was once again based on the distinction between dictionary and encyclopedia. Sanger began by explaining the need for the column, implicitly referring to the LA2 controversy: some people disagree with his view, "or simply do not understand" them. He then went on to state the basic assumption that guided him, invoking the concept of the encyclopedia: "it is obvious that Wikipedia (like encyclopedias generally) is not in the business simply of giving dictionary definitions…in order to be proper encyclopedia articles, they will contain some non-semantic ("synthetic," if you will) information about the subject of the article." Thus, the problem with listing senses of words is that "in place of a proper encyclopedia entry, listing out many different senses of a word has in English, inevitably leads to the listing of very many definitions that are not topics of encyclopedia entries." Sanger columns drew some responses, who pointed out that dictionary like stub articles are important for the creation of more comprehensive articles. Sanger insisted on that he only resisted "lists of the senses of words, some of which are not the subjects of encyclopedia articles", as they "won't become articles unless someone deletes most of them and replaces them with a proper article or two". Both the responses and Sanger's replies to them were written in a calm and reasoned tone. The drama, it seems, had died out. By June 12th, when Sanger added a link to Why a list of the senses of a word is not an encyclopedia article from the older debate Encyclopedia and or versus dictionary (Wikipedia, 2001ax) the discussion had already died out. Sanger use of his personal page as a venue for personal columns instead of a policy page signaled a shift in attitude, as he all but stopped attempting to create or 169 enforce policies but rather chose to set a personal example, and promote his views through unofficial columns on his private page. In the following month, he slowly and carefully began modifying content that he considered at odds with it. He added questions and comments to the pages of very short articles, giving a clear indication that they were flawed and urging other editors to elaborate them. He placed all the external links under a sub-title "links," distinguishing them from the article itself and changed internal links as to leave only those specifically relevant to the context of the article163. In cases where the editor or editors of an article were known, Sanger requested they fix what he saw as their shortcomings – though in a much more polite manner than he did before, and while attempting to avoid direct conflict and arguments, preferring to do the necessary work himself or noting that it should be done rather than confronting and reproaching editors that created articles that needed such work. Sanger deleted content that he considered dictionary-like and with no hope of becoming encyclopedic only in cases where the writer was anonymous, meaning that there was no specific user to engage in discussion and at the same time no one likely to challenge the deletion164. In other such cases, he completely replaced content he considered as un-encyclopedic with original content of his own creation165. Sanger's did not meet any opposition, and in some cases other editors even copied his actions. On June 29th, Sanger ran into a large amount of very short articles, mainly on persons related to medical and cultural subjects, started by a single anonymous editor identifiable through his IP address166. Most of these articles contained only dates of birth and death, and occupation. While they were very short, they clearly did not fall into the category of dictionary articles. Upon fixing a layout problem in an article on Simon Flexner (Wikipedia, 2001bw), containing only his name and dates of birth and 163 As he previously did on LA2's Boston article. For example, on June 8th he added the sub-title "Links:" to the article Disease and explained in the edit summary that "we should not use external links to substitute for info that we ourselves want to include, IMO. So links should go under 'Links:'" (Wikipedia, 2001bt). 164 For example, on June 9th Sanger deleted the content of the article Of, claiming that "[…] this isn't a dictionary...I mean, what encyclopedia-type stuff can you say about 'of'?" (Wikipedia, 2001bu). 165 For example, on June 30th, Sanger edited the basis article, which had two definitions in it, one pertaining to the mathematical meaning and below it other one defining it as "is a set of facts used as a starting point for an argument or larger discussion." Sanger deleted that sentence, noting on the edit summary that he is "removing dictionary definition of 'basis,' which is not ever going to be an encyclopedia article" (Wikipedia, 2001bv). 166 Wikipedia's edit history feature presents the IP of the editor who made a certain edit, unless he chooses a username. 170 death, he wondered in his edit summary "Why create an entry if all you know are the person's birth and death dates?" in response, Hornlo posted a comment on Sanger's userpage that tried to justify such articles (which he claimed that he did not create) as "provocative editing--creating, or changing, a page to get it onto the recent changes list, hoping it will provoke someone to flesh it out." Sanger, in response, explained why he saw such "provocative editing" as problematic: Good point, Larry, but it still annoys me. Yes, it's great to leave things undone, but these wimpy entries do add up and in the shortterm and medium-term, make Wikipedia look, well, superficial. That's what I don't like. –Larry In response, Hornlo suggested it "might be a good topic for a future column167: 'What makes a good /seed/basic/minimal article?', or a follow-on to the 'not a dictionary' column". On July 1st, Sanger answered Hornlo's challenge with a new column titled the perfect stub article (Wikipedia, 2001bx). Sanger began by stressing the openness of Wikipedia, and the importance of stub articles for it: Part of what makes Wikipedia work is that we do not require perfection, and I strongly believe we should continue adding as much imperfect stuff to Wikipedia as we possibly can […] Most of us can agree that stub articles are good things. They help to fill out the breadth of Wikipedia, giving us actual content to work with in our attempts to organize and present all of human knowledge and--important from a practical point of view--giving the search engines more to link to. This means more traffic and therefore more contributors--a virtuous cycle." However, he reasoned, too many stub articles may hurt Wikipedia's legitimacy, as they would make it "look frivolous." As a practical solution, Sanger developed the concept of "the perfect stub article" – one that is productive yet respectable. Among other qualities168, Sanger offered that such article should include "a little more" than just a definition. He also specifically noted that editors should "make sure that your topic (and therefore your definition) is one on which we are 167 168 Link to Larry Sanger/Columns (Wikipedia, 2001ap). Such as the use of proper English, accuracy, lack of bias and proper links to other articles. 171 going to want an actual encyclopedia article. In other words, bear in mind that Wikipedia is not a dictionary."169 In his explanation for insisting on "a little more" than just a definition, Sanger pointed out that the "build the web" attitude may be detrimental to the vision of Wikipedia eventually becoming a "real" encyclopedia: Now, why do I keep harping on this? I have a hypothesis, which I think is a very reasonable hypothesis. I think that it is important to the psychology of Wikipedia that we all understand ourselves as not writing articles that simply identify people, events, and concepts, in a very basic way, but that actually give details, "empirical facts," content. That tiny extra bit of content is very important, psychologically speaking, because every time it is written, or read by another contributor, it makes it clear that the project is indeed eventually going to be about going deep into all these subjects. Finally, Sanger pointed out that "you can always ignore these guidelines entirely, and someone will probably fix the article for you! That's the beauty of a wiki." The perfect stub article summarized Sanger's position regarding the double tension that permeated Wikipedia at the time: the specific tension between the "build the web" approach and the "Wikipedia is an encyclopedia" approach, and the broader and partially overlapping tension between an the freedom of editors to contribute to the project in any way they see fit, and the enforcement of standards derived from encyclopedic concept. Sanger's attitude in both cases was one of compromise: he insisted that there should be a general standard for the evaluation of contributions based on a vision of Wikipedia's future as a "real" encyclopedia, with in-depth articles of the type that can be found in traditional paper encyclopedias. However, in contrast with the initial strict position he took in his conflict with LA2 and with the position the projects founders took on the issue of the Neutral Point of View policy, he was much more liberal with regards to edits that are at odds with this standard, preferring a posteriori editing mechanisms to strict enforcement. This attitude, manifested also in the movement from policy pages to personal commentary pages, represents the 169 However, Sanger introduced two legitimate exceptions to that rule: first, pages that explain jargon necessary to understand genuine articles; Second, pointer pages "consisting of a list of several divergent senses of a word, each defined on the page, and each definition followed by a pointer to an article where the topic, in that sense, is discussed in more depth". These exceptions, and especially the later, build on the counter arguments Sanger ran across in the previous conflict, which he then co-opted to articulate his position. 172 lessons learned from the Wikipedia is not a dictionary conflict, as Sanger explained his point of view, sought to make it as liberal and inclusive as possible, and acknowledged the difficulty of enforcing an unpopular policy in Wikipedia. Sanger continued acting on his policies, editing dictionary like articles and deleting articles dealing with subject he considered un-encyclopedic. More significantly, he added links to the Wikipedia is not a dictionary page on the relevant talk pages, strengthening its position as a privileged passage point and conscription device (Henderson, 1991). In addition, he removed brackets170 from terms he considered unworthy of an article before actual content was added to the resulting articles, thus preventing in advance the creation of dictionary articles. Sanger was not alone in this pursuit: other prominent editors such as Hornlo, Lee Daniel Crocker and Tim Chambers Koyaanis Qatsi (the most industrious editor at the time) explicitly supported his position and began editing contributions that failed to adhere to it. This signaled a deeper change in Sanger's position in the project. While formerly, Sanger took the role of an equal participant in policy discussions, he emerged from this conflict in a position of leadership by example, albeit one with limited authority. It was clear from this episode that Sanger could not expect automatic support when attempting to impose his view of Wikipedia's form, as his first attempt backfired. On the other hand, though adherence to his approach was not universal, it was widespread enough to have a lasting impact on Wikipedia's production practices. Using subtler measures than actual policy formation, Sanger enlisted support for his views in a case-by case manner, and let them be known on his columns, appearing only in his personal user-space. However, this new position did not last long: when of new participants started swarming the project, Sanger attempted once more to achieve a more substantive measure of authority in the project. The relative success of this attempt, described in the next sub-chapter, resulted in the final institutionalization of the "Wikipedia is not a dictionary" policy. 170 Enclosing and expression in brackets in Wikipedia's user interface created a new page for the terms surrounded by them, if one did not exist previously. 173 6.2.5 Crisis and Resolution – Conscription Devices and Scopic Mechanisms Starting on July 25th, unprecedented amounts of traffic followed links from an article Sanger posted about Wikipedia in Kuro5hin.com (Sanger, 2001g). A day later, Wales linked to article in a post he published on the highly popular collaborative content website Slashdot.com (Wales, 2001i), causing another surge in traffic and a wave on new, and inexperienced participants. Many of these editors immediately began editing articles without acquainting themselves with Wikipedia's policies and customs. On the following days Sanger and other editors tried to correct and direct the new editors, and fix the myriad dictionary articles they started, mostly on computer and geek-culture related subjects (see more below). Consequently, Sanger began doing an overhaul of Wikipedia's conscription devices, the core network of pages directed at informing participants of Wikipedia's policies, such as the homepage, Welcome, Newcomers (Wikipedia, 2001by), the FAQ (Wikipedia, 2001bz), Wikipedia Policy (Wikipedia, 2001ca) etc171. On the same occasion, he edited Wikipedia is not a dictionary to look more like a policy page. The new version of the policy, edited by Sanger on July 28th (and later proofread by Koyaanis Qatsi), read less like the semi-personal statement it originally was and more like a general policy. While the old version was directed at veterans, the new one was directed at new editors, as the opening sentence was changed from "Some Wikipedians have had a habit of starting an article, listing several senses of a term, and then stopping working on that article", to "It's natural, when you see Wikipedia for the first time, to mistake it for a dictionary." Where the old version used the first person, the new one used the third. Personal statements like "I am delighted" or "I am coming around to the view" were deleted. The new version even addresses Sanger itself in the third person, as it stated that "If you do just give the meaning of the word, nobody is going to be mad at you (except maybe Larry Sanger, but then, he gets mad at everybody :-) since he cares about Wikipedia so much)." However, as this last example shows, the new formulation is also much softer than the 171 A major policy page that Sanger devised in this occasion was what Wikipedia is not (Wikipedia, 2001cb, and see below), which subsequently became a general standard distinguishing between legitimate and illegitimate content. While the first clause of policy page repeated the content of Wikipedia is not a dictionary, Sanger kept the original Wikipedia is not a dictionary as a separate policy page. 174 former. Sanger did not state his position as a rule, but rather as an opinion not held by everybody, and acknowledged that breaking it is "probably OK, in most cases". This version of the policy stayed rather stable over the following months. While a few dissenting views appeared on Wikipedia is not a dictionary/Talk (Wikipedia, 2001g) which was created in august 17th, Sanger's responses and explanations were ultimately well received, and with the help of other editors backing his point of view no real opposition was raised against what has become an effective consensus and a cornerstone policy in the project172. In the months following the policy's institutionalization, various conscription devices (Henderson, 1991) and scopic media (Knorr-Cetina, 2005) were developed in an effort to enable its enforcement, counter dictionary articles and expand stub articles. The first to develop such a system was Manning Bartlett, who kept a list of stubs in his personal page's to-do list, and later moved them to a page titled Wikipedia Utilities: find or fix a stub (Wikipedia, 2001cd). Wikipedia utilities/Pages needing attention (Wikipedia, 2001ce), created in an unknown later date (prior to November 8th) had a section named articles that seem to be nothing more than definitions, directing editors to delete or expand articles listed in it. Similarly, the Requested articles (Wikipedia, 2001cu) page had a list of articles that were started but remained as stubs. On November 21st, Sanger moved this list to the already existing but largely unused Wikipedia Utilities: find or fix a stub. This page became the main location for handling such articles, spreading the work among a number of veteran editors that made sure these articles were dealt with. To conclude, the articulation of the "Wikipedia is not a dictionary" policy, followed a pattern similar to those preceding it: the arrival of participants with new editing practices and visions of the project resulted in a conflict; that conflict ended in 172 However, the policy page and its Talk page did, for a time, became involved in conflict, as controversies regarding authority in project and quarrels between Sanger and the Cunctator began intensifying in the beginning of October. On October 4th, an editor known as "the Cunctator" added a note on the page itself, below the statement that dissenting from the policy is "probably ok", noting that "Actually, the 'Wikipedia is not a dictionary' philosophy is, as of September 2001, being used as a justification to delete stub entries". However, this comment, and the a similar allegation he raised in his essay how to destroy Wikipedia (Wikipedia, 2001cc) – a consequential essay which was essentially an attack on Sanger and Wales, part of it dedicated to reproach Sanger's deleting of pages he saw as inconsistent with Wikipedia's policies – were not directed at the policy (which he supported) but on Sanger's deleting of pages he saw as inconsistent with Wikipedia's policies. Thus, this controversy paradoxically reaffirmed the policies effective consensus rather than threaten it. 175 the creation of a new policy based on a practical rules transmuted into policies through investment in the encyclopedic form, blackboxing and the creation of conscription devices and scopic media. On the level of policy institutionalization, it uncovered the limitations of power and policy formation in Wikipedia, as it was the first one that was met with reasoned and relatively widespread opposition, and had to be qualified and downplayed. This was, perhaps, a result of the noninvolvement of Jimmy Wales, Wikipedia co-founder and owner, who had an important role in the institutionalization of the previous content policies. On the other hand, its subsequent successful institutionalization demonstrated that Wikipedia is capable of sustaining policies that are not consensual, on the conditions that it was codified in policy pages, had reasonable justifications, that sufficient editors pursued it, and that adequate scopic media were produced to support their efforts. The Wikipedia is not a dictionary policy was responsible for affixing the notion that Wikipedia's goal is more specific than producing an innovative reference work, or a "way to find information", as LDC and LA2 suggested. Wikipedia was constructed through it as an encyclopedia, first and foremost, with unique characteristics derived from the history and conventions of the genre. This articulation paved the way to further policies dealing with types of content not prevalent in traditional encyclopedias. These policies were grouped to create one of Wikipedia's most comprehensive and important policy pages What Wikipedia is not. 6.3 Phase 3: What Wikipedia is Not This sub-chapter describes the institutionalization of one of Wikipedia's most comprehensive and elaborate policies, What Wikipedia is Not (Wikipedia, 2012g). This policy had its humble origins in an appendix to one of Sanger's columns, in which attempted to enlist support for an effort to turn pages containing only lists of links to traditional prose articles. To it, he appended "a list of things encyclopedia articles are not". It summarized several content policies and various issues the founders feared might derail the project. Its subtitle was: "Wikipedia is an encyclopedia. Here is a list of Things That Encyclopedia Articles Are Not." Below was a list of cultural forms and types of content related to them that were proclaimed 176 inappropriate for Wikipedia. After every clause, there was an explanation in parenthesis of the permissible way to engage in similar types of content. Various types of content labeled as illegitimate by the (un)virtue of their associating with unencyclopedic cultural forms, such as discussions (associated with "Discussion forums, or Everything2 or H2G2 nodes"); Dictionary definitions of common English words, or lists of such definitions (associated with dictionaries); descriptions of word usage, and particularly jargon (associated with word usage guides), etc. these types of content all represented actual issues that Wikipedia had to practically engage in and come to terms with. As such, they were all subject to controversy. However, these controversies were transmuted and codified as consensual – even in cases where it was clearly not the case. Their codification produced a general guide for writing in Wikipedia articles that functions as such till this very day. 6.3.1 The Controversy over Lists of Links From Wikipedia's very first days, it contained a considerable amount of pages comprised only of lists of links to other articles. As the means for creating a new page was linking to it, lists of links were an important tool for initiation of articles and for mapping subjects that needed expanding. This was the case with many of Wikipedia's early articles, especially the ones dealing with general and overarching subjects, such as Countries of the world (Wikipedia, 2001cf), Philosophy (Wikipedia, 2001cg), Music (Wikipedia, 2001ch), Sport (Wikipedia, 2001ci), Disease (Wikipedia, 2001bt), Psychiatry (Wikipedia, 2001cj), etc. Such articles were routinely created by various users, including Wikipedia's founders, and wasn't considered problematic. The new "build the web" approach, however, intensified their creation and brought them to the center of attention, with the new emphasis Sanger put on comprehensive and original article content. List of links articles came under scrutiny about the time the LA2 affair was fading. On June 2nd, Stefan Rybo, one of the very few editors in the German Wikipedia who also frequented the English Wikipedia, sent a mail to the Wikipedia-l mailing list (Rybo, 2001) complaining about "huge link lists" that are "often intermixed with text (content)" and suggested a "separation of content and navigation 177 items (as content being true nodes like in a computer data tree)". The allusion to "computer data tree" uncovers the attitude behind it: that of a computer programmer used to working with data arrays and structures. From this point of view, the problem of lists of links was the mix between link lists and actual content. Sanger (Sanger, 2001f), the only editor who replied to the post, did not see the problem and confessed that he found the phenomena "delightful". Moreover, he objected the very premises of Rybo's argument when he claimed that he actually preferred articles that added "some 'content' (i.e., descriptive prose) to a list", and claimed that Wikipedia being hypertext, the separation between content and navigation is redundant. Contrary to Rybo's position, Sanger actively endorsed the addition of prose content to pages comprised of lists of links. Sanger's position was in line with his general attitude towards Wikipedia, stressing the creation of article content and downplaying all aspects of the project that did not contribute to it. From this point of view, the creation of navigational aids was a waste of time and effort – but the transformation of link lists into prose was productive activity. Sanger's opportunity to publicly present his view on Wikipedia presented itself a few days later. On June 9th, when Sanger noticed and answered a question posted by LA2 to the Naming conventions/Talk (Wikipedia, 2001ck) page on the latter's very first day in Wikipedia – before he picked up a username, so the author was not identified. LA2 described his dilemmas while editing the Albert Einstein article. Loyal to his "build the web" approach, he saw it as "tempting to create a link to a page listing various physicists through history, and pointing out that all physicists are also scientists." The main issue that troubled him was […] Whether the structure or catalog of professions should be created before we start writing biographies of other scientists, like a coordinate system is drawn before the data points are plotted. Such catalogs could be: professions, academic disciplines, timelines (like the excellent one on geologic ages), families of biological species, types of organic chemic substances, etc. The risk with such an approach is that the catalogs will dominate over the substance contents (Yahoo without the Internet). The risk with not taking such an approach is that a lot of useful links will be missing from entries (the Internet without Yahoo). While LA2's position, similar to Rybo's, was based on the distinction between content and navigation, and sought to strike a balance between what he saw as 178 opposing tendencies, Sanger's answer denied the very legitimacy of such distinction in Wikipedia: […] lists of links are great and absolutely necessary in a hypertext encyclopedia, so we gotta have 'em. I just think we should discourage the idea that an adequate article might consist just of the links. E.g., I still think actress should address the phenomena that are actresses, and not just give a list of actresses (as if people visiting the "actress" page were interested only in finding their favorite actress). In this answer, Sanger re-articulated the problem, posing it in terms of his "Wikipedia is an encyclopedia" approach, where by every page had to be evaluated in terms of its adherence to the standard posed by the traditional model of encyclopedic articles. As the example of the "actress" page indicates, he refused to separate navigation from content, subsuming the former under the latter. Per his general pattern of action in this period, Sanger stated his position in a much more careful and polite tone than he did in the previous conflict173 and immediately began implementing it himself, as he added short introductory text on pages comprised solely of a list of links, such as the Dewey Decimal System (Wikipedia, 2001cl) page. However, the most notable effort Sanger made on the issue took place on his home territory: philosophy articles. On June 20th, Sanger to engage in a major edit of the philosophy article comprised of a long list of links to various sub-disciplines, famous philosophers, etc. He wrote a general introduction, and added text around the various lists to make them into prose sentences dense with links. Over the following days, Sanger turned the huge link lists in many of the philosophy pages articles to prose articles, and opened new ones based on the prose section he wrote on the philosophy article. In addition to acting on his new initiative, Sanger urged other editors to follow his example. He first attempted to enlist them by placing comments on pages comprised of list of links, but as these pages were generally in the periphery of the network, this strategy had limited success174. Thus, he began making attempts to 173 In the end of his answer, he wrote that "My conclusion is that unless there's a really striking problem (as, ''in my opinion,'' the existence of many mere-dictionary type entries was becoming), it's not worth it to discourage ''anyone'' from doing ''anything.'' The beauty of Wikipedia is that, by being open to all manner of contributions, everyone wants to contribute. It's great that way". 174 One such attempt almost ended in open conflict: as he stumbled upon an article on VIM (a popular UNIX and Linux text editor) comprised mainly of a list of its features, Sanger asked its editors if they could "please take a minute or two and render some of the above list into full prose sentences? Have 179 promote his new initiative through Wikipedia's conscription devices (Henderson, 1991), attempting to reach a larger audience and convince participants to join his efforts. On June 21st, Sanger posted a comment on Wikipedia NEWS (Wikipedia, 2001cm), a page used to publicize new entries and developments on Wikipedia. In it, he stated that The Philosophy main page has been updated, with new content and a new organizational scheme. Larry Sanger is immodestly proud of this work, or so he says, and thinks that it shows how other toplevel pages could be rendered into a prose format. Biology is another top-level example that uses prose sentences intermixed with link lists to both introduce biology and to point the reader to specialized topics within the subject of biology. However, as Wikipedia NEWS was neither a popular arena for debates nor a policy page, and as his attempt of enlistment was very subtle, this announcement didn't have much impact on the project. Another attempt was made on June 29th, when, Sanger posed a question on the Wikipedia help desk (Wikipedia, 2001cn) on June 29th. This page, started as a place to host requests for general information, was yet another conscription device, albeit not a central one. His question read: How do the different parts of a computer (processor, memory, etc.) work together with software to create a functioning desktop computer? I imagine this should be answered on computer, which now has a paltry line, "programmable machine that processes data," and a link to computing, which page consists just of a list of links. Even computer hardware consists of a list of links. All I wanna know is how a computer works. :-) This effort had no immediate consequences, as no one bothered answering Sanger's question or fixing the articles. Thus, in another attempt to enlist other participants to join his efforts, Sanger posted another column on July 5th, that would prove to be of immense consequences, even though it carried the somewhat awkward name List-o-links (Wikipedia, 2001co). fun!" A day later, an anonymous editor erased his message, and he restored it a few hours later. The following day, after another editor added a list of internal links, he rephrased his message, stating that "The above paragraph isn't really what we should expect in an encyclopedia article--i.e., it shouldn't be a list, it should be ordinary English prose sentences." Finally, on June 21st, the anonymous editor wrote a couple of text paragraphs, which Sanger edited a day later to his satisfaction. 180 6.3.2 Things That Encyclopedia Articles Are Not Sanger opened the column with a meek attitude, calling its subject one of his "pet Peeves", warning the reader not to misconstrue it as an attempt to "bury Wikipedia in negativism" and stating that "Almost everyone should read this and then go right back to doing whatever it was that they were doing before they read it--I only need a few, good people to read, agree, and act." Only after these concessions, he turned to the actual subject of the column: "certain pages, such as sport, music, some of the film-related pages, and some of the computing-related pages, that consist of-ugh--lists of internal links." After describing his personal fondness of the concept of lists, he explained what he sees as the problem with such pages: "Now, I would like to gently point out that a list is, well, not an encyclopedia article." His concern was that articles dealing with the most general topics have no prose describing their subject, and readers can learn nothing about them from mere lists. He offered as a counterexample the philosophy article he edited, and suggested making a similar effort on other such articles. Once again, he finished the column with an acknowledgement of his inability to enforce his opinion, as he states: "I know. Go thou and do whatever thou wilt. But, as Spinoza said, 'All excellent things are as difficult as they are rare.' Inversely, all mediocre things are as easy as they are common." What turned this column into a critical moment in Wikipedia's history came as an afterthought to the column itself. After the end of the actual column, Sanger wrote: "I just felt inspired to make a list. Here is a list of Things That Encyclopedia Articles Are Not." what followed was a list of almost all the conflicts Sanger was involved in during the previous weeks, articulated and edited as to become a manual of style and a standard against which to measure the adequacy of article content. In each case, Sanger presented both a general rule he already set forwards in conflicts with other editors or in previous columns, and reservations and alternative that emerged from such interactions. This close connection between actual practices, conflicts and general policies, reiterates the notion that form investment and production of rules stem first and foremost from concrete social problems that emerge when there is no clear standard for the coordination of action, on a bottom up rather 181 than top down process. In what follows, I'll present these clauses, following each of them with a short genealogy and interpretation. Dictionary definitions of common English words that, in the sense in question, name no subject that any respectable encyclopedist would ever think of making the topic of an encyclopedia article. (But an article can and should always ''begin with'' a good175 definition or a clear description of the topic, as in the case of biographies.) It is clear why this clause is the first on the list, given the events of the previous months. It summarizes Sanger's Wikipedia is not a dictionary (Wikipedia, 2001as) policy, specifically addressing such articles as of (Wikipedia, 2001bu) and drunk (Wikipedia, 2001av), that he made a point of deleting or preventing their creation in preceding weeks. Lists of such definitions. (But an article can certain consist of a pointer to other pages, where a word is too general to have any one topic associated with it; see freedom and Columbus for examples.) This clause is a summary of Sanger's column Why a list of the senses of a word is not an encyclopedia article (Wikipedia, 2001bs). The counter-examples are disambiguation pages, listing several meanings of the word, all of which have related articles in Wikipedia. Such pages were (and still) considered vital navigational aids, and this justification was repeatedly used in related conflicts concerning such pages. Sanger, who acknowledged the importance of "pointer pages" in the column and in various other remarks, co-opted this justification, preempting any attempt to use this justification for practices he rejected. Propaganda or advocacy of any kind. (But an article can of course ''report objectively'' on what advocates ''say,'' as long as this is done from the Neutral Point of View. Go to Usenet if you want to convince people of the merits of your favorite views--and good luck.) This clause is, as it makes explicit, a summary of the Neutral Point Of View (Wikipedia, 2001ak) policy. Specifically, it echoes a number of incidents Sanger had on this issue in the in the preceding days. The first of them, with an editor named Philip2 who insisted on debating his opinions regarding the McCarthy period on the Unamerican activities talk page (Wikipedia, 2001cp), took place on June 26th. It ended only when Sanger answered to one of his replies by stating "Beyond this I won't reply, because I am not interested in using Wikipedia as a debate forum. If I 175 Link to the fallacies of definition article. 182 want to debate, I'll use another wiki, or else a mailing list or a newsgroup." Another such incident took place less than three hours before Sanger published List-o-links, when Sanger came across what he conceived as a biased edit of the Abortion article, to which he responded: "This is a perfect example of text that fails to adhere to a Neutral Point of View. If you want to contribute to Wikipedia, you must make a very strong effort to write from a neutral point of view. If you think this is impossible or a bad idea, please see Neutral Point of View, where all is made clear and plausible. Wikipedia is not a platform for advocacy of any kind, period" (Wikipedia, 2001cq). Mere lists of quotations and aphorisms. (But some such lists might be very nice to have to supplement encyclopedia articles, sure.) This clause echoes a debate Sanger had with several other editors, mainly SJC, Kpjas, and KQ on Literary quotations/Talk (Wikipedia, 2001cr) page on July 3rd. Sanger objected the existence of this page (and similar page) on the basis that they were "distracting attention from our main brief, which is to write an encyclopedia." Similar to his strategy in earlier conflicts regarding dictionary articles, he suggested that "Perhaps someone could start a quotations wiki. That might be fun." This clause represented an attempt for a compromise on the subject, as it suggested incorporating them into relevant encyclopedic articles, an innovative feature that traditional encyclopedia lacked. By virtue of articulating the relevant attributes of the encyclopedic article, Sanger was thus able to legitimize variations on it. Mere collections of external links. (But of course there's nothing wrong with adding both lists of links and lists of on-line references you used in writing an article). This clause targets two separate issues: articles comprised of lists of external links (such as LA2's Boston article), and link sections in legitimate prose articles. While the former was de-legitimized by pointing to its inconsistency with the encyclopedic form, the latter was constructed as complacent with that form. Even though the legitimate use links in article pages was still being worked out at the time, it became a common practice to concentrate them on the bottom of the page. This was the result of two separate practices: Sanger's demarcation of external links to a separate section during his conflict with LA2, and the courteous habit of linking to online resources used to create articles (especially when large quantities of content were copied from public domain websites). This practice was added by Sanger to the 183 Rules to Consider page176 about a week earlier (Wikipedia, 2001cs), and was now further institutionalized by virtue of its contrast with an illegitimate practice. Mere collections of internal links. (But of course, there's nothing with pointer pages when a word is too general for any one topic to be associated with it; and of course, it's very important to make collections of relevant internal links, as this conveys useful information and helps navigation.) Finally, the last clause restated the subject of the column that led to the creation of this policy, i.e. list-o-links. Below the list, Sanger added a note stating: "I will continue to add to this list as we discover interesting new ways of not writing encyclopedia articles. :-)", thus attempting to affix his position as a leader and articulator of standards, though he did not have real authority in the project. The list was a hybrid: it was written in a style reminiscent of policy pages, yet constructed through its location in the network and on the page (an afterthought to a column) as an expression of Sanger's personal opinions. It was an attempt to find a compromise between leading by example and through personal interactions on the one hand, and attempts to usurp authority using policy on the other. The list of thing encyclopedia articles are not represented a significant and sophisticated work of articulation. It may seem, at first glance, as distinctly modern (or traditional, in another sense), as its clauses encapsulate the logic of the modern encyclopedia, thus implicitly accepting its basic premises regarding the adequate summary and representation of knowledge. In another sense, however, it is a postmodern document, as it constructs the encyclopedic model as a system of differences, rather than essential characteristics. As each clause, or sign, distinguishes the encyclopedia from a distinct type of text, it is constructed as a part of a binary opposition. The combination of all the clauses does not rise to the level of a positive definition, but rather places Wikipedia in the center of a system of differences (Laclau 176 Sanger added to the rules to consider page on June 29th, suggesting "Don't use external links where we'll want Wikipedia links: Don't put in links, like this [here appeared a link to http://www.Wikipedia.com, L.G.], to external URLs linking text that we will want articles on Nupedia about. Put external links in a 'links' section at the end of the article." The trigger for the creation of this rule was a number of edits Sanger made on the same day while fixing and editing articles concerning statistics (almost exclusively written by the same author, Dick Beldin, during February and March 2001), where he replaced links to external sites in the body of the text to internal links, and moved them to a separate links section in the bottom of the page. 184 and Mouffe, 1985, p. 132). This construction liquefies the reference, to use Baudrillard's (1994) terms, as it replaces the original idea of the encyclopedia as a summary of current expert knowledge with a list of signs whose justification is not substantive or meaningful, but rather symptomatic. This allows for a much more concrete and productive setting of standards, as systems of signs are "more ductile material than meaning, in that it lends itself to all systems of equivalence, all binary oppositions and all combinatory algebra". (Baudrillard, 1994, p.2) Thus, Sanger's list did not just address current critical moments in the project, but also created a generative structure that can help solve future ones. The binary opposition formula stood at the basis of its clauses could be permutated to delegitimize a very wide variety of contents, based on their alleged incompatibility with the encyclopedic form, and assemble a coherent image of the encyclopedia articles the project sought to produce. It did not preclude, however, variations from the modern encyclopedic form, as it did not present a totalizing definition of the encyclopedia. In terms of its immediate effect, Sanger's column was more successful in enlisting support than the measures he tried previously, but less than actual policy pages. Three hours after its original publication, an anonymous editor transformed the list of links on the computer page into a preliminary version of an article (Wikipedia, 2001ct). A day later, on July 6th, Gareth Owen began adding prose descriptions to the music article. However, these were the only practical reaction to the columns. For a few weeks, it seemed that it would be the columns only significance. The long lists of links articles were all created in the very first weeks of Wikipedia existence, and new ones were not being created anyway. The "list of Things That Encyclopedia Articles Are Not" was comprised of guidelines and policies mostly already accepted by veteran editors, and its location, on a subpage of Larry Sanger personal page, made it into a personal commentary rather than a policy and placed it on the margins of the network. This marginal position was sustained by the fact that Sanger did not place links to it on places other than his personal userpage. This relative neglect came to an end on July 24th, when Sanger finally began advancing his column to the core network of policy pages. As Sanger came across the Nazi Germany (Wikipedia, 2001cv) article after some edits were made to it, he discovered a large subpage named Nazi Germany/related terms (Wikipedia, 2001cw) 185 that contained a list of terms from the period and their meaning. In the discussion on the Nazi Germany/Talk page, Sanger joined a dialogue between WojPob and SoniC, two veteran users who discussed a suggestion LA2 made on the main article page a month earlier to create a more general page listing various words originating in foreign languages. Sanger objected this idea, reproducing his formula of binary negative and claiming that "[…] Wikipedia is not a dictionary. It isn't a style guide, it isn't a collection of foreign borrowings, etc., etc. It is an encyclopedia.[…]" about ten minutes later, Sanger edited the Wikipedia is not a dictionary article, adding a link to it on the bottom, embedded in the sentence "Have a look at this personal essay177 for further reflections on what Wikipedia is not" (Wikipedia, 2001cx). This action had two important implications: first, it brought the list much closer to Wikipedia's conscription devices, placing it two links away from the rules to consider page. Second, and no less important, the headline originally referring to a list of Things That Encyclopedia Articles Are Not was changed to a description of what Wikipedia is not. This seemingly minute change in wording is actually an important transformation, as it reduced Wikipedia as a whole to an encyclopedia, and not just concrete article pages, rejecting any alternative formulation of it. In the following weeks, this articulation became the basis for a major new policy initiative. 6.3.3 "The Big Slashdotting" The emphasis on the idea that Wikipedia is first and foremost an encyclopedia proved very timely, as it happened merely two days prior to "the big slashdotting". Starting on July 25th, unprecedented amounts of traffic followed links from an article Sanger posted about Wikipedia in Kuro5hin.com (Sanger, 2001g), and was later promoted by Wales on Slashdot.com (Wales, 2001i) linked to from the highly popular Slashdot.com. The result, as Sanger phrased it, was that there were "many eager, but clueless newbies about" (Wikipedia, 2001a). Some of these editors immediately began editing articles without acquainting themselves with Wikipedia's policies and customs. This proved a problematic moment for the project, as many of them shared common interests in topics that challenged the traditional encyclopedic form, such as hacker culture, conspiracy theories and popular culture fandom, with its emphasis on 177 Link to Larry Sanger/List-o-links (Wikipedia, 2001co). 186 the minute details of said works. Moreover, as many hacker culture devotees were familiar with the Jargon File (Raymond, 2012), a highly popular glossary of hacker terms, many of them modeled their production on its standards. On the following days Sanger and other editors tried to correct and direct the new editors, and fix the myriad articles they started, as many of their edits contradicted Wikipedia's policies and traditions. To sustain this effort, Sanger began doing an overhaul of Wikipedia's conscription devices, the core network of pages directed at informing participants of Wikipedia's policies and standards, such as the homepage, Welcome, Newcomers (Wikipedia, 2001by), the FAQ (Wikipedia, 2001bz), Wikipedia Policy (Wikipedia, 2001ca) etc. The wording, layout and interlinking in the various pages were standardized and improved, enabling quicker orientation and allowing them to function as resources for veteran participants in their attempts to discipline new ones. This process also resulted in the emergence of what Wikipedia is not (Wikipedia, 2001cb) as an independent page and a general guide pointing out to various types of unwanted content, generally accepted as an authoritative document and a privileged passage point in the project. The "list of things that encyclopedia articles are not" was a readymade tool for an orientation of new editors on content policies, and was quickly adapted to answer this need. The trigger for making "the list of things that encyclopedia articles are not" into a Wikipedia policy was an article named w00t (Wikipedia, 2001cy). This article, describing a geek slang term for happiness or excitement associated primarily with young gamers, was itself laden with geek terms. On July 28, Sanger started w00t/Talk page where he posted the following remark: Among the things that Wikipedia is not178 is a usage guide. Much less is it a usage guide for teenage slang--allowing that is to permit Wikipedia to degenerate into something resembling H2G2 and Everything2, which, as Zeus is my witness, I will never allow to happen. If you want to write entries about that, please start a dictionary wiki. We still don't know why no one has started one. (Wikipedia, 2001cz) Similar his post on the Nazi Germany/Talk page, this remark presented the list of things encyclopedia articles are not as a list of things Wikipedia is not. However, 178 Link to Larry Sanger/List-o-links (Wikipedia, 2001co). 187 the fact that this comment was directed at editors who were new to the project made the linking to list-o-links problematic. The list was titled of Things That Encyclopedia Articles Are Not and moreover was at the bottom of a column discussing a very specific issue, so new editors may very well misunderstand Sanger's intention. Thus, two minutes after posting his comment, Sanger moved the list to a new page, replacing it with a link to it. This new page, What Wikipedia is not, was not a subpage of Sanger's user page but an independent policy page, and similar to neutral point of view or Wikipedia is not a dictionary. After pasting the list of Things That Encyclopedia Articles Are Not to it, Sanger began editing and supplementing it for its new role as a rapid orientation guide for new editors. Sanger made both stylistic and substantive changes. Edits of the first type made the policy more general and removed the former personal style. Thus, the description of the page, appearing above the list itself was changed from the colloquial "I just felt inspired to make a list" to "Wikipedia is an encyclopedia. Here is a list of Things That Encyclopedia Articles Are Not". The postscript was also translated from first to third person, editing the sentence "I will continue to add to this list etc." to "We should." Finally, Sanger added a link to Wikipedia is not a dictionary on the first clause, dealing with dictionary definitions, not needed in the former version directed at veteran editors who presumably were familiar with it. The substantive changes to the list included the addition of two new clauses; both of them developed the themes raised by Sanger in his comment on w00t/Talk page, making the page a post-factum substantiation of Sanger's position. The first, inserted after the first two clauses dealing with dictionary definitions, stated that Wikipedia articles are not "a usage guide": Wikipedia is not in the business of saying how idioms, etc., are used. (But, of course, it's often very, very important ''in the context of an encyclopedia article'' to say just how a word is used. E.g., the article on freedom will, if doesn't already, have a long discussion about this.) The second clause, added at the end of the list, stated that Wikipedia articles are not "Everything2 or H2G2 nodes": Wikipedia is not a discussion forum or chat room. (But you can chat with folks on their own pages, and you can resolve article problems on the relevant /Talk pages). 188 After Sanger finished creating and modifying the page, he immediately proceeded to change the link on the w00t/Talk so it will direct to it, and then added a link to it from the Welcome, newcomers page. Three minutes later, he put a link to it on the Wikipedia policy page, describing it as "an attempt to say what Wikipedia is, by saying what it isn't". The link was placed on the second row of the page, a most prominent place – second only to the links pertaining to page creation (i.e. Naming conventions and Topic creation). Finally, he redirected the link on Wikipedia is not a dictionary from list-o-link to the new page, and deleted the list itself from the former. After he finished promoting what Wikipedia is not to the core network of policy pages, Sanger continued editing it and adding new categories to it. His additions continued to reflect his encounters with edits – and editors – he objected. The first addition, inserted on the same day, stated that Wikipedia articles are not "Personal essays, that state your idiosyncratic opinions about a topic": We're reporting on what is in the canon of human knowledge; unless you're unusual, your idiosyncratic opinions aren't part of this canon. (But you can put your essays in Wikipedia commentary.) This clause reflects Sanger fear of new editors will employ such practices – a fear rooted in the fact that he already found such essay (Wikipedia, 2001da, titled Two Main Theories of History) the day before and moved it to Wikipedia commentary (Wikipedia, 2001dc). Three days later, on July 31, Sanger added another clause. Right after the one stating that Wikipedia isn't a usage guide, he added a similar one, targeting directly to the issue of hacker slang. By a simple extension of the latter, a hacker/computer usage guide. We aren't teaching people how to talk like a hacker; we're writing an encyclopedia. (But see jargon file; also, articles, even extremely in-depth articles, on hacker culture are very welcome, ''and'' insofar as guides to some particularly essential piece of hacker slang is necessary to understand those articles, of course articles on that slang would be great to have.) This addition reflected Sanger's growing concern over the proliferation of such articles. The trigger for its inception was probably an article he edited immediately before, titled Yet Another (Wikipedia, 2001dd), defined as "A hacker meme, the use of 'Yet Another' as a way of padding out their acronym is fairly common", but it was 189 only one of many such articles Sanger came across and edited around this time. This clause, clearly directed at the new Slashdot editors, was subsequently edited by one of the most productive and vocal newcomers known only as the Cunctator on November 6th (Wikipedia, 2001de) to make it, as he put it, "a little more general". In order to achieve this goal, he expanded it to refer not only to "a hacker/computer usage guide" but also to "other slang and idiom guide", and added a second example, that of Cockney chimney-sweep to the statement "We aren't teaching people how to talk like a hacker". Over the following months the policy page slowly changed and evolved. After Sanger left the project in the beginning of 2002, other editors took on his role as an articulator of policies, and began updating the policy pages. New types of content were gradually added to what Wikipedia is not, usually after some discussion between participants, as new forms of content surfaced as problematic and were de-legitimized as anti encyclopedic179. The page's layout was also changed, highlighting the most frequently used clauses in the introduction and indicating these clauses that merited close attention by new participants. During these and subsequent changes, the core of the list remained very close to the way it was when Sanger first formulated it180, as very few existing clauses were modified, and they usually remained at the top of the list, ensuring their visibility. Though the page had undergone considerable changes in the last decade, these policies still stand at its core. The use of what Wikipedia is not as a guide to what Wikipedia is allowed Wikipedia's participants to avoid a positive, unequivocal definition of the project they were engaged in, even though its general parameters remained relatively stable. This route had some very positive unintended consequences: it enhanced its flexibility and gave allowed it to experiment and change over time. As there was no clear definition of what knowledge is desirable in Wikipedia, contents were not selected beforehand on the basis of its adherence to an established model. Rather, they were first added to Wikipedia and only then, if participants saw it as problematic, was brought up to 179 For example, Mere collections of public domain source material was added by Mav (Daniel Mayer) on February 25th 2002, news report was added on August 9th by Stephen Gilbert, a vehicle for advertising was added by DW on September 23rd, and original research added by AxelBoldt on January 26th, 2003. See Wikipedia (2012h) 180 Perhaps surprisingly, though the page had undergone considerable changes, these core policies are still visible today. 190 notice, discussed and transmuted into a policy decision. And even then, the fragmented nature of what Wikipedia is not allowed for changes and variation much more than a positive definition would. While creating standards and practices that shape the bulk of Wikipedia's articles and the "rates and gates" of its gift-economy, it allows Wikipedia to be inclusive enough to sustain immediate, individual giving which does not demand anticipatory socialization. Moreover, such definition would almost necessarily demand some sort of theorization or a symbolic universe to justify it, and therefore limit its flexibility and the ability of participants with divergent visions of the project to cooperate. It is precisely this under-theorized, even untheorized nature which underlies Wikipedia's strength as a global micro-structure. 6.4 Conclusion The events described in this chapter set the tone for dealing with content conflicts in the project over the following years. As new forms of content appeared in the project (and were then reintroduced numerous times by new individualized users unaware of the project's policy) the core network of policy pages and the conscription devices (Henderson, 1991) and scopic media (Knorr-Cetina, 2005) built to debate, inform and enforce their content proved vital in maintaining coordination between various individuals and groups. While policies changed and evolved considerably over time, those described here not only remained essential to the project, but also set the precedents that later policy formation initiatives would follow. This episode's implications are thus apparent in Wikipedia till this very day. It generated patterns of action for dealing with problematic editing and resources to be used in them. It shaped Wikipedia's volatile but effective authority structure based primarily on effective consensus and ongoing discussions, allowing it to accommodate new participants and their visions. In modeling the form of Wikipedia's content on a symptomatic model of the traditional encyclopedias, the policies described above created a stable but flexible standard that contents could be tested against, but also allowed for variations, especially on the issue of legitimate article topics. This standard allowed Wikipedia it to successfully strive towards practical legitimacy: a balance acceptable for both participants and readers that constructed 191 their expectations in a way favorable to the project, based on practices rather than theorization. In terms of governance, the pattern of action developed in this episode became standard method for content disputes: problematic contents were usually identified and debated in the periphery of the network initially, in one or few article pages; references, quotations and links were then used to connect them to the core network of policy pages181, where they were further articulated and contrasted with the encyclopedic model established beforehand; finally, an effective consensus would emerge as to the contents' legitimacy and a policy would be declared and implemented through conscription devices or software features. Within this pattern of action, Wikipedia's policy pages became privileged passage points in the project: not completely fixed, but generally stable; not absolutely compelling, but generally effective. Giving the community of editors a concrete, familiar and justifiable model that its editors could usually "grasp intuitively", as Wales (Wales, 2001j) claimed, the reliance on the traditional model of the encyclopedia helped reach decisions in cases of contested edits without reducing the conflict to one of status, seniority or personal preference. Policy pages – and specifically what Wikipedia is not – function as resources, mediators for the encyclopedic form, translating it into a series of relatively stable concrete standards. While what Wikipedia is not was periodically changed to reflect changes in the project's effective consensus, its main clauses were already institutionalized. This kind of ongoing but limited re-articulation was made possible by virtue of the policy page's original structure: failing to produce a positive form of the encyclopedia, it traced it contours through a series of binary oppositions constructed against other cultural forms. This allowed the project to balance between the imposition of form (cf. Bourdieu, 1991, PP. 137-159) and its innovation. The procedures whose development was described in this chapter enabled participants to coordinate their actions in a highly unstable and fragmented social environment. As literature shows, scientific knowledge production projects usually 181 Scopic media in Wikipedia changed over time, including at various times mailing lists, other Wikis such as meatballwiki.org and meta-Wiki and IRC channels. However, the policy pages remained the final arena where the policies were set. 192 achieve a high degree stability, whether by virtue of norms (Merton, 1973; Gieryn, 1982a), a common epistemic culture (Knorr-Cetina, 1999), a work of translation and the production of OPP's (Callon, 1986; Latour, 1988) or shared boundary objects and standard methods (Star and Griesemer, 1989; Bowker and Star, 1999, Star, 2010). Wikipedia's turbulent social environment precluded the viability of a stable, unchallenged consensus or hegemony determining the project's aims and methods. Wikipedia kept re-articulating itself as new editors continually joined the project, each bringing with him a somewhat different form of the encyclopedic model and a different vision for the project. This constant destabilization, however, was contained: the standard methods devised in this chapter generally succeeded in incorporating such visions (while excluding others) into the project. This flexible and reflexive form of coordination marks a distinct mode of translation, different from the one performed in scientific arenas, which is based on larger flexibility and adjustable standard methods. The extant and variations of this mode of translation merits further study, not only in collaborative knowledge production projects in the strict sense, but also in phenomena such as the contemporary global protest movement that seems to employ similar methods. I would like to extend the story to the content policies prevalent in Wikipedia today, even if only by touching their surface. The policies described in this chapter and the previous one were intensively changed and edited, but their essence remained very similar to the content described above. What Wikipedia is not policy remained prominent in the project, and new clauses were periodically added to it. The neutral point of view policy, on the other hand, evolved over time into three "core content policies"182 (Wikipedia, 2012k). First and foremost, the Neutral Point of View policy, dictating that Wikipedia articles should represent their subjects "fairly, proportionately, and as far as possible without bias, all significant views that have been published by reliable sources" (Wikipedia, 2012f). Second, the Verifiability policy, dictating that the threshold for inclusion in Wikipedia is "verifiability, not truth" – that is, "[i]t must be possible to attribute all information in Wikipedia to reliable, published sources that are appropriate for the content in question" (Wikipedia, 2012i, bold in the original). Finally, the No Original Research policy 182 While other content policies exert significant influence, most notably the "what Wikipedia is not" policy discussed above, those policies are considered the most important and general. 193 dictates that "Wikipedia articles must not contain original research. The term "original research" (OR) is used on Wikipedia to refer to material—such as facts, allegations, and ideas—for which no reliable, published sources exist" (Wikipedia, 2012j, bold in the original). These policies are presented as complementary, and are tightly interlinked, both through multiple occurrences in their text and through their existence as a distinct, formally acknowledged category of policies regulating the type and quality of Wikipedia article edits (Wikipedia, 2012k). While still lacking an articulated theoretical underpinning, these policies represent a coherent epistemological strategy present in embryonic state in the policies described above183. In contrast with modern encyclopedias, Wikipedia does not claim to the truth about the world, but rather the truth about human culture, or the opinions prevalent in it about the world. It does not try to function as an arbitrator regarding truth, but rather trace the networks of agreement and controversies and document the power-relations between various cultural standpoints regarding the truth. Wikipedia thus splits the world in to two parts: the world of things, inherently inaccessible within the confines of Wikipedia (as it does not allow original research), and the world of texts, which is Wikipedia's domain. Encyclopedic knowledge, according to this view, is only legitimate when applied to the world of texts, which is accessible through Wikipedia's only legitimate bearer of authority: the textual reference through which one may achieve a textual truth. This standpoint is very close to some trends in postmodernism which uncover the dominance of signifier over the signified in contemporary culture. However, Wikipedia does not break away with the modern culture: expert systems – a cornerstone of modernity according to Giddens (1990) – and especially academic ones are given a monopoly over knowledge of the world of things. Wikipedia's standards construct academic review and acknowledgement as the main forms of legitimation for information: Wikipedia's guidelines regarding hierarchy of information sources state that "academic and peer-reviewed publications, scholarly monographs, and textbooks are usually the most reliable sources" (Wikipedia, 2012l). However, that authority is limited to the world of things, and is denied in the context of the world of texts, as expertise remains unacknowledged inside Wikipedia's 183 This interpretation is solely my own, and is not based on an articulation present in Wikipedia itself. 194 production system. In this system, every editor is considered equal to all others in terms of their critical capacities and their ability to assess or generate knowledge and information, and real life identity is not recognized as a bearer of legitimacy. In this, Wikipedia peculiarly retraces the footsteps of early encyclopedists, those who were involved with the manuscript culture (and later, with early print culture) of the high middle ages and the renaissance. These encyclopedists regarded authoritative texts as auctores which confer the necessary legitimacy on their compositions, rather than their immediate knowledge or the counseling of contemporary experts. As Ribemont (1996) argues, the style of medieval encyclopedias, "is that of assertion, in a discourse in the present tense, made up of a juxtaposition of elements presented in an objectivity justified by the authority [i.e. auctoritas, L.G.] from which encyclopedic thought derives a large part of its origins". Wikipedia's commitment to the traditional form of the encyclopedia also shaped Wikipedia's consumption, as it enabled it to achieve a substantive enough measure of legitimacy to attract readers. While Wikipedia's credibility and legitimacy are still lively debated, I would propose that relying on the model of the encyclopedia allowed it to achieve practical legitimacy – an attitude of relative trust that takes into account and encapsulates its usability, its dangers and its lack of formal legitimacy. As Wikipedia's production processes were incommensurable with the legitimating discourse of expertise which supported traditional encyclopedic projects, it had to develop alternative means to demonstrate its legitimacy, both internally and externally. Wikipedia's reliance on this model and its success in enlisting participants to sustain it generally succeeded in filtering out various types of problematic content that would have made it less likely that the reader finds what he was expecting for. It helped distinguish it among other internet information resources that were less organized and uniform, and convey a measure of trust and credibility as it shaped editors and readers' expectations. Paradoxically, it was the very commitment to the traditional encyclopedic form that allowed Wikipedia to achieve its status as an innovative and dominant reference work. 195 At the same time, the relative disregard for certain aspects of the traditional encyclopedia184 (such as the reliance on a predetermined category scheme; the hierarchical arrangement of topics; the attempt to strike a balance between different subjects in terms of size; the emphasis on stable or static types of knowledge, the western curriculum and high-brow subjects), enabled Wikipedia to be more relevant to web surfers than any other contemporary encyclopedia. The eagerness to engage in articles relating to contemporary news and popular cultural events (in the broadest sense of the term) appealed to a large crowd interested in a relatively dry encyclopedia style articles on such subjects. Other innovative features, such as the existence of talk pages; a full and accessible archive of articles' former versions; the inclusion of bibliographies and external links sections and the use of standardized warning labels in cases of disputed content, also increased its usability – all without threatening the traditional form of the encyclopedia article and the project's practical legitimacy. Wikipedia's usability and the notion of practical legitimacy in general, will be the subject of the next chapter. While in the last chapters I described the production of Wikipedia's general model and some of its main standards, in the following chapter I examine how they affect Wikipedia's audience. Models and standards, by their natures, mold the cultural product in their shape. They are inscribed in them. It is now common wisdom in the sociology of culture and in cultural studies that the interpretation of cultural products should be supplemented and triangulated by studying the way they are constructed by their consumers. Thus, in the following chapter I examine the way a specific group of highly information oriented workers – namely journalists – use Wikipedia, and how their construction of its form and legitimacy are embedded in the patterns of its consumption. 184 such as the reliance on a predetermined category scheme; the hierarchical arrangement of topics; the attempt to strike a balance between different subjects in terms of size; the emphasis on stable or static types of knowledge, the western curriculum and high-brow subjects. 196 Chapter 7: Wikipedia in the Field of Israeli journalism "Wikipedia is like porn – no one talks about it, but everybody does it" (Eyal, Freelance journalist, previously deputy editor of a weekend supplement) In the previous chapters, I described the production of Wikipedia, and examined the practical ways in which participants tried to connect and articulate the Wiki technology and non-professional, volunteer-based editing with the traditional model of the encyclopedia that would buttress its legitimacy. The results of these efforts are not unambiguous: as noted in the introduction, Wikipedia is one of the most popular websites in the world, and by far the most popular reference work in existence, but its legitimacy is still controverted (Reagle, 2010a, 137-168). In this chapter, I shall demonstrate how this attempt failed and succeeded at the same time, as Wikipedia developed into a universally used but unacknowledged resource. I shall try to better understand this strange, paradoxical pattern by examining the pragmatics of its reception and consumption in one context specifically, that of journalism. While some work have been done which attempts to describe how users asses the credibility of Wikipedia’s articles, no studies had been made about the practical, everyday use of Wikipedia by experts. Both Metzger et al. (2010) and Yaari et al (2011) studied under-graduate students’ use of Wikipedia, and Yaari et al. used an experimental setting where participants were asked to assess specific articles chosen by the authors. A similar method was employed by Chesney (2006), who asked experts to assess the credibility of articles either in their area of expertise or outside it, as in both conditions the articles were chosen by the researcher, and there was no attempt to understand the process of assessment in a natural setting. To understand Wikipedia's success and its effects on society, it is crucial to examine how it is used in practice, in the everyday life of its users and not in a controlled experimental setting. Moreover, focusing on students and academics use of Wikipedia may give a distorted view of how Wikipedia articles' credibility is assessed, and the meaning accorded to the information appearing in it by its users. 197 Thus, in line with this work's pragmatist approach, I decided to study how specific knowledge experts – namely journalists – use it in the course of their work. Weinberger (2011) noted that expertise itself is changed in several significant ways once it is embedded in digital networks. First, it seizes to be topic based, and defies conventional disciplinary boundaries. Second, where its value used to lie in the certainty of its conclusions, it now lies in its ability to open various direction and opportunities unthought-of of by layman. Third, expert's knowledge used to be opaque – it was a stopping point beyond which the knowledge seeker could go no further; now it is networked – that is, includes links to various sources which allow the consumer to contextualize it. Four, divorced from its traditional print medium, expertise is now interactive, allowing for more interactive forms of knowledge seeking. Five, expertise are no longer a distinct class – "on On the Net, everyone is potentially an expert in something" (p. 67). Finally, while expertise used to demand speaking in a single voice, current consistency is now polyphonic. The original idea for starting interviewing journalists was that they are a population uniquely posited for examining the relation between the use of Wikipedia, the way in which it is perceived, and conceptions of knowledge and knowledgegiving in general. Journalism is a field where general knowledge is a crucial, necessary tool of the trade, and information is the goal and prize, making Wikipedia an especially valuable resource for professionals in that field. At the same time, the cautious and suspicious ethos towards sources of information, and the extreme visibility of journalists' products, might make them wary of using it. This tension, I hypothesized, would enable me to examine the connection between perceptions and evaluations of Wikipedia, and its actual use. Specifically, I expected to find some sort of correlation between journalists' use of Wikipedia, their readiness to quote or refer readers to it, and their perception of it. Another pattern I was expecting was a variation between the perception and practical utilization of Wikipedia by journalists working in print and online venues. However, when I began interviewing journalists about the way they use Wikipedia, I quickly realized my assumptions were wrong. While the interviewees indeed demonstrated a variety of attitudes towards Wikipedia, ranging from a rooted distrust of the website to much admiration for its reliability and usability, their use of 198 Wikipedia seemed to be almost completely unrelated to their perception of it. The journalists I interviewed reported that their use of Wikipedia is not influenced by the fact that knowledge is given to it by unpaid contributors, or by the fact that anyone can edit it. In general, they dissociate their consumption of Wikipedia from their views of its production. This dissociation is directly connected to the way Wikipedia's role in the world of journalism was institutionalized185: all the journalists I interviewed, without exception use Wikipedia regularly, usually on a daily basis and almost never less than a few times per week. Moreover, all interviewees reported that this practice was habitual in the venues they worked in However, while the use of Wikipedia in journalism seems to be a universal practice, leaving evidence of having made use of it – such as reproducing text passages from it, citing it or acknowledging using it in articles – is a taboo, strictly kept even by those who personally advocate for its use. These seemingly paradoxical findings – a universal and routine utilization coupled with the absence, even denial of public acknowledgment – will serve as the backbone of this chapter. In its first part, I detail these findings, and the structural and cultural conditions underlying them. In the second part, I portray the practices and patterns of operation entailed in journalists' professional use of Wikipedia, involving an apparently smooth and effortless movement between contexts and websites that hides a range of valuations and decisions. I also specifically address the importance of paratextual, para-informational and intertextual elements in this process. Paratext, a term coined by Genette (1997), relates to "verbal or other productions, such as the author's name, a title, a preface, illustrations" that enfold the text and shape its interpretation by the reader. Similarly, para-information relates to verbal or other productions that shape it interpretation by the reader. These obviously include the paratext, but also textual elements such as style, wording, references, equivocations, etc, and intertextual elements imbuing it with the legitimacy of other texts. In the context discussed here, I demonstrate the importance of such elements for journalists using Wikipedia in what is a fast-pace, highly time bound pattern of operation, and the peculiar interchangeability of texts and paratexts in its confines, which blurs the information's source and renders it ambiguous and contingent. Another issue I address 185 At least in the Israeli context. There are indications that this is also true for journalism in the United States, at least (Shaw, 2008). 199 in this section is the importance of scopic elements, which supply readers with a view from afar of texts and information encapsulated in summaries, links, etc, thus enabling them to avoid actually accessing those texts. Before concluding, I stop to discuss issues that may limit the findings of this study, arising from the fact that my interviewees are all Israelis working in Hebrew written languages. In the final section, I try to explain the apparent contradictions discovered in this study, and asses the perception of knowledge encapsulated in journalists use and attitudes. 7.1 Wikipedia in the Journalistic Environment From the interviews conducted, it appears that Wikipedia had become an intrinsic part of the professional toolkit of journalists, and is seamlessly integrated into their work process. All the journalists interviewed, without exception, stated that they use Wikipedia regularly in the course of their everyday work. In addition, interviewees reported that using Wikipedia is a standard procedure used by others in their field. For example, when asked how often Wikipedia is used in the newsroom of his print newspaper, David, a news editor in a daily paper responded All the time. Literally all the time. Every evening, everyone. Reporters, Editors, proofreaders… it surprised me when I first found out that Wikipedia is a routine work tool for what is called fact checkers (English in the original, L.G), that are supposed to be the authority. They're supposed to check the facts. And usually they check them using Wikipedia. Moreover, there are some indications that Wikipedia is being institutionalized and using it is becoming a part of the socialization process of journalists. Matan, senior editor of the foreign new section in a print venue, stated that he actively encourages his subordinates to use it on certain occasions: I try very hard to infuse the practice of using it in my department. For example, the other day a translator made an error in translating a name […] so in such cases I always say, if you have a name to translate, always check it up in Google, in Wikipedia. Often we get translated names, and it's not sure what the source is, I always refer whoever is doing it to Wikipedia, look it up, it has a pronunciation spelling […] The frequency of use differed between interviewees, as the most avid users confessed consulting it more than once an hour on a workday, while the most sporadic 200 users admitted using it at least once or twice a week. Fifteen of the twenty interviewees stated they use it at least once in every workday. While the relatively small number of journalists interviewed (20) precludes any quantification, it seems that the frequency of using Wikipedia is generally correlated with the frequency of article submissions, and the distribution of use is generally correlated with the temporal demands of deadlines. For example, Tal, a proofreader in a weekly culture supplement, said she mostly uses Wikipedia on Mondays, the day the supplement is "closed" (i.e. is ready to be published): "Monday, which is the day '7 nights' is closed, it's a very long day, from the morning to the middle of the night, I'm with Wikipedia all the time." As a means to assess the extent to which Wikipedia is woven into the everyday practices of journalists, interviewees were asked how their work would be affected if Wikipedia ceased to exist. All agreed that while it won't be an insurmountable obstacle, it may seriously slow down their work pace – a state of affairs which in some cases might be extremely problematic. Thus, the extent to which that option was perceived as inhibiting was related to the perceived importance of the fast paced rhythm of their work. David, for example, who is responsible for the news section meeting the publishing deadline, thought that it may have severe consequences: Q. How do you imagine your work would be like without Wikipedia? A. Nasty, brutish, and short (laughter). How would I imagine my life without Wikipedia? More or less like living without Google. It would force me… it would expand the workload enormously. I imagine I wouldn't know how to manage today without Google and Wikipedia. I mean, I would manage eventually, I'd get used to it, like I've gotten used to Wikipedia, there's nothing holy in it in that sense. It just makes the work so much quicker and easier. But a life without Wikipedia… (groans) On the other hand, Noa, an editor in the high-tech news section of a financial, took a similar but much more lenient position: A. Look, there's always Google. If you know how to search with Google, you'll find what you want. But Wikipedia makes it easier. To be honest, it would slow down my pace. Q. Slow down your pace? That's all? Because two other journalists I interviewed… 201 A. Said they couldn't work without it? Oh, no. I don't work like that. It's only… always the starting point, but never the end point. Really, only in extreme cases of like, the editor is sitting here (points to her neck, L.G.), and it's quarter past four (very near the close of the newspaper, L.G.) or something like that that could be the end point. But I’ll try to avoid it. And deny it later on. These quotes uncover some of the most recurrent themes journalists addressed while talking about Wikipedia: its incorporation into the journalistic work process, its connection with issues of speed and time pressure, its inextricable ties to Google, the objection to using it as an exclusive single source of information – and the denial that so often accompanies its use. As the quote equating Wikipedia to porn with which I started this chapter indicates, the use of Wikipedia is as embarrassing as it is widespread. While Wikipedia was presented by all interviewees as vital to contemporary journalistic work (if to a varying degrees), it was just as strongly presented as illegitimate, as it must never, as a rule, be named as a source for information appearing in the article. As Noa noted, "You can't quote Wikipedia. You just don't do that. We're not really big on citing sources anyway, but quoting Wikipedia is forbidden. You don't name it as a source. I remember I searched for some data on something once, and they [Wilkipedia, L.G.] brought an assemblage of data that was exactly what I needed, and I didn't see any problem with using it as a source because I said… it's all taken from sources and websites we work with and know, and we would use them as a source, so it's silly like, in this case not to use it as a source, but no. there's a taboo against it. Q. Taboo? A. Forbidden. Q. Is it something writ… talked about? A. Not written, nothing is written. Q. Talked about? A. Yes. Forbidden. Big no no [English in the original, L.G.]. Similarly, leaving in the text of the article evidence that Wikipedia was used is considered an indication of bad journalism. As Matan said, 202 Sadly, I often see leftovers of Wikipedia texts in articles. And I always preach my reporters to be extremely careful with it. The idea is to read, understand the story and articulate it in your own words, not to copy materials and get into this kind of stuff. But in this age, where journalists, all over the world, have to do, to work much more for the same money, or even less, you see a lot of cutting corners and fudging. People who don't have time, who are stressed, who need to produce results quickly, forgo going in depth. These taboos uncover a deep ambivalence towards Wikipedia: it is almost as universally suspected as it is used, as almost all the interviewees raised concerns regarding Wikipedia's reliability and credibility. There was a consensus that Wikipedia includes at least some errors and omissions186, and that using it should be done with caution. Moreover, there was general agreement that Wikipedia must not be used as a definite single source and that information originating from it should be triangulated with other resources. How can this paradox be explained? How can journalists hail the use of Wikipedia as a resource and denounce it at the same time? How can they incorporate it into their professional toolbox and at the same time demand that its use be hidden? A simple answer would be that they are hypocrites: they use Wikipedia as a means to make their works easier even though they know it is wrong, and thus they have to hide it. However, this interpretation, with all its plausibility, falls short of explaining the fact that the interviewees did not see themselves as lazy or unprofessional, and neither did they see the use of Wikipedia as a breach of their professional “epistemic virtue” (Daston and Galison, 2007). Rather, Wikipedia was constructed as a legitimate and highly useful tool – as long as it is used appropriately and correctly. As a way to solve this seeming paradox, I suggest that at least in the context of journalistic work, Wikipedia enjoys a practical legitimacy (as distinct from a theoretical legitimacy, or one immersed in a symbolic universe. See Berger and Luckmann, 1967, p. 95), embedded in a pattern of action that takes into account and encapsulates its usability, its dangers and its lack of formal legitimacy. As a means of demonstrating this claim, I will now describe an ideal typical pattern of action reconstructed using the interviewees’ description of their use of Wikipedia, using an ethnomethodological approach that requires a description of the minute practices journalists use when 186 However, when asked about problems or mistakes that they ran across, none of the interviewees could remember concrete examples. 203 searching for information in the course of their work. While these details may seem mundane, it is precisely only through such details that Wikipedia's specific mode of practical legitimacy can be uncovered. 7.2 Wikipedia in Journalistic Practice Almost without exception, interviewees reported a similar pattern of action that occurred when they consulted Wikipedia during the course of their work. This pattern, included the following stages, and can be schematized as the following: Problem  Google search  Consulting Wikipedia  Triangulation  Rearticulation The fact that this pattern of action may seem familiar or even obvious to the reader, is an indication of the extent to which it is incorporated with our own everyday lives, as well as those of journalists If the valuations and practical choices made during information seeking on the internet appear smooth and effortless it is because they are incorporated into an institutionalized pattern of action, a routine, where the methods of knowledge acquisition itself become habitual. However, as I'll try to demonstrate, this seamless veneer hides a plethora of supporting conditions and complex considerations worn smooth by habit. Before describing this pattern, however, I would like to quickly survey the structural conditions of the journalistic workflow which support and sustain it, namely its spatial and temporal organization. The first important aspect regarding this pattern of action is that Wikipedia's seamless integration with the journalistic workflow is made possible by the physical organization of the journalistic work. Knowledge acquisition, especially when conducted through the act of reading, is often thought of as somewhat of an abstract, cognitive process, happening between the reader and the text. This is even more so when addressing interaction with the internet, still very often thought of as a nonplace, an un-physical and in some senses anti-physical cyberspace. However, the physical setting is of the utmost importance when attempting to place knowledge acquisition within a sequence of action. This assertion is quite obvious in the case discussed in this chapter: journalists relations with Wikipedia are first and foremost 204 determined by the fact that the bulk of their actual work is done in front of a computer connected to the internet. The work of journalists is very often a desk job, especially in the process of writing the final product. Information acquisition, communication with peers and the actual production and editing of articles are done on personal computers, usually in the offices of newspapers. A second important aspect conditioning journalists' use of Wikipedia is the temporal arrangement of the journalistic workflow. Journalists' work is highly timebound, in both print and internet venues. In traditional newspapers, the workflow is bound by the pressure of deadlines. As the quotes above demonstrate, the actual works practices of journalists are highly dependent on the time pressure they work in, as closer deadlines tends to result in more condensed work and shorter chains of action (i.e. "cutting corners"). The workflow in internet venues is just as time-bound, as that of print venues, if not more. While internet venues don't have deadlines in the traditional sense, they are bound by the competition with other websites and the need to stay in pace with the news. As Alon, a sports section reporter, noted, "When an editor tells a reporter to go make an obituary, the first thing he does is to open Wikipedia. No doubt. However, it's always important to triangulate it with another source. Well, it's problematic with internet venues, because it’s a medium that is… immediate. The competition between venues is, people are unaware of it, only those inside it, but the competition between internet venues, Ynet, Walla, NRG, Ha'aretz… it's mostly over speed. Who goes up first with an item." These spatial and temporal conditions make Wikipedia a highly valuable asset for journalists, and facilitate its incorporation into their workflow. In what follows, I'll outline the typical stages of the chain of action that take place when journalists use Wikipedia. 7.2.1 Problem As the journalist is writing or editing an article, she encounters a Problem in the work process, manifested as a certain piece of information necessary for the article that she is not confident about. According to the interviews, this insecurity is never about the main storyline of the article – as building an article on Wikipedia was 205 unanimously declared as bad journalism, but rather about some information needed to support it. The usual causes for this insecurity are either the novelty of a piece of information, the information, a memory lapse, or a precaution. While deep background knowledge needed to understand a subject in general is considered a very positive thing, the scope of information journalists usually cite as reasons for consulting Wikipedia, at least in this most dominant pattern of action, is of a much more limited and specific nature, such as a chronology of a rolling event, a quick overview of a person, a company, a technological or cultural product, etc. Indeed, all interviewees stressed that the most prevalent reason for using Wikipedia was to find some concrete and minute details, or "mini-facts". Notable details which systematically reoccurred in interviews were the spelling of persons' and places' names, dates (and especially dates of birth and deaths), sums of money (such as revenues of culture commodities, financial worth of firms), etc. The nature of relevant information often changed according to the journalists' specialty. For example, Tal stated that "anything (in an article, L.G) I don't recognize, I check with Google and Wikipedia. Sometimes I use IMDB (Internet Movie Data Base, L.G.) too. Names of songs, movies, books, people…years of publication too. I say 'wait a minute', and go look it up". On the other hand, Alon stated that he uses Wikipedia "For getting details on obscure foreign football players who come to play here (in the Israeli football league, L.G.)… mostly things that have to do with foreign events, like, if I write on something I generally know about but need to fill in the blanks, let's say I'm writing on, the African football cup, and I'm writing about a player who scored a goal, and I'm not sure which club he plays for right now, I'll search for it. […] or when there's some Bosnian midfielder that I don't know how to spell his name. It's a great tool for the sports section, especially for details you're not sure about, or don't know. Ages, heights and the likes." The use of Wikipedia in cases of memory lapses is also usually connected to this kind of "mini-facts". As Or, a newsroom editor in a news website noted: Sometimes I just forget something, like a name, or… just the other day, I wrote an article about [names an Israeli politician], and I just couldn't remember what was his former position in the IDF. I knew I knew it, but I just couldn't remember. So I found his page in Wikipedia, and it was there, and I went, "yeah, that's what it was ". 206 Q. why didn't you just ask someone? A. it was just easier. And quicker too. Alt-Tab, Ctrl-T, Ctrl-E, write his name and press enter. And click on the link. It takes two seconds, and I know I'll get the answer. Also, I didn't have to bother anyone. Another contingency of a similar kind, and closely related to memory lapses, was usually designated by the term "to make sure." That is, the problem in this case wasn't a lack of information at all, but rather an insecurity regarding its accuracy. Events such as this highlight Wikipedia role not only as a source of information, but also as a mnemonic instrument. Information isn't a dichotomous variable you either have or you don't. There are also fuzzy middle positions, where one is not sure of one's knowledge. In these cases, perhaps even more than in the former, Wikipedia's accessibility and rapid response makes it a ready extension of the journalist's cognition, mobilized quickly and effortlessly as he sits and works on a computer connected to the internet and equipped with an open browser. 7.2.2 Google Search As evident from the quotes above, the result of such insecurity is an internet search – that is, almost without exception, a Google search. Only two of the interviewees searched for information using the search function embedded in Wikipedia. In that respect, Google takes precedence over Wikipedia in the journalistic workflow187. This is also evident from the fact that some of the journalists admitted that when they encounter the information they need on the preview presented on Google results page, they will suffice with it. In any case, all interviewees observed that almost in all their searches, a page from Wikipedia will appear as the first result, or very close to it – and that they expect to find it when searching. As noted above, the use of Google is inseparable from the physical and temporal work conditions of journalists. Several of the older interviewees, versed in older ways of information retrieval, such as the venue's internal archive, information specialists or purchased databases, contrasted the use of Google to such methods, 187 This is also evident from the fact that some of them stated that they sometimes use results from other relevant websites, such as the internet movie database (IMDB.com). 207 stressing their slowness or physical remoteness. As Tzvi, a domestic news reporter remarked, "we used to use LexisNexis, all sorts of information databases, which are in fact much more reliable, but we're living in a different age, much quicker, more instant [English in the original, L.G.], much easier. Who needs LexisNexis? It still exists, these premium databases that hardly anyone uses anymore. It's a shame, don't get me wrong, but it's just too slow and complicated. You need a degree to get what you want from it. Google is just so much quicker, who can resist it?" Similarly, Eyal, an editor in a major website and formerly a reporter in a daily newspaper, claimed that "The old procedure was to check the newspaper archive. Yediot scanned all their archive, so you can search it directly from the computer. So did Ha'aretz. But Maariv didn't have the money, so they still use envelopes. They are in the basement, and you have to ask someone to go and get them, they are catalogued in books, and you have these levers and pulleys to get them. It's just like modern times, this story. And the more esoteric the subject, the worse it is covered. It's just impractical for most purposes." Iddo, who worked in a venue with a digitized archive, didn't have a much better view of it, noting that "If I look for a subject in the archive, let's say the Iranian nuclear program. You have a page on the subject, a tag. Let's say I'm looking for something very specific, the archive can have ten articles just from the previous month…it's going to take a lot of time to find what I want. In Wikipedia, because it's a relatively short encyclopedic article… it's faster to look in it, and get from there to the original news article in the original website if I need to." The combination of Google and Wikipedia thus replaces, at least partially, older ways of finding information. These findings raise the fact that Wikipedia, from the point of view of a journalist seeking information, is practically inseparable from Google – as Amir, a former sports reporter and current culture section editor of a major website noted, "Wikipedia, for me, is another section in Google." And as Eyal stated, "I wouldn't even say 'Wikipedia', you can't disconnect Wikipedia from Google in this respect. The placing Google gives it, and the fact that it presents it alongside the most authoritative resources makes it what it is.… so I use Google, in the first stage. It is my regular starting point from which I enter Wikipedia. The only starting point, absolutely." As this quote illustrates, Wikipedia's usability and legitimacy is 208 inseparable from the ranking it receives from Google, and its dominant status in the realm of online information retrieval is inseparable from Google's monopoly in that realm, as it bestows Wikipedia with its dominant status as a source of knowledge. This legitimacy, it needs be stressed, is articulated in a practical, rather than formal or epistemological terms. It is rooted in the socialization and previous experiences of the journalist, who learns over time that Google brings her quicker and better results than alternatives, considering the constraints of her time and work frame. The fact that Wikipedia articles continuously occupy the top positions in Google pages, and are usually found useful to journalists, frames and shapes the way it is perceived by them. Google thus functions as a sort of paratextual or intertextual element, shaping Wikipedia's reception and determining the way it is received and read. However, the reverse is also true up to a point, as Wikipedia also shapes journalists' expectations of Google. As noted above, journalists often search Google with a clear intention to find a Wikipedia result. Thus, the relation between Google and Wikipedia seems to be more symbiotic than parasitical, as Google benefits from their expectation to find results from Wikipedia. In a very important sense, Wikipedia transcends Google's framing, as is evident from the fact that most of the journalists stated they will click on a link to a Wikipedia article when one shows up in the results page even if other websites rank higher than it. Moreover, five of the interviewees stated that they occasionally add the terms "Wikipedia" or "Wiki" to the search form to ensure they get a result from Wikipedia. 7.2.3 Consulting and Evaluating Wikipedia Articles While in some cases journalists extract the information they need from the preview presented on a Google results page, in most cases they will click on the link leading to the relevant Wikipedia article. Following links from Google to specific Wikipedia articles has an important role in shaping journalists’ perception of Wikipedia. According to the interviews, journalists consulting Wikipedia during the course of their work usually encounter just the single page linked to from Google. 209 They seldom, if ever, frequent its homepage188. They hardly ever use such navigational aids as portals and category schemes in the course of their work, and neither do they consult help or community pages. Furthermore, while twelve of the twenty interviewees were aware of the existence of History and talk pages, only one of them actually stated they use them on a regular basis. As evident from the previous section, they almost always enter a specific article Wikipedia from a Google link that functions as its paratext, frames it and bestows it with legitimacy. Thus, Wikipedia as a totality, a website with an internal structure and readily available, transparent sustaining mechanisms, is kept near-invisible, in this habitual pattern of action. Wikipedia's reception is thus, to an extent, fragmented. While texts appearing in the Encyclopaedia Britannica, Associated Press or Bloomberg News are more or less uniformly endowed with a legitimacy stemming from their venue of publication, the fact that a certain page appears in Wikipedia frames its reception to a much smaller extent. This fragmentation is inseparable from the heterogeneity of Wikipedia's articles themselves. While all journalists noted that they regarded Wikipedia as a source of information superior to most other websites, they also, without exceptions, noted the great differences in quality to be found among articles. As Orly, a proofreader in a daily newspaper noted: There are huge differences between articles, really very huge. There are articles that look amazing, incredible. Well structured, with a good introduction, with these little squares on the side that give you all the information you need (i.e. infoboxes, L.G.), good links... Other articles look like they were written by a 12 years old kid. And in some cases, they're so short you can't get anything at all out of them. This heterogeneity may partly explain the attitude journalists exhibit towards the information they find in Wikipedia (to be discussed at greater length below), as they usually stress the fact that they judge it on an ad hoc, "case-by-case" basis. What this actually means is that the main measure for the reliability of a looked-for detail is the perceived reliability of the article page it appears in. Thus, while journalists are searching for information on an article page, they are also forced to make some evaluation of its quality. This process is so habitual that it is done almost unconsciously, and in any case it is not temporarily differentiated 188 One major exception, the use of the "on this day…" section, will be discussed below. 210 from the search itself, allowing it to smoothly blend into the pattern of action. When asked to explain how they assess the reliability of information in Wikipedia, journalists named several elements, all highly visible, appearing within the confines of the accessed page and external to the information itself. While in cases of memory lapse it is possible to assess the validity of information based on recollection, in the case of new information the evaluation process is assisted, to a large extent, by paratextual and par-informational elements. In the absence of such traditional elements such as a respectable publisher, an author's name or a leather binding, journalists have to draw on signs appearing on the web pages themselves to evaluate specific Wikipedia articles. In a paradoxical dialectic similar to the kind outlined by Goffman (1971), the need to assess the accuracy of the information they seek forces journalists to focus more on the signs external to it, and at least partly irrelevant to it. Such signs can be generally divided into two categories: framing elements and hypertextual elements. Framing elements are signs journalists use to evaluate the information they seek, they are those visible elements on the page which indicate to the reader the amount of trust that should be accorded to the article can be trusted. These elements can be broadly divided into paratextual and textual elements. Paratextual elements, in this context, are usually devices and trappings of traditional print works translated into Wikipedia's digital realm, such as infoboxes, headlines, illustrations, footnotes and a table of contents189. Textual elements are aspects of the text itself, such as a good introduction, correct grammar, length etc. These are para-informational but not para-textual elements, as they are incorporated in the text of the article itself. The elements noted above all have two important characteristics in common: first, they are highly visible and can be identified with a quick glance that does not have to be differentiated from the action of seeking the relevant information on the page; second, they serve as a measure for the amount of attention and work invested in the article. Transcending the basic work needed to generate a simple presentation of information, these elaborate elements are important exactly because they are peripheral to the information itself. They indicate that enough extra work was done on 189 The fact that these paratextual elements have that effect may lead one to speculate about the legitimating effect print culture still enjoys in digitally mediated texts. However, most journalists noted that it is rather the feeling that time and effort has been invested in the article that makes them trust it. 211 the article that its basic elements were already taken care of.190As Ronen noted, "a good article is always written better articulated. There's no mess. The text is orderly, you have references, footnotes, the text is usually well written, I mean, it gives you an impression that…you know, it’s the difference between three and three hundred people creating it". Another important type of paratextual elements occupying a unique position in comparison to the ones noted above, in that it diminishes rather than enhances trust, are warning templates, noted by more than half of the interviewees as modifying their judgment. Such templates are prefabricated standard comments and warnings that can be conveniently pasted by Wikipedia's editors to an article as indications of problematic or insufficient information, in the form of standardized warning templates. Over the years, many such templates were developed to indicate certain problems in an article as a whole, or in certain parts of it. Notable among those are the stub template, indicating a very short and insufficient article, disputed content template indicating that the factual accuracy of an article, a section or a sentence is disputed, NPOV template indicating the neutrality of an article, a section or a sentence is disputed, and reference necessary (or as it is more often referred to, citation needed) template, indicating that information appearing in the article is unreferenced. Such elements differ from the ones described above as they are at once an indication of investment (of the editor placing the waning) and lack of it (as it warns against a text that may otherwise seem perfectly sound). All interviewees addressing it noted that they are more suspicious of information flagged with such templates. A second type of signs used to evaluate the quality of article pages are intertextual elements. Intertextual elements that connect Wikipedia pages to external textual resources usually appear in the form of hyperlinks to external websites but can also manifest themselves as references to offline sources. Such elements have a unique position in journalists' pattern of action and valuation process. The majority of the journalists interviewed stated that external links are the most important measure for the reliability of a Wikipedia page. Such elements include references, footnotes, 190 An important paratextual element occupying a unique position in this respect is the "citation needed" template, noted by almost half the interviewees as modifying their judgment. This element is at once an indication of investment and lack of it (as it warns against a text that may otherwise seem perfectly sound). All interviewees addressing it noted that they are more suspicious of information flagged with this template. 212 bibliographies, "further reading" sections, and most importantly, "external links" sections, etc. These elements are obviously also a part of the article's framing, as they are evidence of the work invested in it. However, they have a more important and more radical meaning that justifies placing them in a category of their own. More than just an indication to measure the work invested in an article, journalists' reliance on hypertextual elements is usually shaped by their confidence that they are lead to the sources from which the information in the article was produced. As Noa noted, "Generally, I asses the reliability of an article just like I would that of any other text. What sources do they use, like…some time they have references there, so if they rely on obscure stuff, like blogs, for example, I won’t perceive it as very reliable. When they rely on newspapers I consider reliable, that’s another thing." Hypertextual elements thus function as a unique kind of sign, imbuing the article with the legitimacy attached to the source they lead to. Similar to the relation between Wikipedia and Google, the relations between Wikipedia and the sources it points to are at least partially reciprocal, as Wikipedia also shapes journalists' expectations of its external resources. As I show below, journalists often incorporate Wikipedia into their pattern of action as a path or proxy for its external links, which they use as the actual source of information. Thus, Wikipedia often functions within this pattern of action as an intertextual element shaping the reception of the sources it links to (who are the actual text they seek), just as Google functions as an intertextual element for Wikipedia. Wikipedia's status is thus ambiguous: it is sometimes text, sometime paratext, sometimes hypertext and sometimes hypotext, sometimes the means and some time the end. Google, the Wikipedia article and the external links cited in it are thus all entangled in this pattern of action, framing each other and reinforcing each other's legitimacy and usability. Within this chain, Wikipedia functions as a synecdoche, a blackbox encapsulating all other objects, representing them and focusing their respective characteristics191. The 191 This entanglement, it should be stressed, is not present in the text itself, or "objective" in any way, as it is constructed through this specific pattern of action and absent in other ones. However, while this is beyond the scope of the current study, I would like to suggest that the possible existence of chains positing similar relations will tend to strengthen this entanglement. Google engineers, Wikipedia editors and owners of websites linked to Wikipedia seem to be aware of the synergetic connections between the objects they nurture, and actively foster it. An interesting indication supporting this hypothesis came from Alon, who commented that his editor in the culture section of a news website occasionally tries to add links in Wikipedia articles to relevant projects in the website. 213 decision when to stop searching – whether to use the information from the preview in Google's results page, the Wikipedia page, or the sources Wikipedia links to – is contingent on the concrete conditions of the search and it is what ultimately stabilizes the relative status of the various objects in the given case. After locating the relevant information and assessing its reliability based on the signs surrounding it, the pattern of action surveyed here forks, as journalists face an alternative: whether to use the information they found on Wikipedia immediately, or to try validating it using triangulation. In the following section I'll describe journalists' generic triangulation practices when using information from Wikipedia. 7.2.4 Triangulation While in many cases the search for information ends when it is located in the Wikipedia page, sixteen of the interviewees stressed, with varying degrees of passion, that they usually do not content themselves with this information, and attempt to verify it through triangulation with other sources. In this narrative, Wikipedia is constructed, to cite Amir, as "just another source, one of the many that I routinely use." While triangulation is constructed as an integral part of the journalist's epistemic virtue, and as especially important in the case of Wikipedia, the actual use of this practice is a modular part of the discussed pattern of action, as its existence is contingent upon the concrete circumstances of the search and in particular, time pressure. Moreover, the concrete actions involved in it confound, to a certain extent, the usual meaning of the term triangulation, as they usually stay within the confines of the series of objects discussed above. The process of triangulation is constructed by journalists as an essential feature of the journalistic epistemic virtue (Daston and Galison, 2007) – that is, it is essential to their construction as professionals. Regardless of the extent of their actual engagement in this practice, all interviewees associated it with proficient and serious journalism, and the use of Wikipedia without triangulating was often presented as an example of the superficiality and lack of professionalism in contemporary journalism. In some cases, this association was used to contrast between the interviewees' virtues, 214 and the prevalent vices she is surrounded by. For example, Yael, vice editor of a weekly culture and entertainment supplement, stated that "Whenever I use Wikipedia – and I do it all the time, obviously – but it's always in parallel with other sources. If it is to get facts, which is the most critical thing in my view, it was never the only source. I always cross-referenced it with other places, two or three places. It was always to make sure. But that's also because I'm oldfashioned… old-fashioned, and also, I say this objectively, more thorough." However, even while hailing the virtues of triangulation, only five interviewees insisted that they never use information from Wikipedia without triangulating it. Most journalists held a more flexible position, stressing the importance of triangulation in principle while admitting that in some cases they avoid doing so, mainly due to time pressure. As Amir noted, I never… not never, but almost never finish my research with Wikipedia. Wikipedia is not a dead end (English in the original), the end of the road in terms of an investigation. It's a launching pad perhaps, or a junction from which I can go somewhere. Q. don't you have cases where you say, "actually, that's ok, that's enough for me? A. no, there are cases; it depends on the level of… how deep I need to go. To be honest, it depends on how much time I have." As noted above, the constraints of time pressure and deadlines are essential to the work of most journalists. Under these circumstances, it is often difficult or impractical to triangulate information. Thus, triangulation functions as a justification for action as much as it does as a pattern of action. The ideal pattern of action including Wikipedia as well as its triangulation was essential to many interviewees construction of it as a legitimate tool in journalism, even when they themselves admitted they don't always do so. This gap between justification and action is especially important when addressing the concrete ways in which journalists actually triangulate information, as the actual enactment of triangulation is done in a way that often confounds the usual meaning of the word. When asked to specify what practices they use when triangulating the information they got from Wikipedia, interviewees usually referred to one of three 215 options: using Wikipedia pages in other languages, using other Google results or trusted websites, or using external links from the Wikipedia article. All these sources are pragmatically interlinked, as they are – some to larger extent than others – part of the same environment and setting as Wikipedia, demanding only a handful of mouse gestures or keyboard typing to arrive at. While triangulation is often constructed as a digression from one context of action to another, in our case it is more an extension of a preexisting context. To triangulate information using these sources, all a journalist has to do is move back or forward in our familiar pattern of action, sidestepping and then backtracking to the routine course with minimum effort. This may explain why the most popular means of triangulation, by far, was the use of external links from the Wikipedia article, an action almost all the interviewees declared they engage in regularly. The construction of this action as triangulation is problematic, as the consulted sources are clearly not independent. Obviously, the links placed in Wikipedia article will, as a rule, support the text and information appearing in the article itself. This is a direct result of Wikipedia's policies and practices, which stipulate that claims should be backed by external resources (see chapter 3). Relying on an alternative description appearing in the interviewees, this action may be better understood as the tracing of a chain of references (Latour, 1999), a continuation of the pattern of action starting with Google, continuing in Wikipedia and ending in the websites it links to. This chain of references represents information in various levels of processing and proximity that the journalist can decide to follow through – in principle, indefinitely, if she decides to use the links or references in the websites that Wikipedia links to – until she is satisfied with its measure of validity. Obviously, as she faces a deadline the question of the information's "ultimate source" usually remains beyond the scope of the journalist's actual actions. What she requires is a practical means to make sure that she is not totally confounded – that is, that a source is credible enough so that she may be sure that the chain leading from it onwards is unbroken, or at least that such hypothesis in that regard is plausible and can later be justified192. The assurance that such a source exits somewhere down the 192 This of course echoes Berger and Luckmann's discussion of plausibility structures, the "social base for the particular suspension of doubt without which the definition of reality in question cannot be maintained in consciousness" (Berger and Luckmann, 1967, p. 175), and their reliance on a generally 216 chain of references is the foundational hypothesis that allows journalists to decide, at any given level, that she can suffice with the measure of proximity she achieved, but it is not necessary, and usually not practical, to actually access it. The decision when to stop along the chain of references is mainly determined by the contingencies of time, already addressed above. In the competitive world of internet venues, as well as the strict deadline regime in newspapers, information delivered too late is of no use. Thus, journalists have to use signs and indices to replace the actual reading of the material present on the next link in the chain of texts (which is not actually read) as means to decide when to stop the chain of action. Such signs are scopic elements that enable the journalist to peek into the next link on the chain, helping them to determine the necessity of actually engaging it beforehand. Wikipedia articles are usually rich with such elements, in the form of references and external links and their description in the text. As noted above, these elements are essential to Wikipedia's ability to become a trusted and privileged passage point. This quality is enhanced by Wikipedia's hierarchy of resources, which shows a preference towards conservative and respectable sources. Thus, in many of the interviews appeared an alternative definition of Wikipedia, which constructed it not as an encyclopedia, but rather as an index. Iddo, for example, when equating Wikipedia to newspaper archives, noted that "In Wikipedia, because it's a relatively short encyclopedic article, you get only the basic information. And it's easier to use, because it's really a sort of index, it's faster to look in it, and get from there to the original news article in the original website if I need to. Wikipedia is a lot of people going over a lot of information, summarizing it and giving you references." These features do more than make Wikipedia a privileged passage point: under conditions of time pressure, they also make Wikipedia increasingly a privileged end point, allowing journalists to stop the chain of references being relatively reassured about its following links. The richness and adequacy of scopic elements in Wikipedia articles very often allows the journalist to determine that it is trustworthy enough, and if the article itself looks reliable he can neglect actually checking the next accessible typology of experts which legitimize them, while their own expertise remains inaccessible (ibid, pp. 95-96). See below in chapter 8. 217 level. As such episodes reoccur, Wikipedia turns more and more into a privileged end point in the chain. As David noted, Wikipedia created, at least for me, eh… some sort of authority, Otorita, in a sense that using it so often in a sense dulled the... the tendency to check, I mean the more you use it the more you become entangled in it, institutionalized into the Wikipedia, it increasingly becomes your main tool for work. The duality of Wikipedia, existing both as a privileged passage point and as an end point in the search of information, is vital to its functioning in the journalistic workflow. As discussed above, Wikipedia is illegitimate as a final or single source of information. Thus, its usefulness and legitimacy as a privileged passage point is vital to its slow and gradual construction as a privileged – though illegitimate – end point, and enable journalists to use it while denying its status as an actual source of information. However, this situation also means that journalists who use Wikipedia as a source of information cannot publicly acknowledge its use, and sometimes have to actively cover their track. Indeed, such actions are an integral part of the next and final stage in the pattern of action – the re-articulation of the information gathered and its incorporation into the article. 7.2.5 Re-Articulation After the journalist returns from her journey, she now has to translate the information gathered to the context of the article she works on. The main actions journalists engage in at this stage are the re-articulation and translation of the information to fit it to the new context, and the erasure of any traces that can lead the readers back to Wikipedia as a source of information. The re-articulation of information is considered an integral part of the process, vital to the epistemic virtue of the journalist. The pattern described above, it should be remembered, is embedded in a longer and more elaborate pattern of action – that of writing an article. In this last stage of the pattern, the problem described in the first phase of the pattern is overcome, and the lack of confidence regarding the information needed to complete the article is overcome. The journalist, now in firm possession of the information, 218 goes back to her word processor, where the article draft is located, and inserts the information into the article. The extent of re-articulation needed is dependent, for the most part, on the size of information: The smaller it is, the closer it is to a "mini-fact", the less adaptation it needs. Information like spelling of names, or dates, need no adaptation at all, and is usually imported using a "copy-paste" method, while larger chunks of information probably warrant a larger measure of adaptation to the concrete narrative of the article. In any case, the journalist attempts to strip all the parainformational elements still present in the original text as a part of this process. The erasure of parainformational elements has both practical and emblematic reasons: from the practical point of view, old para-informational elements need to make room to those specifically relevant to the article. However, the emblematic point of view is just as important, as it severs the link between the new article and the chain of objects implicated in the journalist's search – and especially Wikipedia. As noted above, all the interviewees, without exception, reported that the text of completed articles should be purged from any sign that the information contained in it originated from Wikipedia. This imperative is obviously relevant mainly to sizable and composite chunks of information (manifested as larger blocks of text), as details such as spelling, dates and amounts are almost pure information and warrants little translation, if any. In addition, such details, in contrast with larger blocks of text, can rarely function as traces leading with any kind of certainty back to the original source in Wikipedia. The dual nature of re-articulation enables journalists to engage in it without feeling they are cheating, or performing a professional vice. Where credit has to be given, when an important part of the story is arrived to through Wikipedia, it is not withheld, at least not in principle. It is simply transposed, moved from Wikipedia to the next link in the chain of texts, just as it is transposed from Google. And when no credit is due, the severing of the information's link to Wikipedia is folded inside the more obvious task of writing the article in the journalist’s own words. In this sense, copying from Wikipedia is not very different from copying from any other source – except that neglecting to do the necessary work of “translation”, or transposition may have more embarrassing repercussions, as Wikipedia's style is easy to trace and equate. In this last stage of the pattern, than, the formal illegitimacy of Wikipedia as a 219 source of information is neutralized by incorporating it and negating it into the very pattern of action through which it is used. The epistemic virtue of the journalist using Wikipedia remains intact precisely because her ideal use of it is indistinguishable from the act of erasing the traces of using it – while she justifies that action in terms seemingly unrelated to it. While this action may be seen as hypocritical, it is indistinguishable from the general practices of journalism. 7.3 Using Israeli Journalists as Interviewees As noted in the methodology chapter, my interviewees were all Israelis working in Hebrew written venues. This may limit the generalizability of the findings presented above, both with regards to other fields and cultures, and to other language versions of Wikipedia. Quick generalization of the practices presented here to other institutional and cultural contexts will be irresponsible: they may be the result of a local epistemic culture, different from ones to be found in other context. As Nicholas Johns' (2011a, 2011b) studies about the diffusion of the internet to Israel during the eighties and early nineties suggest, the institutionalization of new media is highly dependent on individual and state level machinations. This means that further research is needed to understand the differences between the ways journalists in different countries use Wikipedia. However, I would like to suggest that a situation similar to that described above may exist in other professional fields – namely, that of higher education. From a few exploratory interviews, numerous discussions with fellow academics, and indeed from my own experience, it seems that very similar patterns of use may be found in this field too – more so with regard to teaching, but to some extent also with regard to research and writing. In any case, further studies are needed to assess the extent to which Wikipedia is used in various fields, and how patterns of action which incorporate it differ between fields. A second problem may steam from the fact that the most relevant version of Wikipedia for the needs of my interviewees was the Hebrew one, while this study's assessment of Wikipedia is mainly based on the dominant English language version. This problem is compounded by the fact that Wikipedias in different languages do not 220 comply to the same polices (with a few exceptions, including the Neutral point of view policy, considered binding in all Wikipedias), and there is a large degree of variance in the way they operate. However all interviewees, without exception, noted that they frequently use the English Wikipedia, some of them more often than they did the Hebrew one. Their reports of their experience with both language versions indicate that they consider them a part of a whole, alternating between them according to their needs and the quality of the information found in them. While they do differentiate between them, journalists perform identical patterns of use and evaluation in both Wikipedias, and more often than not use the English version as a higher authority than the Hebrew one. Thus, I believe that their reports may be used to construct a pattern of action generalizable to audiences of various language versions of Wikipedia. This belief is supported by the fact that several of my interviewees who used Wikipedias in various other languages – such as German, Russian, and French – reported performing identical patterns of action and evaluation when using these versions. 7.4 Conclusion This chapter started with a puzzle: how is it that the use of Wikipedia among journalists is widespread, habitual and institutionalized, but so is the taboo against citing it or leaving traces leading to it in an article? How come journalists acknowledge this taboo and usually view it as justified, while at the same time they deny that using Wikipedia is an offense against epistemic virtue? I attempted to demonstrate how these seemingly contradicting attitudes are possible by describing the actual pattern of action journalists use while consulting Wikipedia. This pattern of action includes, in each of its stages and in their cumulative existence as a pattern, various elements that enable journalists to use Wikipedia while denying it a status of a source of information warranting a reference, and to facilitate the reconstruction of the para-information engulfing it. The problem that drives journalists to consult Wikipedia is constructed as a lack of confidence regarding pieces of information that are peripheral to the main story of the article, are usually rather small, and often either previously known but 221 forgotten or in need of verification. The search for the information is usually done using Google and not Wikipedia itself, constructing the former as a more critical link on the way to the information. This construction then is strengthened by the fact that journalists sometimes use other resources appearing in the Google results page, and on other occasions make do with the information appearing on the results page itself. When journalists actually use the Wikipedia page, they do not take the information presented in it at face value, but rather use various signs to verify its quality, making their expertise at assessing the information an integral part of the process. Moreover, Wikipedia is often constructed as a sort of index or threshold leading to information whose "actual" location can be traced through the external links appearing in the article. Following these links to their origin, or using some other method of triangulation is constructed as obligatory in the process, thus making Wikipedia – in principle, if not in practice – just one of several sources and defusing the responsibility over the information. Finally, the actual utilization of the information includes a transposition of credit from Wikipedia to a text further down the chain, and a process of purging the information from exterior textual elements and paraphrasing it, which is universally acknowledged as a legitimate journalistic practice. While those elements all facilitate the ability to use Wikipedia and at the same time deny its role as a source for the article, the most important element facilitating such a pattern probably resides in the relation between these stages, in the very structure of this pattern of action. As demonstrated above, the journalist using this pattern traces a chunk of information through a number of interlinked texts (corresponding to the stages of the pattern), searching for a source that he considers good enough to substantiate those preceding it or for a reference193 that points to the existence of such a text even if it is not actually accessed. In this pattern, Wikipedia serves as a privileged passage point – not obligatory, but highly useful and usable – summarizing the content of such texts, and even more importantly, giving trustworthy links pointing to such texts, and making superfluous the need to actually access them, making it thus often – and increasingly - the eligible, privileged end point of the search. The information found in it is thus perceived as simultaneously present in all the texts that lie beyond it, whether they are actually accessed in during the enactment of the pattern or not. It remains “floating,” as it were, between various texts, either 193 Thus functioning as a scopic tool or paratext. 222 one of which can be arbitrarily named as its source, enabling the use of Wikipedia as a source of information while denying its status as such. This pattern bears an interesting resemblance to Weinberger's (2011, ch. 4) notion of networked expertise – the term he uses to explain how expert knowledge – and indeed, expertise itself – is changed once it is embedded in digital networks. A central difference between networked expertise and traditional expertise lies in the fact that expert's knowledge used to be opaque – it was a stopping point beyond which the knowledge seeker could go no further; networked expertise, however, includes links to various sources which allow the consumer to contextualize it. Moreover, divorced from its traditional print medium, expertise is now interactive, allowing for more interactive forms of knowledge seeking. While Weinberger's argument is quite compelling, it lacks examples and evidence. Journalists' use of Wikipedia and the notion of privileged passage point both serve as concrete manifestations of this concept. The fact the last part of the habitual process for using Wikipedia includes an erasure of any evidence to its existence indicates that the process in its entirety encapsulates both the practical utility and the threat Wikipedia poses to the status of journalistic practice. The taboo on referencing to it or leaving parts of its text in journalistic articles is incorporated into the very same pattern of action that organizes its use, and is a condition for its successful performance as a practice. This pattern thus incorporates both the seamless integration of Wikipedia search in the journalistic practice, and the prevalent suspicions regarding its use, thanks to the multiplicity of information sources and the floating and inherently unfixed nature of that information. As this description demonstrates, the action of using Wikipedia for professional purposes in journalism is infused with the tension arising from the universality of the taboo against citing or acknowledging its use. However, it is important to stress that while the interviewees all acknowledged – and by and large approved – the taboo against leaving evidence to the use of Wikipedia in the text, they did not see their use of Wikipedia as unprofessional or unethical. Wikipedia thus enjoys a status of practical legitimacy within the field of journalism: it is embedded in 223 a routinized and institutionalized pattern of action that takes into account and encapsulates its usability, its dangers and its lack of formal legitimacy. This type of legitimacy allows journalists to use Wikipedia as a crucial tool for locating information, and at the same time retain their sense of epistemic virtue. However, it must be remembered that practical legitimacy, by its very nature, presumes a lack of formal legitimacy. This ambivalence is often present in the way interviewees addressed it. While most of them viewed it as inherently positive, it is nevertheless constructed as problematic. As Iddo, an internet and technology reporter in a major website noted, The problem, one of the problems with Wikipedia is that it has no substitute. Its accessibility…its availability, its ease of use, and the fact that it has most of the information you need, and the fact that it has no substitute makes it a thing journalists are bound to use. The construction of Wikipedia as a "problem" in a quote hailing its advantages is highly symptomatic in this regard. This attitude makes Wikipedia's usability an open secret; one that great caution is needed in repeating or publicizing. Returning to the quotation with which I opened this chapter, this type of practical legitimacy leaves Wikipedia in the socially ambivalent position that porn is located in: highly popular, at times – mostly in private – discussed in positive terms, but always somewhat obscene, never achieving the status of formal legitimacy. This attitude may of course change over time – and my personal impression is that it is changing - but paradoxically, a very important factor in such a change, i.e. the support of the press, is confounded by the very fact that journalists use Wikipedia so very often, and yet cannot openly admit it. The following years may be a crucial time to research the development of this type of legitimacy, as it may evolve into formal legitimacy, remain as it is, or acquire new forms and articulation. 224 Chapter 8: Conclusion and Discussion This work delineated the ways through which Wikipedia cemented its position and practices of knowledge production over time, and how it is used by one class of professionals, journalists. In this concluding chapter, I sum up the main arguments and draw out their theoretical significance. The first part recapitulates the main steps whereby, a peculiar epistemic culture which "only works in practice" evolved in Wikipedia. In the second part, I elaborate on two theoretical concepts which I found helpful in accounting for this culture and its development, namely the idea of privileged passage points and that of practical legitimacy. Finally, I try to assess: to what extent is Wikipedia symptomatic of the current emergence of a new, perhaps global epistemic culture? While previous explanations of Wikipedia's success were predominantly ahistorical, this work traced its origins in locally situated processes which led to the emergence of Wikipedia's peculiar “unstructured structure.” As this study demonstrated, to understand Wikipedia's success before it became a dominant actor in its field, we need to look at the intertwining of diverse networks. The coordination and enlistment of these networks constructed Wikipedia's articles as effective boundary objects, mediating between different actors and social worlds (Star and Griesemer, 1989; Bowker and Star, 1999; Star, 2010). Through negotiation and translation, Wikipedia's founders articulated some minimal standards crucial for Wikipedia’s operation and development, and constructed mechanisms that allowed for its relatively smooth coordination in the future. While often described in hindsight as predetermined and shaped by broader cultural and technological phenomena, such as its underlying technology (Bruns, 2008; Tapscott and Williams, 2006), the open source tradition (Mulgan, Salem and Steinberg, 2005) or contemporary "cult of the amateur" (Keen, 2007), Wikipedia was shown to be, in fact, the contingent product of negotiations and interactions which gradually amassed local solutions to concrete problems, translated them into culturally justifiable and technologically feasible patterns of action, and covered up their own traces in the end product. 225 One of the main goals of this work was precisely to recover these traces. While such Reconstruction is a common goal in STS, this work offers a unique focus on an arena that lies outside the more usual purview of science and technology studies, and extends its implications to a situation in which central actors are cultural models, rather than physical objects and/or persons. Wikipedia incorporated several historical epistemic models, such as the Socratic dialogue, the encyclopedia and the philosophy of Enlightenment, just as it did the innovative Wiki technology, the FOSS movement ideological creed and groups of computer programmers and internet enthusiasts. They all combined in shaping Wikipedia – and were, in turn, shaped by it. Cultural models are extremely evasive entities: they often operate in an unarticulated manner, operating through the practices of actors with various previously unrelated interests. They are not "norms" which govern behavior, but rather assemblages of meanings and practices that other actors can mobilize to forge stronger networks. They function as relatively fixed entities, which help orchestrate more mobile elements around them. However, as the case of Nupedia demonstrated, relying too heavily on established cultural models may prevent innovative knowledge production projects from acquiring the flexibility necessary to succeed. Nupedia, Wikipedia's failed predecessor adhered to culturally sanctioned models of knowledge production: its form and contents were based on the Britannica's print encyclopedia, its production process and gift-repertoire on academic practices, and its governance on the authority of experts with academic credentials. While it did include some innovative elements – such as a public review process for article content – those were subjected to the dominance of the principle of expertise. However, the project's slow rate of production triggered a movement that attempted to reshape the project in a more open and inclusive direction. In this attempt, the Britannica Online was used as a qualified object (Thévenot, 2001a), generating a standard for article quality. This move enlisted it to the project, and by ensuring that the quality of articles would remain high even if they were produced by non-experts, allowed for the incorporation of less orthodox methods. While these attempts did have some effect, they did not change Nupedia's basic problems: its inability to sustain a viable gift economy attractive to the experts 226 needed to maintain it. However, as this change proved insufficient in accelerating the rate of production, its founders decided to start a new project, more open than Nupedia could ever be. Thus emerged Wikipedia. Wikipedia stood, from the start, in a complex relation with Nupedia. In some ways, it was shaped as a complete opposite of its predecessor, relinquishing its models of knowledge production and suggesting a new one, based to a large extent on the practices developed in other Wiki communities which focused on knowledge production. It abandoned expertise as a model for knowledge production, replacing it with a model of radical participation and collaboration, in which everyone could write on any topic they felt they knew something about and each participant could alter, erase or comment on the work of any other participant. Nupedia's standards for articles and content policies were discarded, and it took some time and effort to form new ones. However, highly important schemas and resources originally developed in Nupedia were transferred to Wikipedia by participants in the former who also took part in the new venture. These included an insistence on a traditional encyclopedic image; a strong and elaborate stance on issues of bias in article content; key actors who were already committed to the idea of producing a FOSS encyclopedia, and funding from Bomis, Jimmy Wales' and Tim Shell's company. As soon became evident, controversies rising in the midst of the radically open and collaborative writing of article contents, reviewed in chapters 5 and 6, were endemic features of this production process. Wikipedia's founders and participants thus had to quickly develop conscription devices (Henderson, 1991) and scopic media (Knorr-Cetina, 2005) that could prevent or resolve such controversies. As these were the first controversies to emerge in the project, their resolution was accompanied by the institutionalization of elements that still sustain Wikipedia's standard practices, policies and governance mechanisms. Through this process of institutionalization, a number of key principles of coordination emerged: the primacy of the traditional encyclopedic image as a "standard for standards"; the legitimacy of the Wiki process as a complementary resource for schemas and practices; a proactive and improvisational attitude in matters of production and technology; the reliance on effective consensus between editors to establish policy; the dialogical nature of 227 knowledge construction; and the reliance on references to other texts rather than on expertise and original research to establish knowledge. The initial major controversies (reviewed in chapter 5) emerged from production problems stemming from the reliance on radical, transparent, collaboration. In the first of these controversies, the problem emerged from discussions between editors appearing in article pages. This practice, an integral part of the prevalent Wiki culture, stood in opposition to the traditional form of encyclopedic articles. Thus, participants had to decide which had primacy in the project. After an intense discussion, the controversy was forcefully resolved by the institutionalization of a new socially constructed category of "talk" pages. This resolution was later inscribed in the software of the website, solidifying it by translating it into the very technological fabric of the project. In the second major controversy, the problem of mediating conflicting opinions regarding article content was forcefully resolved by Wales's articulation of the “Neutral Point of View” policy, dictating that disputed content would be declared as such and that all sides of the disputes should be fairly represented. The resolution of these controversies formed a general pattern for disagreement resolution that was later repeated in other conflict-situations in Wikipedia: the creation of a solution for a particular problem; the generalization of this solution, orchestrated through policy pages or mailing lists; its effective translation to a legitimate practice through "form investment", to use Thévenot's (2001a) term, 194 in previously sanctioned modes of coordination (such as the traditional encyclopedic form, Wiki culture, etc.); and its technological or social inscription. These steps thus became what I termed "privileged passage points" (PPPs), coordinating action in most cases while allowing for some variations and modifications. While some of these PPPs later emerged as fully-fledged Obligatory Passage Points (Callon, 1986; Latour, 1988; Bijker and Law, 1992) – for example, the neutral point of view policy and the separation of article pages from talk pages – others remained a more voluntary or flexible mode of coordination. This semi-flexible 194 Form investment is the creation of equivalences across time and space between various practices, objects, and persons, in order to stabilize and anchor the participants' interpretation of an allegedly intuitive concept (Thévenot, 2001). 228 organization allowed Wikipedia to change and develop over time to react to new challenges and – at least until now – to face them successfully. Just as crucial as facilitating radically collaborative production, another key problem was the need to decide on the image of the encyclopedia, and set standards for article content. The translation of the encyclopedic model to Wikipedia (described in chapter 6), through interpretation and selection of relevant attributes, was a contentious process, intimately related to struggle over power and authority in the project. It also gave rise to some of Wikipedia's central governance mechanisms, still active at the present time. It culminated in the construction of effective mechanisms for creating, modifying and enforcing content policies and in the articulation of an effective and relevant encyclopedic model. This model relied on attributes of the traditional encyclopedic article form, but supplemented them with innovative elements. These included talk pages, the introduction of links and online references as means of backing any potentially controversial claim, the availability of previous versions of articles, and warnings in cases of bias, partial or unsourced information, or otherwise problematic content. These elements, all developed to assist Wikipedia's production, played a crucial part in enabling Wikipedia's growth and success, as they encouraged and helped casual readers in forming the stance needed to make it relevant to their various lifeworlds. Relying on the model of the encyclopedia, and supplementing it with features designed to help assess the quality of articles, allowed Wikipedia to achieve what I term practical legitimacy195 for its producers and consumers alike (see below). Chapter 7 sets to explore the everyday manifestations of this mode of consumption and usage through interviews with Israeli journalists about their use of Wikipedia. Based on these interviews, I constructed a model of journalistic use of Wikipedia comprised of five stages: Problem Google search Consulting Wikipedia Triangulation Re-articulation. In this process, the journalist moves from her immediate practical needs and environmental constraints following a chain of references (Latour, 1999) leading to a hypothesized origin of information, until she 195 Practical Legitimacy is a specific mode of action that is based on an attitude of relative trust that takes into account and encapsulates the action's utility, risk and lack of formal legitimacy (see more below). 229 reaches a point where following further references becomes impossible or unpractical. In this process, Wikipedia serves as a privileged passage point – a practical shortcut to knowledge that originated elsewhere. Over time, however, It gradually tends to develops into a practical end-point in this process, as users tend more and more to rely on the validity of the information it contains, based on evidence found in its articles that further valid references were involved in their compilation. As users increasingly tend to suffice with information found in Wikipedia articles, they are forced to perform their validation processes exclusively vis-à-vis those articles. Thus, consuming them increasingly involve judgmental and critical practices meant to establish the validity of the sought after information. This critical and selective mode of consumption is actively encouraged by elements appearing in Wikipedia's articles, which give readers indications of its quality. First, a range of framing elements – such as infoboxes, headlines, illustrations, footnotes and a table of contents – were found to help journalists estimate the amount of attention given to the article, which in turn serves as an estimate for its worth. Second, intertextual elements, most often in the form of hyperlinks to external websites,196 allow journalists to assess the sources through which the article was (allegedly) assembled, thus imbuing it with the legitimacy attached to the source they lead to. Finally, warning templates indicate certain problems in an article as a whole, or in certain parts of it.197 In addition, other elements of Wikipedia that appear outside the actual article-page encourage their evaluation by readers. These include mainly highly visible article histories (previous versions), which allow readers to view the actual composition of the article; and talk pages, which enable them to access previous discussions about the articles' content. The re-articulation of information obtained from Wikipedia brings up a curious finding: while consulting it emerged as a routine and universal feature of journalistic work processes, mentioning it as a source was a strict taboo. The 196 Such elements can be found in Wikipedia in the form of references, footnotes, bibliographies, "further reading" sections, and most importantly, "external links" sections. 197 Warning templates are indications of problematic or insufficient information, in the form of conspicuous standardized hypertextual elements and banners. Over the years, many such templates were developed to notable among those are the stub template, indicating a very short and insufficient article, disputed content template indicating that the factual accuracy of an article, a section or a sentence is disputed, NPOV template indicating the neutrality of an article, a section or a sentence is disputed, and reference necessary (or as it is more often referred to, citation needed) template, indicating that information appearing in the article is unreferenced. 230 journalist must cover up her use of Wikipedia as she re-articulates the information she obtained, attributing it (if necessary) to the farthest link in the chain – and in any case, not to Wikipedia. This attests to Wikipedia's limited success in endowing its model of knowledge production with universally acknowledged legitimacy. It cannot be considered legitimate as a whole, and each article has to be assured piecemeal and assessed individually, and its use has to be kept hidden. This implicit contract between producers and consumers – problematic and secretive as it is - stands at the heart of Wikipedia's success. 8.1 Theoretical Implications While the narrative I construct concerning the institutionalization of collaboration and production practices in Wikipedia and its use shed a pragmatist light on the emergence of one of the major cultural product in the beginning of the 21st century198 (as it accounts for Wikipedia's success by describing the practices performed by individuals and groups who partook in producing and using it), this work also aims to contribute to a better understanding of the formation of epistemic cultures more generally. Conceptually, I submit, the institutionalization and dissemination of Wikipedia's epistemic culture is better understood in light of two concepts: the notion of privileged passing points (PPP) and that of practical legitimacy. 8.1.1 The Strength of “Weak” Passage Points The notion of privileged passage point emerged when I tried to understand how standards and policies function in Wikipedia: they are clearly not Obligatory Passage Points (OPPs), in the sense given to this term by Actor-Network theorists 198 Pragmatist theory proved highly useful in explaining Wikipedia's temporal trajectory. As I show throughout this work, Wikipedia's history follows a very similar trail, by which the object of knowledge and the means to construct it are gradually articulated and co-constructed, and habit and creativity alternated as problems led to re-articulations of both the project and its goals. This approach is also highly relevant to understanding how Wikipedia is incorporated into knowledge production practices of users, as its use is conditioned upon and conditions the concrete environments in which it operates. Moreover, this view helps redirect the focus from Wikipedia's legitimacy as a knowledge production institution to the actual practices and procedures it generates and which attempt to construct it as legitimate. 231 (Callon, 1986; Latour, 1988, p. 43, 2005, p.31), as they can be changed, circumvented, or even, at times, ignored.199 In contrast with the construction of OPPs as a “single locus that could shape and mobilize the local network” and “have control over all transactions between the local and the global networks" (Bijker and Law, 1992, p. 31), Wikipedia policy pages (with few exceptions) were far less omnipotent: most of them could not shape and mobilize the entire local network (of Wikipedia editors, technological apparatus, etc.) as some actors did not enforce or even adhere to them. Likewise, these passage points do not control all transactions between local and global networks, as evident from the fact that problematic or controversial content was regularly added to the website.200 However, as demonstrated in this study, PPPs did cause a major realignment of the networks involved in Wikipedia, and quite effectively directed the actions of actors in the project. Moreover, I observed a recurrent pattern in which practices were transformed by PPPs, through routinization and institutionalization, into habits. Some habits were transformed into PPPs through the intensification of links to other practices and actors enlisted to sustain and disseminate them; some PPPs (such as the Neutral Point of View Policy) were transformed into OPPs, when they became highly effective and immutable parts of the project. PPPs, as I theorize them here, are somewhat closer to boundary objects (Star and Griesemer, 1989; Bowker and Star, 1999; Star, 2010) or conscription devices (Henderson, 1991), but differ from them in their limited impact on the arrangement of the arena. Both boundary objects and conscription devices, with all their flexibility, are means of generating a structured totality: a well-coordinated social arena, its different parts re-aligned to contribute to a joint task. PPPs do not go as far, since they afford a greater measure of variability and reversibility, allowing the social space to remain more fluid and dynamic. They themselves are not completely fixed, though generally stable; not absolutely compelling, but generally effective. Moreover, while boundary objects are, first and foremost, objects they form a narrower concept than 199 A major formal policy, named "Ignore all rules" (Wikipedia, 2012a) even encourages them to do so, under certain circumstances. 200 Though it is often erased quite quickly, it does exist on the website for long periods of time, at time because it remains unnoticed, but more often because its debated between editors with conflicting views about what should be done with it. 232 PPPs who may function as actors, actively reshaping and redefining the actions and identities of others. This difference became clearer when I started applying the concept, and soon realized its usefulness in understanding not only policy pages, but also the actions of other actors in the field: Wikipedia's founders, for example, who held (at different times) similar positions as leaders of the project in its institutionalization period. They too had a directing role, but were usually unable – and seldom tried – to forcefully dictate behavior against participants' will or to enforce their positions. Rather, they developed alliances with other actors that allowed them to steer certain aspects of the project against opposition. With the assistance of allies – some human and others nonhumans, some contingent and others more stable – they could win most major challenges, although they did lose some, and had to modify or give up their positions in specific cases I describe. These two instances are related: the flexibility and leniency of leadership (see Reagle, 2010a, chapter 6) is mirrored in the flexibility of policies.201 Thus, it may be claimed that the term PPP is relevant not only to this specific network, but to a wide range of networks that are more open or flexible than institutionalized ones built around scientific projects. Indeed, many virtual communities, and most of the global information networks Manuel Castells (1996, 1997, 1998) describes, seem to fall within these parameters. Checking the applicability of this hypothesis is one possible way to extend this work (see below). The usefulness of this conceptualization is apparent when we move from production to consumption – the analysis of the patterns of action enacted by journalists when searching for information online. In this context, Wikipedia appeared as itself a PPP: not obligatory, but highly useful and usable. Its privileged position stems from the fact that it helps selecting and articulating relevant chunks from the endless text fields of the internet. While it is seldom completely trusted, it quickly becomes an integral part of journalists' habitual tool-kit in attempting to look for information. Once using Wikipedia was transformed into a routine practice, distrust towards it remained mostly in the background, while in practice it significantly 201 However, the reverse is also true: Wikipedia's leadership patterns were shaped alongside the institutionalization of project policies, and in reaction to controversies about them. 233 reshaped journalistic information seeking practices – thus making it more than just another mere "resource" among others. I would like to propose that structural links exist between journalists’ use of Wikipedia, and its construction. Both the production of Wikipedia and the search for information online involve highly individualized trajectories and patterns of action in an extremely open space. While stabilization and closure are, to a large extent, the goals of technoscientific projects (Pinch and Bijker, 1987), it had already been noted that openness and instability are inherent both to the internet as a whole and to Wikis in particular (see above, chapter 1). Thus, there is no clear entrance and exit points and no fixed borders, trajectories are contingent upon the inclinations of divergent actors, and it itself can change significantly as a result of their actions. These features, embedded in technology, prevent any attempt to structure them completely. But, at the same time, they make the reduction of complexity and the creation of roadmaps and shortcuts highly effective for large numbers of actors, which in turn helps in enlisting them – although only partially. However, as Nupedia's example illustrates, closed and stable structures fare worse, not better, than open ones in enlisting allies in the context described here. They operate much better in stable social environments, where strong and steady ties can be formed and institutionalized and become integrated with their surrounding structures. More open environments, such as the internet, are more welcoming to more open and dynamic structures. Closed systems are slower in adapting to the quickly changing environment of the internet, as their OPPs limit their flexibility. They are less inclusive, and thus fare worse in attempting to mediate the needs, interests and attitudes of those multiple and shifting social worlds of which the internet is comprised. And finally, flexible systems fare better when insufficient legitimation prevents projects from becoming stable and taken for granted. The concept of PPP provides opportunities for further research. First, what are general conditions under which PPPs evolve into OPPs? Second, the relationship between PPPs and open technology needs to be further examined: are PPPs technology/environment specific, or can they be found in other social arenas, even inside the realm of science? It may also be relevant to the sociology of cultural production. In contemporary consumer society, where few symbols have universal 234 appeal, the flows in the global mediascape (Appaduari, 1986, 1990) are a perfect arena for the emergence of such nexuses of action. Similarly, this concept can help develop the ideas put forward by Antoine Hennion in his study of popular music producers. Hennion describes the sound studio as "impermeable to systems; it dissolves obligatory associations; it undoes rationalizations. Inversely, all connections are permitted, whether or not they are specified in the user’s manuals. The studio is an apparatus for capturing raw material by extracting it from the structured networks along which it circulates in 'normal' life" (Hennion, 1989, p.410). This description may be made more theoretically cogent by pointing to the seemingly weak privileged passage points and actor-network connections that, nonetheless, give its patterns of action some privileged structure. 8.1.2 Practical Legitimacy A complementary conceptual question which this study emphasizes is the issue of Wikipedia's peculiar mode of legitimacy. This issue became especially salient when trying to make sense of the paradoxical way journalists engage Wikipedia, using it routinely yet at the same time constructing it as an unreliable source and erasing all evidence for having used it from their texts. As journalists claimed that Wikipedia cannot be trusted, and that divulging its use to readers conflicted with their sense of duty, this pattern certainly falls short of Weber’s canonical definition of a legitimate order as one in which actors' actions are guided by a sense of duty, are viewed as "obligatory" or "exemplary," and create relations of domination and obedience (Weber 1978: 31). Wikipedia is thus not endowed with legitimacy in the full sense of the word, whether traditional, rational-legal or charismatic202. However, Wikipedia certainly had some measure of legitimacy, as its use seems to have become part of journalists' standard socialization process, and just as important – in that most interviewees explicitly declared that it is a legitimate journalistic tool. Thus, for journalists, Wikipedia occupies an intermediary position between legitimacy and illegitimacy, which I call practical legitimacy. This concept designates 202 As Weber's types of legitimacy are ideal types, it is not surprising that Wikipedia does not neatly conform to one of them (see Weber, 1978, p. 216). However, I suggest that the case of Wikipedia manifests a type of legitimacy which warrants independent conceptualization (see below). 235 a form of legitimation which grants actions (or objects, knowledge, persons etc.) a limited yet observable measure of authority, as it takes into account and encapsulates both its usability and its liabilities, both its de-facto legitimacy—in the sense that "everybody does that," (Johnson, Dowd and Ridgeway, 2006) and lack of de-jure legitimacy (manifest in repeated expressions of doubts and criticism).203 Wikipedia stands out in its ability to sustain an extremely large and diverse readership despite – or perhaps even because of – this tension-ridden and contradictory form of legitimacy. Wikipedia plays a central role in shaping the very conditions of its consumption, as manifest in several of its salient features: First, the use of the culturally prevalent and previously-legitimated model of encyclopedic articles helps in shaping editors and readers' expectations of Wikipedia, as it generates a certain measure of trust, credibility and legitimacy that distinguish it from other internet information resources, and serve as a standard for determining article quality204. Second, mechanisms such as talk pages and warning templates, originally built for the benefit of Wikipedia's editors, construct Wikipedia as an unfinished project and its ideal reader as a knowledge producer, interested in engaging tools and mechanisms designed to construct and assess the articles she reads. Wikipedia's legitimacy thus remains inherently contextual, and can only be determined on a case-by-case manner. The price of this limited authority is manifested in restricted public recognition and acknowledgement of its use. Weber's definition of legitimacy, ultimately rooted in political ideas of totalistic duty and order (Weber, 1978, pp. 212-215), is ill-suited to conceptualize such a notion of legitimacy. Neither is the influential idea of regimes of justification developed in the context of the new French pragmatic sociology or as it is also called, 203 This concept obviously has some affinities with Webber's Practical Rationality, as its name attests. However, while Weber's concept designated a way of life which "accepts given realities and calculates the most expedient means of dealing with the difficulties they present," (Kalberg, 1980) practical legitimacy nonetheless is preoccupied with the legitimacy of the act as much as it is preoccupied with its consequences. Moreover, it is not distinctly rational, as it relies on habituation and socialization processes. 204 However, as demonstrated in chapter 6, Wikipedia did diverge from the traditional encyclopedic model in important aspects which helped shape its perception and increase its usability at the price of limiting its ability to construct itself as legitimate. Perhaps the most notable exception was the inclusion of topics usually not found in traditional encyclopedias, such as popular culture, sports and current events. The willingness to engage topics related to contemporary news and popular cultural events (in the broadest sense of the term) appealed to a large crowd interested in a relatively dry encyclopedia style articles on such subjects. 236 sociology of criticism (Boltanski and Thévenot, 2006). As Silber (2003) noted, theirs has affinities with a form of strong cultural sociology program, aimed at going beyond repertoire theory and providing a general order or structure to the cultural "tool-kit" (Swidler, 1986), in the form of regimes of justification or orders of worth. However, the justifications and legitimations found in the context of this study do not fit well any of the regimes they describe. Moreover, as they focus on the articulation of universal modes of justification, their theory it is less attuned to Wikipedia's local and contingent form of legitimation, rooted in specific practical-epistemic concerns205. Moreover, Thévenot’s later distinction between regimes of engagement (Thévenot, 2001a) falls short of accounting for phenomena which precisely challenge the separation between familiarity, regular planned action and justification. Practical legitimacy, emphasizing the rapid alternation between personal habits, conventional precepts and general notions of knowledge, does exactly that. A more flexible approach may be found in Berger and Luckmann's (1967) treatise in the sociology of knowledge, which moved the focus away from the binary oppositions of “particular vs. universal” and “duty vs. free will” to a continuum of possible measures of institutionalization and legitimation. They see legitimation as an indispensible part of the ongoing process of institutionalization that gradually transforms social practices into objectified and culturally legitimized institutions. Berger and Luckmann differentiate between four levels of legitimation: incipient legitimation, implicit in every utterance which objectifies the social world; theoretical propositions, i.e. various maxims, proverbs and explanatory schemes which explain and justify the institutional world; explicit theories, which appear when a differentiated body of knowledge is erected to provide legitimacy, usually becoming the domain of experts; and symbolic universes, which "integrate different provinces of meaning and encompass the institutional order of a symbolic totality" (ibid, p. 95). It should be noted that Berger and Luckmann view this process as gradual and incremental. This is evident both from their naming these forms of legitimation "levels" – suggesting a hierarchy between them – and in their nested deployment, where each level assumes and incorporate the level "below him." There seems to be 205 However, I find their notions of "tests" and form investment very helpful, and indeed used it in various parts of this work. 237 here a logic that sees institutionalization as completed only after adequate theorization has taken place. Similarly, Johnson et al. (2006) claim that the legitimation of new objects goes through four stages: innovation, local validation, diffusion and general validation, suggesting that it only becomes truly legitimate when it is perceived as inevitable. While this may seems as a very different criterion, it is similar to Berger and Luckmann's conceptualization in that it assumes a universally sanctioned status as the natural end point of an incremental process of institutionalization. The notion of practical legitimacy, on the other hand, focuses the gaze on intermediary phenomena, which do not yet have—and, indeed, may never have— theoretical justifications or the veneer of inevitability. Whether such phenomena are inherently temporary or not, bound to end in full legitimation, dissolve or simply remain suspended between the practical and the epistemic, they warrant research in their own right. Such research may improve our understanding of institutionalization processes, tracing concrete trajectories of legitimation and constructing more refined analytic structures to capture their similarities and variations. Starting with practices rather than with fully fleshed out legitimations, the question of whether practical legitimacy is a general characteristic of under-theorized legitimacy formations, or a locally bounded form of legitimation whose features are determined by its social, cultural and technological environment, is inherently an empirical one. Two research strategies may stem from it: the first attempting to identify similar modes of legitimacy in innovative projects whose features and environment are very different from those of Wikipedia, and to establish whether practical legitimacy is a mode of legitimation endemic in to such projects; the second attempting to identify similar modes of legitimacy in digital, widely distributed knowledge production projects, to assess whether Wikipedia takes part in a wider epistemic culture endemic to such environments. Taking the first route means looking at historically removed knowledge production projects, at real life (rather than online) environments, and at institutions that that deal with processes other than knowledge production; taking the second means looking at start-up ventures, at FOSS projects, and at internet environments such as Youtube or file-sharing networks. In any case, both routes are similar in their attempt to uncover the practical tools and objects which may engender and sustain this distinctive type of legitimacy. 238 To summarize this section, I wish to describe the relations between PPPs and pragmatic legitimacy. Both relate to a situation in which some actors are able to direct the actions of others in a limited yet effective manner and retain their trust and compliance, though not unconditionally and without reservations. While this is merely a hypothesis in this stage, I suggest that while they are analytically distinct phenomena – one preoccupied with knowledge production and the other with its consumption or use - they tend to go hand in hand. They both share a logic of flexibility which defies, at least temporarily and partially, the traditional logic of institutionalization and legitimation, and emphasize agency to a much larger extent. While many (if not most) knowledge production projects, and especially scientific ones, tend to be fully institutionalized, immersed in expert-systems, theorizations and symbolic universes, projects that do not entail these features will be more likely to adopt mechanisms and practices which manifest both the formation of PPPs and a reliance on practical legitimacy (or other alternatives to it), and thus enhance their flexibility and reflexivity at the expanse of their stability and legitimacy. In the following section, I discuss the implications of these differences on such projects' utilization of gifts of knowledge, as reflected by the differences between Nupedia and Wikipedia. 8.2 The Gift of Knowledge The review of the literature regarding the gift focused on a salient trend in contemporary conceptualizations of it, namely the heightened awareness to the diversity of gifts and gift practices. A major strategy in dealing with this diversity involves attempts to construct analytic typologies of types or dimensions of giftgiving practices (Vandevelde, 2000; Caille, 2001; Graeber, 2001; Hénaff, 2010b). Following Silber (2007, 2013), however, I preferred to use the notion of a gift repertoire, which emphasizes the need to contextualize different types of gifts, their coexistence and competition, in specific historical and cultural circumstances. Closer to the pragmatic orientation which guided my research, this approach emphasizes the importance of creative agency in selecting from existing repertoires, and in broadening them. Thus, I tried to demonstrate how attempts at creating a free online encyclopedia took into account and at the same time shaped such a repertoire of acts 239 of knowledge giving. The findings of this research highlight the substantive contribution of the gift repertoire approach: rather than generally specifying types of knowledge, it allows for a more nuanced understanding of gift practices which helps in explaining the varying success of projects attempting to solicit gifts and contributions of knowledge in specific social and cultural contexts. This contribution was most apparent in my attempts to explain why Wikipedia succeeded while its predecessor, Nupedia, failed. Both Wikipedia and its predecessor Nupedia incorporated existing modes of giving and attempted to adjust them to their needs. However, they drew gift-practices from different cultural contexts with distinct gift-repertoires and different tolerance to changes and adjustments. The flexibility of these repertoires and their congruence with Nupedia and Wikipedia's social and technological environments had important effect on their success. Nupedia's production relied predominantly on the repertoire of practices prevalent in the world of higher education, with which both founders were familiar. Imitating the model of peer-reviewed scholarly publications (which rely on unpaid contributions of knowledge), they attempted to construct a viable gift economy attractive to credentialed experts. However, this attempt to translate academic practices into a voluntary online context failed, due to the discrepancies between the imported repertoire and its new setting, which obstructed both reciprocity-based, group-oriented and individual giving. In contrast with the bulk of online based gift-economies, in which contribution builds reputation in a "virtual community", Nupedia attempted to harness the real life reputation of its contributors by offering them to further enhance it. However, being affiliated with a start-up business venture rather than established academic institutions Nupedia lacked the symbolic capital needed to create and sustain an attractive, viable gift-economy. This problem was further complicated by the fact that most of the tasks needed to build Nupedia – such as writing, editing and peer-reviewing articles – were disturbingly similar to academic chores. Offering a lot less recognition and symbolic capital for a task very similar to one's daily chores made it very hard for Nupedia to compete for experts' spare time. In addition, recognition and acknowledgement (manifested in the form of bylines, personal biography pages etc.) was not proportional to the amount of work done, but rather to the credentials of the contributors. Severing the connection between gift and retribution may have undermined enthusiasm for reciprocity based giving. 240 The reliance on the academic hierarchy and disciplinary divisions inherent to its gift repertoire also prevented Nupedia from creating a more generic onlinecommunitarian context for giving, based on interest in intra-community reputation. This problem was manifested in its sociotechnical structure, based on many specialized disciplinary-defined mailing lists206 and lacking a central location for social integration. Thus, it failed to form a global microstructure207 (Knorr-Cetina, 2005) that could transcend its nature as a formal organization, and generate the feeling of communal participation and engagement needed to encourage group-oriented giving and generalized exchange. Nupedia's attempts to sustain a viable gift economy also hampered its ability to attract individual giving. Giving knowledge to Nupedia usually demanded participants to undergo a rather lengthy identification process, divulge their real life identity, and become incorporated in its formal structure and task-designation mechanism before beginning to contribute to it. These demands posed a barrier between the willingness to give and its fulfillment, which may have deterred potential contributors interested in free, immediate and unconstrained giving of knowledge. Last but not least, Nupedia failed to abide by one of the most essential aspects of the gift process: the obligation to receive (Mauss, 1990 [1925]). Knowledge given to Nupedia was not accepted as is, but rather had to undergo a rigorous review process by experts. This method of operation may very well have damaged individual givers' enthusiasm towards giving knowledge – especially since it made their act of giving even more similar to academic chores. Wikipedia's founders' decision to abandon Nupedia's constraining process had a dramatic effect on the gift-related aspects of the project. Free from Nupedia's need to reciprocate the serious work of its contributors with substantial symbolic capital, it could become a much more flexible, playful and fun project, enhancing its potential contributors' sense of freedom in the very act of giving. This move instilled in Wikipedia a pluralistic and liberal approach which encouraged a variety of giving practices that Wikipedia's founders themselves often did not foresee. However, this does not mean that extant gift-repertoires did not play an 206 Nupedia did have a general mailing list, but it failed to attract a large following as it seldom included significant discussions. 207 Global microstructures, based on features characteristic to face to face interactions such as trust and acquaintance, rather than on clear structure and hierarchy, seem more supportive of gift relations than formal organizations. 241 important part in Wikipedia. A significant factor explaining its success is the congruity between the gift-repertoire it appropriated and its environmental context, reflected in expectations of its potential contributors. In contrast with Nupedia reliance on academic-like practices, Wikipedia attempted (and evidently succeeded) to reproduce if also reconstruct gift-practices prevalent in the online environment it operated in. Specifically, it borrowed heavily from the repertoire associated with socalled "virtual communities". As the literature indicates, there is an institutionalized gift-repertoire which developed in such context over the nineties: acts of giving in such contexts are justified through a "gift giving-culture," which constructs of sharing and openness as virtues (Kollock, 1999); they are sustained through social acknowledgement and prestige (Bruns, 2008; Raymond, 1998; Tapscott and Williams, 2006), and are reciprocated through generalized exchange, where each participant contribute to the community rather than to a specific beneficiary (Bauwens, 2005; Bergquist and Ljungberg, 2001; Giesler, 2006; Jenkins, 2006; Kollock, 1999; Rheingold, 2000). This repertoire was familiar to the Wikipedia's potential contributors, who used similar (though not identical) practices in their contribution to mailing lists and internet forums. Transforming the skeletal Wiki technology Wikipedia utilized in its early days to the locus of a virtual community demanded constructing mechanisms for informal appreciation and retribution. Thanks to the flexibility of the Wiki platform, such mechanisms, including informal discussion pages, talk pages, and individual userpages were promptly built, supplementing the central mailing list (rather than Nupedia's many lists), opened by Wikipedia's founders. Such mechanisms, which provided spaces for dialogue, community building and reputation accumulation, aided in securing the participation of contributors by embedding their gifts of knowledge in a concrete social context. However, the most conspicuous change in the gift practices signified by the transition from Nupedia to Wikipedia was encouraging immediate, instant individual giving. Doing away with Nupedia's cumbersome registration procedure and lengthy review process, Wikipedia allowed new potential participants to try and contribute without any constraints or delays. New or occasional users had ample evidence for the project's openness to such giving: first, the standard text at the bottom of each page 242 reading "you can edit this page right now! It's a free, community project"; second, the lack of named credit in article pages prevented the feeling of interfering with someone else's work; third, the unfinished and in-process appearance of Wikipedia's contents, which invited readers to fix and expand them; and finally, the intuitive and simple edit mechanism and instant publication which allowed its participants to experience the immediate joy of giving one's knowledge and watching it get published – a joy that is so often delayed, frustrated or denied in professional or academic lifestyle. While this unorthodox mode of knowledge giving did not provide the lion's share of contents added to Wikipedia, it was crucial as an initial phase on the path to a more sustained group-oriented giving. As time passed, the formation of mechanisms sustaining group-oriented giving and their institutionalization gradually laid more constraints on individual giving. The formation of standards and policies which determining the relative value of different contributions, inextricably connected to group-oriented giving, narrowed the range of possibilities open to potential individual contributors. As rules became more restrictive, more initial socialization was needed to give knowledge successfully. However, as my discussion of PPP's suggests, this institutionalization process consistently left room for individual, unconstrained contributions. As Wikipedia's founders insisted on the availability and importance of immediate contribution, its standards and policies functioned as PPP's rather than strict rules, allowing for a measure of variability and reversibility. Keeping Wikipedia's giftrepertoire open to individual gifts, and to many forms of acceptable content, was crucial to its success just as its institutionalization was. Indeed, the comparison between Wikipedia and Nupedia illustrates of the importance of flexibility and of balancing freedom and obligation in the construction of a gift-repertoire. However, in the long run, this relatively inclusive approach seems to have waned – and along with it, some of Wikipedia's success. As a recent study (Halfaker et al., 2013) found, the number of active editors208 in Wikipedia peaked around the beginning of 2007 with more than 50,000, and has gradually been declining ever since. According to this study, the main reason for this decline is its growing restrictiveness: to deal with the massive inflow of traffic brought about by its growing popularity, Wikipedia gradually became more rigid and less welcoming. It began 208 Active editors are ones who made at least 5 edits per month. 243 incorporating technological tools to prevent possibly harmful edits which took the place of deliberation. Moreover, its standards and policies also became increasingly institutionalized, and suggestions for changes to them are usually rejected – especially when they come from newcomers. These steps, which prevent immediate individual giving, distance new editors that could have become part of Wikipedia's core community of active contributors. Losing its balance which allows for various types of giving, Wikipedia is increasingly having trouble replacing editors to replace ones that leave the project. This study further affirms the conclusions described above, as it highlights the crucial importance of flexibility and the coexistence of various modes of giving, even at the price of diminished control. However, it should be noted that Wikipedia is still very live and active, and that there are still a great number of participants who are willing to contribute to it. This willingness, essential to Wikipedia's functioning both in the past and in the present, relies in an understanding of knowledge giving as a virtuous act and on the attribution of value to the specific knowledge given209. Thus, understanding Wikipedia's success demands not only understanding the particular mechanisms of knowledge giving it was built with, but also the larger social and cultural context in which it was embedded and which was and still is essential to its ability to attract and retain contributors. The nature of this context will be discussed in the next, final section of this chapter. 8.3 Wikipedia and Contemporary Epistemic Culture In the introduction to this work, I noted the widespread consensus that contemporary society is increasingly becoming an information (or knowledge) society – a society in which economic, political and cultural structures and processes are increasingly related to, and dependent upon the production and dissemination of knowledge and information. At the same time, highly contentious disagreements remain regarding the viability of the concept of modernity in this context, at least if used to connote an existence of a common cultural center, a sense of historical continuity with the vision of enlightenment, and the centrality of rationality, science 209 For a discussion of the ethical context of online knowledge giving in terms of virtue ethics see Benkler and Nissenbaum (2006). 244 and expertise. While some claim that contemporary knowledge society is a continuation or even intensification of modernity in that sense (cf. Giddens, 1991; Beck, 2000), resulting in so-called late or second modernity, others maintain that it represents a break from it, with new and dissenting knowledge production structures, processes and legitimation principles challenging and replacing those prevalent in modernity (cf. Best and Kellner, 1997; Featherstone, 1989, 1991; Latour, 1993). Yet others claim that a cultural center based on the vision described above remains dominant, while various subcultures flourish at its margins, fostering immanent standards, practices and justifications but failing to achieve widely acknowledged legitimacy (Laermans, 1992; Lash, 1994). This work seems at first glance to buttress this last assertion, as the internal standards whose it has internal standards that are not generally acknowledged as legitimate, and is similar in many of its aspects to other subcultures flourishing in the context of the new media, whether appearing in the shape of networks210, reflexive communities (Lash, 1994) or global microstructures (Knorr-Cetina 2005; KnorrCetina and Brugger, 2000). Still, and as opposed to marginal subcultures, it has come to occupy a consistent, relatively stable and highly influential position to the side of modernistic knowledge production arenas such as the academic field, the medical field and the field of journalism, feeding on them and at the same time supplying them with usable information. This position, I contend, would not have been achievable if Wikipedia's unorthodox practices and structures were not constructed and articulated in a way that kept intact, as much as possible, the modern ethos of expertise and the established hierarchies of knowledge (see chapter 6, especially its conclusion). In such terms, Wikipedia touches right to the heart of the debate concerning modernity and postmodernity, in a way that posits yet another paradox: defining Wikipedia's epistemic culture as either a modern or a postmodern phenomenon will not do it justice, as some of its elements will stay unaccounted for. Thus, the question of modernity may be better addressed if we avoid attempting to reach a "clear and distinct" concept of modernity, whether relevant or irrelevant in current knowledge society. Rather, it may be better to see modernity as a potent, dynamic and relatively 210 In the meaning Castells (1996, 1997, 1998) uses – i.e., as a distinct form of social structure – and in contrast with Latour's view of it as it as the universal social structure. 245 unstructured mix of practices and justifications, often conflicting with one another211. Modernity should thus be read as a cultural repertoire (Lamont and Thévenot, 2000; Silber, 2003) or configuration so huge, historically layered and wrought with inner tensions, contradictions and controversies that its contours and concrete manifestations can only be assessed partially and locally. So how does Wikipedia's epistemic culture relate to this configuration or repertoire? In some senses, it challenges major components related to it, as it is based on the decoupling of knowledge and expertise and of theoretical and practical legitimation, two amalgams standing at the core of modern theories and methodologies. However, it does so in a way that assumes the legitimacy of modern epistemologies, rather than confronts or undermines them, as it leaves the final judgment of reality in the hands of experts and scientists. Moreover, it is both explicitly and implicitly committed to the Enlightenment's critical and empiricist ethos, as it demands of its readers to function as autonomous judges of the authority of its contents. Thus, while Wikipedia represents a new epistemic culture, it does not reject but rather builds on, incorporates and extends central elements of the modern vision of knowledge. This approach has some close affinities with those of a number of theorists who insist on the prevalence of modernity in contemporary culture, yet at the same conceptualize it as multifarious, problematic, and prone to reconceptualization or opposition, such as S.N. Eisenstadt (2001), Peter Wagner (2008), Bjorn Wittrock (2000) or – from a very different perspective, but perhaps all the more relevant – Nissim Mizrachi (Mizrachi, Drori and Anspach, 2007). The Latter's characterization of modernity as an "awareness of the loss of the markers of certainty and of a continuous search for them, while conscious of the impossibility to attain such certainty" (Delanty, 2004: 395–396) seems especially relevant in this respect. Wikipedia's struggles to find an epistemic foothold are direct heirs to modern 211 While Latour (1993) gave a compelling and provocative description of the modern Constitution as a series of conflicting principles and paradoxical statements, his scheme is ultimately to symmetrical and ordered, constructed to address solely the issue of separation between humans and non humans and avoiding the complexities stemming from the plethora of concepts and practices associated with modernity. In doing so, he violates the very logic of networks he himself advocates in different contexts. 246 knowledge production projects' search for certainty, as well as their doubts towards its attainability. However, even as it holds on to the categories and justifications of modernity – an attitude most apparent in Jimmy Wales assertion that "what people believe is a matter of objective fact, and we can present that quite easily" (Wikipedia, 2001ak) Wikipedia advances a new epistemic culture and challenges major modern institutions. In doing so, it changes our very definition of ourselves as knowledge acquiring subjects as well as our everyday practices, and brings about new sorts of epistemic virtues, gradually gaining public recognition and use. This rapidly growing culture should be understood in its own terms, rather than in terms of a simple Modern/Postmodern dichotomy. The reconstruction of contemporary popular, voluntary epistemic arenas is of vital concern to the sociology of knowledge, if the latter is to remain relevant. It is in this context that the social construction of knowledge takes place on a huge, global scale, possibly larger and more pervasive than research done by experts in laboratories and scientific institutions212. Rather than judged as ephemeral or inadequate (cf. Sanger, 2009, Wray, 2009), the popular online production of knowledge should be understood as the frontier of contemporary information society, where new insights into the nature of knowledge and its production may be gained, if we engage in reconstruction, rather than (mis)judgment. As Daston and Galison (2007, p. 41) noted, "new epistemic virtues come into being; old ones do not necessarily pass away... but the older ones are inevitably modified by the very existence of newer ones, even if they are not replaced by it…the very multiplicity of epistemic virtues can cause confusion or even accusation, if adherents of ones are judged by the standards of the other." 212 In some areas, there is already a growing awareness to the importance of knowledge produced by "unqualified" persons (c.f. Indyk and Rier, 1993; Rier and Indyk, 2006; Wynne, 1996; Yearley, 2000). 247 Bibliography Abercrombie, Nicholas, and Brian Longhurst. 1998. Audiences: A Sociological Theory of Performance and Imagination. London: Sage. Alexa, the Web Information Company (2011). "Top 500 Sites on the Web." Alexa. (http://www.alexa.com/topsites, retrieved September 10, 2011). Almeida, Rodrigo B., Barzan Mozafari, and Junghoo Cho. 2007. "On the Evolution of Wikipedia." In Proceedings of the International Conference on Weblogs and Social Media. (http://oak.cs.ucla.edu/~cho/papers/almeida-icwsm07.pdf, retrieved January 18, 2011). Alvesson, Mats. 2003. "Beyond Neopositivists, Romantics, and Localists: a Reflexive Approach to Interviews in Organizational Research." Academy of Management Review, 28(1): 13-33. Anthony Denise, Sean W. Smith and Tim Williamson. 2007. "Explaining Quality in Internet Collective Goods: Zealots and Good Samaritans in the Case of Wikipedia." Hanover : Dartmouth College. (web.mit.edu/iandeseminar/Papers/Fall2005/anthony.pdf, retrieved September 17, 2011) Appadurai, Arjun. 1986. “Introduction: Commodities and the Politics of Value," in Arjun Appadurai (ed.) The Social Life of Things: Commodities in Cultural Perspective. Cambridge: Cambridge University Press, pp. 3-63. Appadurai, Arjuan. 1990. "Disjuncture and Difference in the Global Cultural Economy." Theory,Culture and Society. 7: 295-310. Arazy, Ofer, Wayne Morgan and Raymond Patterson. 2006. "Wisdom of the Crowds: Decentralized Knowledge Construction in Wikipedia." Paper presented at 16th Annual Workshop on Information Technologies and Systems (WITS) Paper. (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1025624, retrieved September 17, 2011) Baker, Daniel. 2009. "Citations to Wikipedia in Law Reviews." University of Houston Law Center No. 2009-A-42. (http://ssrn.com/abstract=1525619, retrieved December 11, 2011) Barbrook, Richard. 1998. "The Hi-Tech Gift Economy." First Monday 3(12). Barbrook, Richard. 2000. "Cyber-Communism: How the Americans are Superseding Capitalism in Cyberspace." Science as Culture 9.1: 5-40. Barbrook, Richard, and Andy Cameron. 1996. "The Californian Ideology." Science as Culture 6(1): 44-72. 248 Barnes, Barry, David Bloor and John Henry. 1996. Scientific Knowledge: A sociological Analysis, London: Athlone Press, Ch. 7. Pp. 169-199. Barton, Matt. (2005). "The Future of Rational-Critical Debate in Online Public Spheres." Computers and Composition 22(2): 177–190. Baudrillard, Jean. 1983. In the Shadow of the Silent Majorities. New York: Semiotext(e). Baudrillard, Jean. 1994. Simulacra and Simulation. Ann Arbor: University of Michigan. Bauwens, Michel. 2005. "The Political Economy of Peer Production." CTheory. (http://www.ctheory.net/articles.aspx?id=499, retrieved June 9, 2011). Beck, Ulrich and Natan Sznaider. 2010. "Self-Limitation of Modernity? The Theory of Reflexive Taboos." Theory and Society 40(4): Pp. 417-436 Beck, Ulrich, Anthony Giddens, and Scott Lash. 1994. Reflexive Modernization: Politics, Tradition and Aesthetics in the Modern Social Order. Stanford: Stanford University Press. Beck, Ulrich. 2000. "The Cosmopolitan Perspective: Sociology of the Second Age of Modernity." British Journal of Sociology 51(1): 79—105. Bell, Daniel. 1973. The Coming of Post-Industrial Society: A Venture in Social Forecasting. New York: Basic Press. Bell, Mark. 2007. "The Transformation of the Encyclopedia: A Textual Analysis and Comparison of the Encyclopaedia Britannica and Wikipedia." Unpublished master’s thesis, Ball State University, Muncie, IN. (http://jwchesebro.iweb.bsu.edu/digitalstorytelling/Theses/Bell_Final_Version _Thesis_April_4_2007.pdf, retrieved June 7, 2011). Bellomi, Francesco and Roberto Bonato. 2005. "Network Analysis for Wikipedia." Paper presented in Wikimania 2005, Frankfurt, Germany. (http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.178.2233, retrieved June 9, 2011). Benjamin, Walter. 1968. "The Work of Art in the Age of Mechanical Reproduction", pp. 214–218 in Hannah Arendt (ed.) Illuminations. London: Fontana. Benkler, Yochai. 2006. The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven: Yale University Press. Benkler, Yochai. 2010. "Law, Policy, and Cooperation," pp. 299-332 in Edward Balleisen and David Moss (eds.) Government and Markets: Toward a New Theory of Regulation. Cambridge: Cambridge University Press Benkler, Yochai. 2011. The penguin and the leviathan: How cooperation triumphs over self-interest. Random House LLC. 249 Benkler, Yochai and Helen Nissenbaum. 2006. "Commons-based Peer Production and Virtue." The Journal of Political Philosophy 14, (4): 394–419. Berg, Bruce. 1998. Qualitative Research Methods for The Social Sciences (3rd ed). Boston: Allyn and Bacon. Berger, Peter L., and Thomas Luckmann. 1967. The Social Construction of Reality; a Treatise in the Sociology of Knowledge. Garden City, NY: Doubleday. Bergquist, Magnus and Jan Ljungberg. 2001. "The Power of Gifts: Organizing Social Relationships in Open Source Communities." Information Systems Journal 11, Pp. 305–320. Berthoin-Antal, Ariane and Richebé, Nathalie. 2009. "A Passion for Giving, a Passion for Sharing. Understanding Knowledge Sharing as Gift Exchange in Academia." Journal of Management Inquiry, Vol. 18, No. 1, 78-95. Best, Steven and Douglas Kellner. 1997. The Postmodern Turn. New York: Guilford Press. Bijker, Wiebe. 1993. “Do not Despair: There is Life after Constructivism.” Science, Technology and Human Values 18 (1): 113-138. Bijker, Wiebe. 1995a. “Sociohistorical Technology Studies” in Sheila Jasanoff, Gerald E. Markle, James C. Peterson, Trevor J. Pinch (eds.) Handbook of Science and Technology Studies. London: Sage Publication. Bijker, Wiebe. 1995b. Of Bicycles, Bakelites, and Bulbs: Toward a Theory of Sociotechnical Change. Cambridge: MIT Press. Bijker, Wiebe and John Law. 1992. Shaping Technology/Building Society Studies in Sociotechnical Change. Cambridge: MIT Press. Blau, Peter. 1964. Exchange and Power in Social Life. New York: John Wiley and Sons. Bloor, David. 1976. Knowledge and Social Imagery. London: Routledge and Kegan Paul. Bloor, David. 1986. "Wittgenstein and Mannheim on the Sociology of Mathematics.” Stuart Shanker (ed.) Ludwig Wittgenstein: Critical Assessments, Vol. 3. Pp. 378-394. Blumer, Herbert. 1969. Symbolic Interactionism; Perspective and Method. Englewood Cliffs, NJ: Prentice-Hall inc. Bo฀hme, Gernot, and Nico Stehr. 1986. The Knowledge Society: The Growing Impact of Scientific Knowledge on Social Relations. Dordrecht: D. Reidel Publishing Company. 250 Boltanski, Luc and Ève Chiapello. 2005. The New Spirit of Capitalism. London: Verso. Boltanski, Luc and Laurent Thévenot. 2006. On Justification: Economies of Worth. Princeton: Princeton University Press. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Bourdieu, Pierre. 1984. Distinction: A Social Critique of The Judgment of Taste. Cambridge: Harvard University Press. Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Harvard University Press. Bourdieu, Pierre. 1998. Practical Reason: On the Theory of Action. Stanford University Press. Bowker, Geoffrey C., and Susan Leigh Star. 1999. Sorting Things Out: Classification and Its Consequences. Cambridge: MIT Press. Breslau, Daniel, and Yuval Yonay. 1999. "Beyond metaphor: Mathematical models in economics as empirical research." Science in Context 12(2): 317-332. Brown, Adam R. 2011. "Wikipedia as a Data Source for Political Scientists: Accuracy and Completeness of Coverage." PS: Political Science and Politics, 44: 339343. Bruns, Axel. 2008. Blogs, Wikipedia, Second Life and Beyond: From Production to Produsage. New York: Peter Lang. Bryant, Susan L., Andrea Forte and Amy Bruckman. 2005. "Becoming Wikipedian: Transformation of Participation in a Collaborative Online Encyclopedia." In Proceedings of the 2005 International ACM SIGGROUP Conference on Supporting Group Work. New York: ACM. (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.62.5337&rep=rep1 &type=pdf&BryantForteBruckBecomingWikipedian.pdf, retrieved February 08, 2007). Buriol, Luciana, Carlos Castillo, Debora Donato, Stafano Leonardi and Stefano Millozzi. 2006. "Temporal Analysis of the Wikigraph." In WI '06: Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence): 45-51. Caille, Alain. 2001. "The double inconceivability of the pure gift." Angelaki 6. 2):23 – 39. Callon, Michel. 1986. "Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fishermen of St. Brieuc Bay." Pp. 196-233 in John Law (ed.) Power, Action, Belief: A New Sociology of Knowledge?. London: Routledge and Kegan Paul. 251 Capocci, Andre, Vito D. P. Servedio, Francesca Colaiori, Luciana Buriol, Debora Donato, Stafano Leonardi and Guido Caldarelli. 2006. "Preferential Attachment in The Growth of Social Networks: The Internet Encyclopedia Wikipedia." Physical Review E, 74(3). Castells, Manuel. 1996. The Rise of Network Society, The Information Age: Economy, Society and Culture Vol. I. Oxford: Blackwell. Castells, Manuel. 1997. The Power of Identity, The Information Age: Economy, Society and Culture Vol. II. Cambridge, Oxford: Blackwell. Castells, Manuel. 1998. End of Millennium, The Information Age: Economy, Society and Culture Vol. III. Cambridge, Oxford: Blackwell. Castells, Manuel. 2001. The Internet Galaxy: Reflections on the Internet, Business, and Society. Oxford: Oxford University Press. Caverly, David C. and Ward, Anne. 2008. "Techtalk: Wikis and Collaborative Knowledge Construction." Journal of Developmental Education, 32(2): 36-37. Chandler, Cullen J. and Alison S, Gregory. 2010. "Sleeping with the Enemy: Wikipedia in the College Classroom." History Teacher, 43(2): 247-257. Chandler-Olcott, Kelly. 2009. "A Tale of Two Tasks: Editing in the Era of Digital Literacies." Journal of Adolescent and Adult Literacy, 53(1): 71-74. Ciffolilli, Andrea. 2003. "Phantom Authority, Self-Selective Recruitment and Retention of Members in Virtual Communities: The Case of Wikipedia." First Monday 8(12). Clarke, Adele E. and Susan Leigh Star. 2003. "Symbolic Interactionist Studies of Science, Technology and Medicine." Pp. 539-574 in Handbook of Symbolic Interactionism, edited by Larry Reynolds and Nancy Herman. Walnut Creek: Alta Mira Press. Clauson, K.A., Polen H.H., Boulos M.N.K., Dzenowagis J.H. 2008. "Scope, Completeness, and Accuracy of Drug Information in Wikipedia." Ann Pharmacother, 42(12):1814–21. Cohen, Eyal. 2007. "The Latest on Virginia Tech, from Wikipedia." New York Times. (http://www.nytimes.com/2007/04/23/world/americas/23ihtwiki.1.5405005.html, retrieved April 9, 2011). Cohen, Eyal. 2011. "Wikipedia." New York Times. (http://topics.nytimes.com/top/news/business/companies/Wikipedia/, retrieved May 25, 2011). Collins, Harry. M. 1982. "Knowledge, Norms and Rules in the Sociology of Science." Social Studies of Science, Vol 12. Pp. 299-30. Collins, Jim. 1989. Uncommon Cultures: Popular Culture and Post-Modernism. New-York: Routledge. 252 Comor, Edward. 2011. "Contextualizing and Critiquing The Fantastic Prosumer: Power, Alienation and Hegemony" Critical Sociology, 37 (3), pp. 309-327. Cronin, Blaise. 1995.The Scholar's Courtesy. London: Taylor Grahm. Daston, Lorraine and Peter Galison. 2007. Objectivity. New York: Zone. Delanty, Gerard. 2004. "An Interview with S.N. Eisenstadt: Pluralism and the Multiple Forms of Modernity." European Journal of Social Theory 7(3): 391404. Denning, Peter, Jim Horning, David Parnas and Lauren Weinstein. 2005. "Wikipedia Risks." Communications of the ACM, 48(12): 152. Denzin, Norman K. 2007. "Analytic Induction." in George Ritzer (ed.) Blackwell Encyclopedia of Sociology. Blackwell Publishing. Dewey, John. 1941. "Propositions, Warranted Assertibility, and Truth." The Journal of Philosophy, Vol. 38, No. 7. Dewey, John. 1944 [1916]. Democracy and Education: an Introduction to the Philosophy of Education. New York: The Free Press. Drucker, Peter F. 1993. Post-Capitalist Society. New York: HarperBusiness. Durkheim, Émile. 1995 [1912] The Elementary Forms of Religious Life. New York: Free, 1995. Durkheim, Emill and Marcel Mauss. 2010 [1903]. Primitive Classification. London: Routledge. Eijkman, Henk. 2010. "Academics and Wikipedia: Reframing Web 2.0+ as a disruptor of traditional academic power-knowledge arrangements." CampusWide Information Systems, Vol. 27(3) pp.173 - 185 Eisenstadt, Shmuel N. 2001. "The Civilizational Dimension of Modernity: Modernity as a Distinct Civilization." International Sociology 16(3): 320-340. Elia, Antonella. 2006. "An Analysis of Wikipedia Digital Writing." In Proceedings of the EACL 2006 Workshop on New Text - Wikis and blogs and other dynamic text sources, Trento, Italy, 3-7 of April 2006. Emigh, William and Susan C. Herring. 2005. "Collaborative Authoring on the Web: A Genre Analysis of Online Encyclopedias." Pp. 99 in Proceedings of the 38th Annual Hawaii International Conference on System Sciences. Vol. 4. Los Halamatos: IEEE Press. Emirbayer, Mustafa and Douglas Maynard. 2011. "Pragmatism and Ethnomethodology." Qualitative Sociology, Volume 34, Number 1. Pp. 221261. 253 Encyclopaedia Britannica. 2006. "Fatally flawed: Refuting the recent study on encyclopedic accuracy by the journal Nature." Encyclopaedia Britannica (http://corporate.britannica.com/britannica_nature_response.pdf, retrieved November 15, 2011). Farrell, Henry and Melissa Schwartzberg. 2008. "Norms, Minorities, and Collective Choice Online." Ethics and International Affairs, 22(4): 357-367. Featherstone, Mike. 1989. "Towards a Sociology of Postmodern Culture." Pp. 147172 in Hans Hoferkamp (ed.) Social Structure and Culture. Berlin: Walter de Gruyter. Featherstone, Mike. 1991. Consumer Culture and Postmodernism. London: Sage. Flichy, Patrice. 2007a. The Internet Imaginaire. Cambridge: MIT Press. Flichy, Patrice. 2007b. Understanding Technological Innovation: a Socio-technical Approach. Cheltenham, Glos, UK: Edward Elgar. Forte, Andrea and Amy Bruckman. 2008. "Scaling Consensus: Increasing Decentralization in Wikipedia Governance." Pp.157–166 in proceedings of HICCS 2008. Forte, Andrea, Vanessa Larco and Amy Bruckman. 2009. “Decentralization in Wikipedia Governance.” Journal of Management Information Systems. 26(1): 49-72. Foucault, Michel. 2001. The Order of Things: an Archaeology of the Human Sciences. New York: Routledge. Fuchs, Christian. 2008. Internet and Society: Social Theory in the Information Age. New York: Routledge Fujimura, Joan H. 1992. "Crafting Science: Standardized Packages, Boundary Objects, and 'Translation'." Pp 168-211 in Andrew Pickering (Ed.), Science as Practice and Culture. Chicago: University of Chicago Press. Galloway, Alexander. 2004. Protocol: How Control Exists after Decentralization. Cambridge: MIT Press. Geertz, Clifford. 1973. "Thick Description: Toward an Interpretive Theory of Culture." Pp. 3-30 in The Interpretation of Cultures: Selected Essays. New York: Basic Books. Genette, Gérard. 1997. Paratexts: Thresholds of Interpretation. Cambridge: Cambridge University Press. Giddens, Anthony. 1990. The Consequences of Modernity. Stanford: Stanford University press. Giddens, Anthony. 1991. Modernity and Self-Identity: Self and Society in the Late Modern Age. Cambridge: Polity. 254 Giddens, Anthony. 1994. “Living in a Post-Traditional Society.” In Ulrich Beck, Anthony Giddens and Scot Lash, Reflexive Modernization. Stanford, CA: Stanford University Press. Gieryn, Thomas F. 1982a. "Relativist/Constructivist Programs in the Sociology of Science: Redundancy and Retreat." Social Studies of Science, vol. 12: 279297. Gieryn, Thomas F. 1982b. "Not-Last Words: Worn-Out Dichotomies in the Sociology of Science (Reply)." Social Studies of Science Vol. 12: 329-335. Giesler, Markus. 2006. "Consumer Gift Systems." Journal of Consumer Research Vol. 33, No. 2, pp. 283-290. Giles, Jim. 2005. "Internet Encyclopaedias Go Head To Head." Nature, no. 438. Pp. 900–901. Godelier, Maurice. 1999. The Enigma of the Gift. Chicago: University of Chicago. Goffman, Erving. 1971. The Presentation of Self in Everyday Life. Harmondsworth: Penguin. Gouldner, Alvin Ward. 1979. The Future of Intellectuals and the Rise of the New Class: a Frame of Reference, Theses, Conjectures, Arguments, and an Historical Perspective on the Role of Intellectuals and Intelligentsia in the International Class Contest of the Modern Era. New York: Seabury. Graeber, David. 2001. Towards an Anthropological Theory of Value: The False Coin of our own Dreams. New York: Palgrave. Gross, Neil. 2007. “Pragmatism and Phenomenology in 20th-Century American Sociology.” Pp. 183–224 in Craing Calhoun (ed.) Sociology in America: A History. Chicago: University of Chicago Press. Gross, Neil. 2009. "A Pragmatist Theory of Social Mechanisms." American Sociological Review, Vol. 74, No. 3. Pp. 358-379. Guimares Mario J. Jr. 2005. "Doing Anthropology in Cyberspace: Fieldwork Boundaries and Social Environments." Pp.141-156 in Hine, C. (Ed.) Virtual Methods: Issues in Social Research on the Internet. Oxford. Berg. Habermas, Ju฀rgen. 1984. The Theory of Communicative Action. Boston: Beacon. Hafner, Katie and Matthew Lyon. 1996. Where Wizards Stay up Late: The Origins of the Internet. New York: Simon and Schuster. Hagstrom, Warren. 1982. "Gift-Giving as anOrganizing Principle in Science." Pp. 2134 in Barry Barnes and David O. Edge (eds.) Science in Context. Cambridge, Mass.: Mit Press. Halavais, Alexander and Derek Lackaff. 2008. "An Analysis of Topical Coverage of Wikipedia." Journal of Computer-Mediated Communication, 13(2), 429-440. 255 Halfaker, Aaron, R. Stuart Geiger, Jonathan Morgan and John Riedl. 2013. "The Rise and Decline of an Open Collaboration System: How Wikipedia’s Reaction to Popularity Is Causing Its Decline." American Behavioral Scientist. 57: 664688. Harwood, Jonathan. 1993. Styles of Scientific Thought: The German Genetics Community, 1900-1933. Chicago: University of Chicago Press. Hayden, Patrick. 2009. Globalization and Utopia: Critical Essays. Basingstoke: Palgrave Macmillan. Head, Alison J. 2007. "Beyond Google: How Do Students Conduct Academic Research?" First Monday 12(8. (http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/1998/187 3, retrieved December 11, 2011). Head, Alison J. and Michael Eisenberg. 2010. "How Today’s College Students Use Wikipedia for Course-Related Research" First Monday 15(3). Heath, Christian, Hubert Knoblauch and Paul Luff. 2000. "Technology and Social Interaction: The Emergence of ‘Workplace Studies’." The British Journal of Sociology, 51: 299–320. Hebdige, Dick. 1991. Subculture: The Meaning of Style. London: Routledge. Hénaff, Marcel. 2010a. The Price of Truth: Gift, Money, and Philosophy. Stanford, CA: Stanford UP. Hénaff, Marcel. 2010b. "On the Norm of Reciprocity." Teoria e Critica della Regolazione sociale. (http://www.lex.unict.it/tcrs/temi/henaff.pdf, retrieved September 8, 2012) Henderson, Kathryn. 1991. "Flexible Sketches and Inflexible Data Bases: Visual Communication, Conscription Devices, and Boundary Objects in Design Engineering." Science, Technology, and Human Values, 16: 448-47. Hennion, Antoine. 1989. "An Intermediary between Production and Consumption: the Producer of Popular Music", Science, Technology and Human Values 14 (4): 400-424. Hine, Christine. 2000. Virtual Ethnography. Thousand Oaks: Sage. Holloway, Todd, Miran Božicevic and Katy Börner. 2007." Analyzing and Visualizing the Semantic Coverage of Wikipedia and Its Authors." Complexity, 12(3): 30-40. Homans, George. 1958. "Social Behavior as Exchange." American Journal of Sociology 63:597-606. Indyk, Debbie, and David A. Rier. 1993. "Grassroots AIDS Knowledge Implications for the Boundaries of Science and Collective Action." Science Communication 15(1): 3-43. 256 James, William. 2008 [1907]. Pragmatism: A New Name for Some Old Ways of Thinking. Rockville: Arc Manor. Jameson, Fredric. 1991. Postmodernism, or, the Cultural Logic of Late Capitalism. Durham: Duke University Press. Jenkins, Henry. 1992. Textual Poachers: Television Fans and Participatory Culture. New York: Routledge. Jenkins, Henry. 2006. Convergence Culture: Where Old and New Media Collide. New York: New York UP. Jesus, Rut. 2010. "What Cognition does for Wikis." In Proceedings of the 6th International Symposium on Wikis and Open Collaboration (WikiSym '10. ACM, New York, Article 12. (http://doi.acm.org/10.1145/1832772.1832789, retrieved April 15, 2011). Joas, Hans. 1993. Pragmatism and Social Theory. Chicago: University of Chicago Press. Joas, Hans. 1996. The Creativity of Action. Trans. by J.Gaines and P. Keast. Chicago: University of Chicago Press. John, Nicholas A. 2011b. "The diffusion of the Internet to Israel: the first 10 years." Israel Affairs, 17(3), 327-340. John, Nicholas A. 2011a. "Representing the Israeli Internet: The Press, the Pioneers, and the Practitioners." International Journal of Communication (19328036) 5. Johnson, Cathryn, Timothy J. Dowd. and Cecilia L. Ridgeway. 2006. "Legitimacy as a Social Process." Annual Review of Sociology, Vol. 32(1); 53-78. Judd, Terry and Gregor Kennedy. 2010. "A Five-Year Study of On-Campus Internet Use by Undergraduate Biomedical Students." Computers and Education, Volume 55, Issue 4: 1564-1571. Kalberg, Stephen. 1980. "Max Weber’s Types of Rationality: Cornerstones for the Analysis of Rationalization Processes in History." The American Journal of Sociology 85(5): 1145–1179. Keen, Andrew. 2007. The Cult of the Amateur: How Today's Internet Is Killing Our Culture. New York: Doubleday/Currency. Kelly, Kevin. 2008. "What Have You Changed Your Mind About?" New York Times. (http://www.edge.org/q2008/q08_6.html#kelly, retrieved April 9, 2011). Kittur, Aniket and Robert E. Kraut. 2008. "Harnessing the Wisdom of Crowds in Wikipedia: Quality Through Coordination." In Proceedings of the ACM 2008 Conference on Computer Supported Cooperative Work. New York: ACM. Pp. 37–46. 257 Kittur, Aniket, Ed Chi, Bryan A. Pendleton, Bongwon Sun and Todd Mytkowicz. 2008. "Power of the Few Vs Wisdom of the Crowd: Wikipedia and the Rise of the Bourgeoisie." Paper presented at ‘CHI 2007’, San Jose, 28 April–3 May, 2007. Knorr-Cetina, Karin. 1981. The Manufacture of Knowledge - an Essay on the Constructivist and Contextual Nature of Science. Oxford: Pergamon Press. Knorr-Cetina, Karin. 1982. "The Constructivist Programme in the Sociology of Science: Retreats or Advances?" Social Studies of Science Vol. 12: 320-324. Knorr-Cetina, Karin. 1999. Epistemic Cultures. How the Science Make Knowledge. Cambridge: Harvard University Press. Knorr-Cetina, Karin. 2005. "Complex Micro Structures- the New Terrorist Societies." Theory, Culture and Society, 22(5): 213-34. Knorr-cetina, Karin and Urs Brugger. 2000. "The Market as an Object of Attachment: Exploring Postsocial Relations in Financial Markets." In Canadian Journal of Sociology, 25(2): 141-168. Knorr-Centina, Karin and Michael Mulkay (eds.). 1983. Science Observed. Beverly Hills: Sage. Kolbitsch, Josef and Hermann Maurer. 2005. "Community Building around Encyclopaedic Knowledge." In Journal of Computing and Information Technology, 14(3): 175–190. Kolbitsch, Josef and Hermann Maurer. 2006. "The Transformation of the Web: How Emerging Communities Shape the Information we Consume." In Journal of Universal Computer Science, 12(2): 187–213. Kollock, Peter. 1999. "The Economics of on Line Cooperation: Gift and Public Goods in Cyberspace." Pp. 220–246 in M. Smith and P. Kollock (Eds.), Communities in cyberspace. London: Routledge. Konieczny, Piotr. 2009. "Governance, organization, and democracy on the Internet: The iron law and the evolution of Wikipedia." Sociological Forum, 24(1): 162–192. Konieczny, Piotr. 2010. "Adhocratic Governance in the Internet Age: A Case of Wikipedia." in Journal of Information Technology and Politics, 7(4): 263-283. Korfiatis, Nikolaos, Marios Poulos and George Bokos. 2006. "Evaluating Authoritative Sources using Social Network: an Insight from Wikipedia." Online Information Review 29(3): 252–262. Krohn, Roger. 1982. "On Gieryn on the `Relativist/Constructivist' Programme in the Sociology of Science: Naïveté and Reaction." Social Studies of Science 12: 325-328. 258 Kuznetsov, Stacey. 2006. "Motivations of contributors to Wikipedia." SIGCAS Computer Society., 36(2): 1. Kvale, Steinar. 2007. Doing Interviews. London: SAGE Publications. Laclau, Ernesto and Chantal Mouffe. 1985. Hegemony and Socialist Strategy: Towards a Radical Democratic Politics. London: Verso. Laermans, Rudi. 1992. "The Relative Rightness of Pierre Bourdieu: Some Sociological Comments on the Legitimacy of Postmodern Art, Literature and Culture." Cultural Studies vol. 6(2): 248-260. Laidlaw, James. 2000. “A Free Gift Makes No Friends.” Journal of the Royal Anthropological Institute 6, 4: 1467-96. Lamont, Michel and Lurent Thévenot. 2000. Rethinking Comparative Cultural Sociology: Repertoires of Evaluation in France and the United States. Cambridge: Cambridge University Press. Lanier, Jaron. 2006. "Digital Maoism: The hazards of the new online collectivism." The Edge 183. Lash, Scot. 1990. Sociology of Postmodernism. London: Routledge. Lash, Scot. 1994. “Reflexivity and its Doubles: Structure, Aesthetics, Community.” In Ulrich Beck, Anthony Giddens and Scot Lash, Reflexive Modernization. Stanford, CA: Stanford University Press. Lash, Scott and John Urry. 1994. Economies of signs and space. London: Sage. Latour, Bruno, and Steve Woolgar. 1986. Laboratory Life. Princeton: Princeton University Press. Latour, Bruno. 1986. “The Powers of Association.” Pp. 264-280 In John Law (ed.) Power, Action and Belief. A new sociology of knowledge? Sociological Review monograph 32. London: Routledge and Kegan Paul. Latour, Bruno. 1987. Science in Action: How To Follow Scientists and Engineers Through Society. Cambridge: Harvard University Press. Latour, Bruno. 1988. The Pasteurization of France. Harvard: Harvard University Press. Latour, Bruno. 1992. “Where are the Missing Masses? The sociology of a few mundane artifacts.” Pp. 225-258 in Wiebe Bijker and John Law (eds.) Shaping Technology/ Building Society: Studies in Sociotechnical Change. Cambridge: MIT Press. Latour, Bruno. 1993. We Have Never Been Modern. Cambridge: Harvard University Press. 259 Latour, Bruno. 1999. Pandora's Hope: Essays on the Reality of Science Studies. Cambridge: Harvard University Press. Latour, Bruno. 2004. “On Using ANT for Studying Information Systems: A (Somewhat) Socratic Dialogue.” Pp. 62–76 in Chrisanthi Avgerou, Claudio Ciborra and Frank Land (eds.) The Social Study of Information and Communication Technology: Innovations, Actors and Contexts. New York: Oxford University Press. Latour, Bruno. 2005. Reassembling the Social: An Introduction To Actor-NetworkTheory. New York: Oxford University Press. Lave, Jean and Etienne Wenger. 1991. Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press Law, John. 1991. "Introduction: Monsters, Machines and Sociotechnical Relations." Pp. 1-23. in J. Law (Ed.) A Sociology of Monsters? Essays on Power, Technology and Domination. London, Routledge. Lee, Kiljae K. and Gilbert G. Karuga. 2010. "The Role of Cognitive Conflict in OpenContent Collaboration." AMCIS 2010 Proceedings, Paper 486. (http://aisel.aisnet.org/amcis2010/486, retrieved April 15, 2011). Leithner, Andreas., Maurer-Erti Werner, Mathias Glehr, Joerg Friesenbichler, Katharina Leithner and Reinhard Windhager. 2010. "Wikipedia and Osteosarcoma: A Trustworthy Patients' Information?" In Journal of the Medical Informatics Association, 17(4): 373-374. Lessig, Lawrence. 1999. Code and Other Laws of Cyberspace. New York: Basic Books. Leuf, Bo and Ward Cunningham. 2001. The Wiki Way: Quick Collaboration on the Web. Boston: Addison-Wesley. Lueg, Christopher and Danyel Fisher. 2003. From Usenet to CoWebs: Interacting with Social Information Spaces. London: Springer. Lévi-Strauss, Claude. 1987. Introduction to the Work of Marcel Mauss. London: Routledge and Kegan Paul. Lévy, Pierre. 1997. Collective Intelligence: Mankind's Emerging World in Cyberspace. New York: Plenum Trade. Lih, Andrew 2004a. "The Foundations of Participatory Journalism and the Wikipedia Project." In Association for Education in Journalism and Mass Communication (AEJMC), AEJMC Convention, Toronto, Canada Lih, Andrew (2004b). "Wikipedia as Participatory Journalism: Reliable Sources? Metrics for evaluating collaborative media as a news resource." In 5th International Symposium on Online Journalism, Austin, Texas. (http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.117.9104, Retrived November 1, 2011). 260 Lih, Andrew (2004c). "Participatory Journalism and Asia: From Web Logs to Wikipedia." In Asian Media Information & Communications Centre, 13th Asian Media Information & Communications Centre Annual Conference: ICT & Media Inputs & Development Outcomes, Bangkok, Thailand, juillet 2004. (http://jmsc.hku.hk/faculty/alih/publications/amic-2004-Wikipedia-rc2wtitle.pdf, Retrived November 1, 2011). Lih, Andrew. 2009. The Wikipedia Revolution. New York: Hyperion. Lim, Sook. 2009. "How and Why Do College Students Use Wikipedia?" In Journal of the American Society for Information Science and Technology, 60(11): 21892202. Logan, Robert K. 2010. Understanding New Media: Extending Marshall McLuhan. New York: Peter Lang. Long, Tony. 2007. "March 10, 2000: Pop!" Wired (http://www.wired.com/science/discoveries/news/2007/03/72804, November 1, 2011). Online Retrived Lorenzen, Michel. 2006. "Vandals, Administrators, and Sockpuppets, Oh My! An Ethnographic Study of Wikipedia's Handling of Problem Behavior." MLA Forum, 5(2, http://www.mlaforum.org/volumeV/issue2/article2.html, retrieved February 14, 2011) Lynch, Michael. 1985. Art and Artifact in Laboratory Science: A Study of Shop Work and Shop Talk in a Research Laboratory. London: Routledge and Kegan Paul. Lyotard, Jean-Francois. 1984. The Post-modern Condition: a Report on Knowledge. Minnesota: University of Minnesota Press. MacKenzie, Donald. 1981. Statistics in Britain, 1865-1930: The Social Construction of Scientific Knowledge, Edinburgh: Edinburgh University Press. MacKenzie, Donald. 1990. Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance. Cambridge: MIT Press. Maehre, Jeff. 2009. "What It Means to Ban Wikipedia." College Teaching, 57(4): 229-236. Mallard, Grégoire, Michèle Lamont and Joshua Guetzkow. 2009. “Fairness as Appropriateness: Epistemological Pluralism and Peer Review in the Social Sciences and the Humanities.” Science, Technology and Human Values, 34(5): 573-606. Mannheim, Karl. 1936. Ideology and Utopia; an Introduction to the Sociology of Knowledge. London: Routledge and Paul. Markham, Annette N. 1998. Life Online. Walnut Creek: AltaMira Press. Marx, Karl. 1913. A Contribution To the Critique of Political Economy. Chicago: Kerr. 261 Mauss, Marcel. 1990. The Gift: The Form and Reason for Exchange in Archaic Societies. New York: W.W. Norton Mckiernan, Gerry. 2005. "WikimediaWorlds Part I: Wikipedia." Library Hi Tech News 22(8), Pp. 46–54. McPherson, Keith. 2006. "Wikis and Literacy Development." Teacher Librarian, 34(1): 67-69. Mead, George Herbert. 1967 [1934]. Mind, Self and Society. Chicago: University of Chicago Press. Mercer, Jean. 2007. "Wikipedia and 'Open Source' Mental Health Information." Scientific Review of Mental Health Practice, 5(1): 88-92. Merton, Robert K. 1973. The Sociology of Science: Theoretical and Empirical Investigations. Chicago: University of Chicago Press. Meyers, Peter. 2001. "Fact-Driven? Collegial? This Site Wants You." New York Times. (http://www.nytimes.com/2001/09/20/technology/fact-driven-collegialthis-site-wants-you.html, retrieved April 9, 2011). Mills, C. Wright. 1940. “Situated Actions and Vocabularies of Motive,” in American Sociological Review, V (December): 904-9 13. Mizrachi, Nissim, Israel Drori, and Renee R. Anspach. 2007. "Repertoires of trust: The practice of trust in a multinational organization amid political conflict." American Sociological Review 72(1): 143-165. Mulgan Geoff, Omar Salem and Tom Steinberg. 2005. Wide Open : Open Source Methods and Their Future Potential. London : Demos. Mulkay, Michael. 1976. "Norms and Ideology in Science." Social Science Information 15(4): 637 -56. Mulkay, Michael and Gilbert, G. Nigel. 1982. "What is the Ultimate Question? Some Remarks in Defence of the Analysis of Scientific Discourse." Social Studies of Science Vol. 12: 309-319 Murray, Hannah B. and Jason C. Miller. 2010. "Wikipedia in Court: When and How Citing Wikipedia and Other Consensus Websites is Appropriate." St. John's Law Review, Vol. 84(2). (http://ssrn.com/abstract=1502759, retrieved December 11, 2011). Nature. 2006. "Encyclopaedia Britannica and Nature: a response." Nature Press release. (http://www.nature.com/nature/britannica/eb_advert_response_final.pdf, retrieved November 21, 2011). Niederer, Sabine and José van Dijck. 2010. "Wisdom of the Crowd or Technicity of Content? Wikipedia as a Sociotechnical System." New Media and Society 12(8): 1368–1387. 262 Nix, Elizabeth M. 2010. "Wikipedia: How it Works and How it Can Work for You." History Teacher, 43(2): 259-264. Nov, Oded. 2007. "What Motivates Wikipedians?" Communications of the ACM, 50(11): 60-64. Okoli, Chitu and Wonseok Oh. 2007. "Investigating Recognition-Based Performance in an Open Content Community: A Social Capital Perspective." Information and Management, 44(3): 240-252. Oreg, Shaul and Oded Nov. 2008. "Exploring Motivations for Contributing to Open Source Initiatives: The Roles of Contribution Context and Personal Values." Computers in Human Behavior 24 (5): 2055-2073. O'reilly, Tim. (2007). "What is Web 2.0: Design Patterns and Business Models for the Next Generation of Software." Communications & Strategies 1: 17. Ortega, Felipa. 2009. Wikipedia: a Quantitative Analysis. Ph.D. Dissertation. Móstole: Universidad Rey Juan Carlos. Ortega, Felipe, and Jesus Gonzalez-Barahona. 2008. “On the Inequality of Contributions to Wikipedia." In Proceedings of the 41st Hawaiian International Conference on System Sciences (HICSS 2008. (http://docencia.etsit.urjc.es/moodle/file.php/40/Slides/S4_S5_Intro_WKP_Co mpanies/Ineq_Wikipedia.pdf, retrieved November 8, 2011). O'Sullivan, Dan. 2009. Wikipedia: A New Community of Practice? Farnham, England: Ashgate. Panciera, Katherine, Aaron Halfaker and Loren Terveen. 2009. "Wikipedians are Born, Not Made: A Study of Power Editors in Wikipedia." Proceedings of GROUP 2009: 51-60. Parry, Jonathan. 1986. "The Gift, the Indian Gift and the 'Indian Gift'." Man (N.S.), 21: 453-473. Patton, Michael. Q. 2002. Qualitative Research and Evaluation Methods, 3rd edn. Thousand Oaks, CA: Sage. Pechacek, Ann. 2007. "I Can't Live without My …: Teens' Top Ten High-Tech Gadgets and Web Sites." Young Adult Library Services, 5(2): 9-16. Pentzold, Christian. 2011. "Imagining the Wikipedia Community. What do Wikipedia authors mean when they write about their community?" New Media and Society, 13(5): 704-721. Peoples, Lee. F. 2009. "The Citation of Wikipedia in Judicial Opinions." Yale Journal of Law and Technology, 12(1). (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1272437, retrieved December 11, 2011). 263 Pickering, Andrew. 1984. Constructing Quarks: A Sociological History of Particle Physics. Chicago: University of Chicago Press. Pierce, Charles. 2012 [1878]. How to Make Our Ideas Clear. (http://www.peirce.org/writings/p119.html, retrieved December 4, 2011) Pinch, Trevor J. 1986. Confronting Nature: The Sociology of Solar-Neutrino Detection. Dordrecht: Reidel. Pinch, Trevor J. and Wiebe E. Bijker. 1987. “The Social Construction of Facts and Artifacts: Or How the Sociology of Science and the Sociology of Technology Might Benefit Each Other.” Pp. 17-50 in Wiebe E. Bijker, Thomas P. Hughes and Trevor J. Pinch (eds.) The Social Construction of Technological Systems. Cambridge: MIT Press. Pollard, Elizabeth A. 2008. "Raising the Stakes: Writing about Witchcraft on Wikipedia." History Teacher, 42(1): 9-24. Poster, Mark. 1997. "Cyberdemocracy: the Internet and the Public Sphere." Pp. 21228 In Mark Holmes (ed.) Virtual Politics: Identity and Community in Cyberspace. London: Sage Publications. Priedhorsky, Reid, Jilin Chen, Shyong Lam, Kathering Panciera, Loren Terveen, and John Riedl. 2007. “Creating, Destroying, and Restoring Value in Wikipedia.” Pp. 259–268 in Proceedings of GROUP 2007. New York: ACM. (http://www.cs.umn.edu/~reid/papers/group282-priedhorsky.pdf, retrieved September 15, 2011). Rafaeli, Sheizaf and Yaron Ariel. 2008. "Online Motivational Factors: Incentives for Participation and Contribution In Wikipedia." Pp. 243-267 in Azy Barak (Ed.) Psychological Aspects of Cyberspace: Theory, Research, Applications. Cambridge, UK: Cambridge University Press. Rafaeli, Sheizaf, Yaron Ariel and Tsahi Hayat. 2007. "Virtual Knowledge-Building Communities." Pp. 1792-1796 in G. D. Putnik and M. M. Cunha (Eds.), Encyclopedia of Networked and Virtual Organizations (Volume 3). Hershey: Idea Group. Rafaeli, Sheizaf, Tsahi Hayat and Yaron Ariel. 2009. "Knowledge building and motivations in Wikipedia: Participation as Ba” Pp. 51-69 in Francisco J. Ricardo. (Ed.) Cyberculture and New Media. Amsterdam: Rodopi Press. Rainie, Harrison, and Barry Wellman. 2012. Networked: The new social operating system. MIT press. Raymond, Eric. 1998. Homesteading the (http://www.catb.org/esr/writings/homesteading/homesteading, January 18, 2011). Noosphere. retrieved Raymond, Eric. 2012. The Jargon File, Version 4.4.8. (http://www.catb.org/jargon/, retrieved March 10, 2012) 264 Reagle, Joseph M. 2006. "Open Codex Historical Entry: Nupedia-l Archives." Open Communities, Media, Source, and Standards. (http://reagle.org/joseph/blog/method/nupedia-l-archives, retrieved April 11, 2011). Reagle, Joseph M. 2010a. Good Faith Collaboration: the Culture of Wikipedia. Cambridge: MIT Press. Reagle, Joseph M. 2010b. "'Be Nice': Wikipedia Norms for Supportive Communication." New Review of Hypermedia and Multimedia 16: 161-180. Reavley, N. J., Mackinnon, A. J., Morgan, A. J., Alvarez-Jimenez, M., Hetrick, S. E., Killackey, E., Nelson, B., Purcell, R., Yap, M. B. H., and A. F. Jorm. 2011. "Quality of Information Sources about Mental Disorders: A Comparison of Wikipedia with Centrally Controlled Web and Printed Sources." Psychological Medicine 42(8): 1-10. Reinhardt, Christian. 2003. Collaborative Knowledge Creation In Virtual Communities of Practice. A dissertation towards a Masters degree in social and economic sciences. Department of Value-Prcoess Management, Marketing. Innsbruck: University of Innsbruck. Rheingold, Howard. 2000. The Virtual Community: Homesteading on the Electronic Frontier. London: MIT Press. Rheingold, Howard. 2003. Smart mobs. No. 1. De Boeck Supérieur. Ribemont, Bernard. 1996. "On the Definition of an Encyclopaedic Genre in the Middle Ages." Pp. 47-61 in Peter Binkley (ed.) Pre-modern Encyclopaedic Texts. Leiden: Brill. Rier, David A., and Debbie Indyk. 2006. "Flexible rigidity: Supporting HIV treatment adherence in a rapidly-changing treatment environment." Social work in health care 42(3-4): 133-150. Robinson, William Sidney. 1951. "The Logical Structure of Analytic Induction." American Sociological Review 16: 812-818. Rosenberg, Daniel. 2001. "An Eighteenth-Century Time Machine: The Encyclopedia of Diderot." In Postmodernism and the Enlightenment. New York: Routledge. Rosenzweig, Roy. 2006. "Can History be Open Source? Wikipedia and the Future of the Past." Journal of American History, 93(1): 117–146. Rutter, Jason and Gregory W. H. Smith. 2005. "Ethnographic Presence in a Nebulous Setting." Pp. 81-92 in Hine, C. (Ed.) Virtual Methods: Issues in Social Research on the Internet. Oxford: Berg. Sanger, Lawrence. 2009. "The Fate of Expertise after Wikipedia." Episteme, Vol. 6, No. 1. pp. 52-73. 265 Savolainen, Reijo. 1999. "Academic Capital and Information Seeking Career." Swedish Library Research 3: 5-19. Schaffer, Simon. 1996. "Where Experiments End: Tabletop Trials in Victorian Astronomy." Pp. 257-99 in J.Z. Buchwald (ed.) Scientific Practice: Theories and Stories of Doing Physics. Chicago: University of Chicago Press. Schroer, Joachim and Guido Hertel. 2009. "Voluntary Engagement in an Open WebBased Encyclopedia: Wikipedians and Why They Do It." Media Psychology, 12(1): 96-120. Schweitzer, Nick. 2008. "Wikipedia and Psychology: Coverage of Concepts and Its Use by Undergraduate Students." Teaching of Psychology, 35(2): 81-85. Segal, Howard P. 1985. Technological Utopianism in American Culture. Chicago: University of Chicago. Sewell, William H. Jr. 1992. "A Theory of Structure: Duality, Agency and Transformation." American Journal of Sociology, 98(1): 1-29. Shalin, Dmitri. 1986. "Pragmatism and Social Interactionism." American Sociological Review 51: 9–29. Shapin, Steven and Simon Schaffer. 1985. Leviathan and the Air Pump: Hobbes, Boyle and the Experimental Life. Princeton: Princeton University Press. Shapin, Steven. 1994. A Social History of Truth: Gentility, Credibility, and Scientific Knowledge in Seventeenth-Century England. Chicago and London: University of Chicago Press. Shapin, Steven. 1995. "Here and Everywhere – Sociology of Scientific Knowledge", Annual Review of Sociology 21: 289-321. Shapin, Steven. 1996. The Scientific Revolution, Chicago: University of Chicago Press. Shirky, Clay. 2008. Here Comes Everybody: The Power of Organizing without Organizations. New York: Penguin. Shirky, Clay. 2010. Cognitive Surplus: Creativity and Generosity in a Connected Age. New York: Penguin. Shaw, Danna. 2008. "Wikipedia in the Newsroom." American Journalism Review. 30(1): 40-45. Silber, Ilana F. 1995. “Gift-Giving in the Great Traditions: The Case of Donations to Monasteries in the Medieval West.” European Journal of Sociology 36(2): 209-243. Silber, Ilana F. 2000 “Beyond Purity and Danger: Gift-Giving in the Monotheistic Religions,” pp. 115-132in T. Vandevelde (ed.) Gifts and Interests. Louvain: Peeters. 266 Silber, Ilana F. 2003. “Pragmatic Sociology as Cultural Sociology: Beyond Repertoire Theory?” European Journal of Social Theory, 6, 4: 427-449. Silber, Ilana. F. 2007. “Registres et Répertoires du Don: Avec Mais Aussi Après Mauss?” pp.124-144 in Eliana Magnani (ed.) Don et Sciences Sociales. Théories et Pratiques Croisées. Dijon: Editions de l’Université de Dijon. Silber, Ilana. F. 2011. "Deciphering Transcendence and the Open Code of Modernity: S.N. Eisenstadt’s Comparative Hermeneutics of Civilizations." Journal of Classical Sociology 11(3): 269-28. Silber, Ilana. F. 2013. “Neither Mauss, nor Veyne? Peter Brown’s Interpretative Path to the Gift.” pp. 202-220 M. Satlow (ed.) The Gift in Antiquity. Hoboken, NJ: Wiley-Blackwell. Spence, Des. 2009. "A Wicked Encyclopaedia." BMJ: British Medical Journal 339: b3814. Spoerri, Anselm. 2007. "What is Popular on Wikipedia and Why?" First Monday 4(2). (http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/1765/ 1645, retrieved April 10, 2011) Star, Susan Leigh. 2010. "This is Not a Boundary Object: Reflections on the Origin of a Concept." Science, Technology and Human Values 35: 601-617 Star, Susan Leigh and James Griesemer. 1989. “Institutional Ecology, Translations and Boundary-Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–1939.” Social Studies of Science 19(3): 387-420. Starling, Tim. 2010. "Old Wikipedia Backups Discovered." Foundation-l Mailing List (http://wikimedia.7.n6.nabble.com/Old-Wikipedia-backups-discoveredtd906781.html, retrieved December 12, 2011). Strauss, Anselm L. 1978. "A Social Worlds Perspective." Studies in Symbolic Interaction 1: 119-128. Sunstein, Cass R. 2006. Infotopia: How Many Minds Produce Knowledge. Oxford: Oxford University Press. Surowiecki, James. 2004. The Wisdom of Crowds: Why the Many are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations. USA : Doubleday. Swartz, Aaron. 2006. "Who Writes Wikipedia?" Raw Thought Blog. (http://www.aaronsw.com/weblog/whowritesWikipedia, retrieved December 3, 2011). Swidler, Ann. 1986. "Culture in Action: Symbols and Strategies." American Sociological Review, 51(2): 273-286. 267 Tapscott, Don. and Anthony D. Williams. 2006. Wikinomics: How Mass Collaboration Changes Everything. New York: Portfolio. Tarasiuk, Tracy J. 2010. "Combining Traditional and Contemporary Texts: Moving My English Class to the Computer Lab." Journal of Adolescent and Adult Literacy, 53(7): 543-552. Tarot, Camille. 2000. “Gift and Grace: A Family to be Recomposed?” pp. 133-156 in T. Vandevelde (ed.) Gifts and Interests. Louvain: Peeters. Tenner, Edward. 1996. Why Things Bite Back: Technology and the Revenge of Unintended Consequences. New York: Knopf. Thévenot, Laurent. 1984. "Rules and Implements: Investments in Form." Social Sciences Information, 23(1): 1-45. Thévenot, Laurent. 2001a. "Pragmatic Regimes Governing the Engagement with the World." Pp. 56-73 in Theodore R. Schatzki, Karin Knorr-Cetina and Eike Von Savigny (eds.) The Practice Turn in Contemporary Theory. London: Routledge. Thévenot, Laurent. 2001b. "Organized Complexity: Conventions of Coordination and the Composition of Economic Arrangements." Eurpea Journal of Social Theory 4(4): 405-425. Thévenot, Laurent. 2002a. "Conventions of Co-Ordination and the Framing of Uncertainty." Pp.181-197 in Edward Fullbrook (ed.) Intersubjectivity in Economics. London: Routledge. Thévenot, Laurent. 2002b. "Which Road to Follow? The Moral Complexity of an ‘Equipped' Humanity." Pp. 53-87 in Law, John and Mol Annemarie (eds) Complexities: Social Studies of Knowledge Practices. Durham and London: Duke University Press. Titmuss, Richard M. 1971. The Gift Relationship : From Human Blood to Social Policy. New York: the New York Press. Tkacz, Nathaniel. 2011. "The Spanish Fork: Wikipedia's Ad-Fuelled Mutiny." Wired.co.uk. (http://www.wired.co.uk/news/archive/2011-01/20/Wikipediaspanish-fork?page=all, retrieved April 9, 2011). Toffler, Alvin. 1980. The Third Wave. New York: Morrow. Touraine, Alain. 1971. The Post-Industrial Society. Tomorrow's Social History: Classes, Conflicts and Culture in the Programmed Society. New York: Random House. Traweek, Sharon. 1988. Beamtimes and Lifetimes: The World of High Energy Physicists. Cambridge: Harvard University Press. Turkle, Sherry. 1995. Life on the Screen: Identity in the Age of the Internet. NY: Simon and Schuster. 268 Turkle, Sherry. 1996. "Parallel Lives: Working on Identity in Virtual Space." Pp. 156175 in Debra Grodin and Thomas Lindlof (Eds.), Constructing the Self in a Mediated World. Thousand Oaks: Sage. Turner, Fred. 2006. From Counterculture to Cyberculture: Stewart Brand, the Whole Earth Network, and the Rise of Digital Utopianism. Chicago: University of Chicago. Turner, Victor. 1967. "Carnival, Ritual, and play in Rio de Janeiro." Pp. 74- 92 in Alessandro Falassi (Ed.) Time Out of Time: Essays on the Festival. Albuquerque: University of New Mexico Press. Turner, Victor. 1968. Schism and Continuity in an African Society. Manchester University Press. Turner, Victor. 1974. Dramas, Fields, and Metaphors: Symbolic Action in Human Society. Ithaca/London: Cornell University Press. Vandevelde, Antoon. 2000. ”Conceptual Map of Gift Practices.” pp. 1-20 in A. Vandevelde (ed.) Gifts and Interests. Louvain: Peeters. Venturini, Tommaso. 2009. "Diving in Magma: How To Explore Controversies with Actor-Network Theory." Public Understanding of Science 19(3): 258-273. Viégas, Fernanda B., Martin Wattenberg, and Matthew M. McKeon. 2007. "The Hidden Order of Wikipedia." in D. Schuler (Ed.) Online Communities and Social Computing, 234: 445-454. (http://www.research.ibm.com/visual/papers/hidden_order_Wikipedia.pdf, retrieved September 15, 2011). Voss, Jakob. 2005. "Measuring Wikipedia." In Proceedings of 10th International Conference of the International Society for Scientometrics and Informetrics, Stockholm: Juillet. Wagner, Christian. 2006. "Breaking the Knowledge Acquisition Bottleneck Through Conversational Knowledge Management." Information Resources Management Journal, 19(1): 70-83. Wagner, Peter. 2008. Modernity as Experience and Interpretation: A New Sociology of Modernity. Cambridge, UK: Polity Press. Warwick, Andrew. 1993. "Cambridge Mathematics and Cavendish Physics – Cunningham, Campbell and Einstein's Relativity, 1905-1911. Comparing Traditions in Cambridge Physics." Studies in History and Philosophy of Science 24: 1-25. Weber, Max. 1930. The Protestant Ethic and the Spirit of Capitalism. London: Allen and Unwin. Weber, Max. 1978. Economy and Society. Berkeley: Univ. of California Press. 269 Weinberger, David. 2007. Everything is miscellaneous: The power of the new digital disorder. Macmillan. Weinberger, David. 2011. Too big to know: Rethinking knowledge now that the facts aren't the facts, experts are everywhere, and the smartest person in the room is the room. Basic Books. Wellman, Barry and Caroline Haythornthwaite. 2002. The Internet in Everyday Life. Cambridge: Blackwell. Wenger, Etienne. 1998. Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press. Wilcox, Bryce. 1994. "Patronage of #AynRand Doesn't Equal Agreement (was: Of IRC Channels and Picnic Tables)." alt.philosophy.objectivism Usenet group (https://groups.google.com/d/msg/alt.philosophy.objectivism/smgN1GbY3w/OTgH1pJLwB8J, Rerieved June 7 2011) Wilson, Chris. 2008. "The Wisdom of the Chaperones: Digg, Wikipedia, and the Myth of Web 2.0 Democracy." Slate. (http://www.slate.com/articles/technology/technology/2008/02/the_wisdom_of _the_chaperones.html, retrieved August 6 2011). Wilson, Samuel M., and Leighton C. Peterson. 2002. "The Anthropology of Online Communities." Annual Review of Anthropology 31(1): 449-67. Wittrock, Bjorn. 2000. "Modernity: One, None, or Many? European Origins and Modernity as a Global Condition" Daedalus: Journal of the American Academy of Arts and Sciences 129(1): 31-60. Reprinted in Shmuel N. Eisenstadt (ed), Multiple Modernities. New Brunswick, N.J.: 2002, pp. 31-60. Wood, W. Jay. 1998. Epistemology: Becoming Intellectually Virtuous. Downers Grove: InterVarsity Press. Wray, Brad K. 2009. "The Epistemic Cultures of Science and Wikipedia: A Comparison." Episteme 6(1): 38-51. Wuthnow, Robert. 1989. Communities of Discourse: Ideology and Social Structure in the Reformation, the Enlightenment, and European Socialism. Cambridge: Harvard UP. Yonay, Yuval. 1994. "When black boxes clash: competing ideas of what science is in economics, 1924-39." Social Studies of Science 24(1):39-80. Yonay, Yuval and Daniel Breslau. 2006. "Marketing Models: The Culture of Mathematical Economics." Sociological Forum 21(3):345-385. Zickuhr, Kathryn and Lee Rainie. 2011. "Wikipedia, Past and Present." Pew Internet and American Life Project. (http://pewinternet.org/Reports/2011/Wikipedia.aspx, retrieved August 6, 2011). 270 Zlatic, Vinko, Miran Bozicevic, Hrvoje Stefancic and Mirjana Domazet. 2006. "Wikipedias: Collaborative Web-Based Encyclopedias as Complex Networks." Physical Review E, 74(1): Pp. 6–11. Zuckerman, Harriet and Robert K. Merton. 1971. "Patterns of Evaluation in Science: Institutionalization, Structure and Functions of the Referee System." Minerva 9: 66-100. 271 Appendix: Primary Resources Banerjee, Robin. 2000. "Reading Level/S and Hypertext." Nupedia-L Mailing List (http://web.archive.org/web/20030328165619, retrieved April 9, 2011). Bomis. 2006. "Bomis in Its 10th Year." Bomis. (http://web.archive.org/web/20080723231711/http://www.bomis.com/about/bo mis10year.html, retrieved June 9, 2011). Campeu, Elizabeth. 2000. "T-Shirts." Nupedia-l. (http://web.archive.org/web/20030516083921/http://www.nupedia.com/piperm ail/nupedia-l/2000-August/000560.html, retrieved June 7, 2011). Cunctator. 2001. "No Subpages. Is That Your Final Answer?" Wikipedia-l Mailing List. (http://markmail.org/message/6qbkpjqldgl4tp34, retrieved April 9, 2011). Gérard, Olivier. 2001. "Wiktionary." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-April/thread.html, retrieved April 9, 2011). Gouthro, Liane. 2000. "Building the World’s Biggest Encyclopedia." PC World Online Magazin. (www.pcworld.com/article/id,15676c,techindustrytrends/article.html, retrieved May 8, 2008). Kissane, Simon J. 2001. "Admins (I Suspect LMS) Permanently Deleting Things without Reason." Wikipedia-l Mailing List. (http://marc.info/?l=Wikipedia&m=104216623606505&w=2, retrieved April 9, 2011). Lamb, Brian. 2005. "Q and A: Jimmy Wales, Wikipedia Founder." C-SPAN (http://www.q-and-a.org/Transcript/?ProgramID=1042, retrieved April 15, 2011). Manske, Magnus. 2001. "PHP Wikipedia, Part 2." Wikipedia-l Mailing List. (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-August/thread.html, retrieved April 9, 2011). Meta-Wiki. 2001a. "Is Wikipedia anExperiment in Anarchy?" Meta-Wiki (http://meta.wikimedia.org/wiki/Is_Wikipedia_an_experiment_in_anarchy, retrieved April 11, 2011). Meta-Wiki. 2001b. "Is Wikipedia anExperiment in Anarchy/Talk." Meta-Wiki (http://meta.wikimedia.org/wiki/Talk:Is Wikipedia an experiment in anarchy, retrieved April 11, 2011). Meta-Wiki. 2001c. "Main Page." Meta-Wiki (http://meta.wikimedia.org/wiki/Main Page, retrieved April 11, 2011). Nostalgia Wikipedia. 2012. "Home Page." Nostalgia (http://nostalgia.Wikipedia.org, retrieved April 11, 2011). 272 Wikipedia Nupedia. 2000a. "Nupedia News." Nupedia (http://web.archive.org/web/20000817023708/http://www.nupedia.com/news.s html, retrieved June 1, 2012). Nupedia. 2000b. "Nupedia News." Nupedia (http://web.archive.org/web/20010607072300/www.nupedia.com/news.shtml, retrieved June 1, 2012). Nupedia. 2000c. "Nupedia.Com Editorial Policy Guidelines, Version 2.1 (May 10, 2000)." Nupedia (http://web.archive.org/web/20000619165525/http://www.nupedia.com/policy print.html, retrieved June 1, 2012). Nupedia. 2000d. "Nupedia.Com Editorial Policy Guidelines, Version 3.2 (June 23, 2000)." Nupedia (http://web.archive.org/web/20000829075115/http://www.nupedia.com/policy print.html, retrieved June 1, 2012). Nupedia. 2000e. "About Nupedia." Nupedia (http://web.archive.org/web/200012031738/http://www.nupedia.com/about.sht ml, retrieved June 1, 2012). Nupedia. 2000f. "Join Us!" Nupedia. (http://web.archive.org/web/200012040154/http://www.nupedia.com/join.shtm l, retrieved June 1, 2012). Nupedia. 2000g. "Other Frequently Asked Questions about Nupedia" Nupedia. (http://web.archive.org/web/20000829075108/http://www.nupedia.com/faq.sht ml, retrieved June 1, 2012). Nupedia. 2001a. "Nupedia: Newest Articles." Nupedia (http://web.archive.org/web/20010804024346/http://www.nupedia.com/newest .phtml, retrieved June 1, 2012) Nupedia. 2001b. "Nupedia's Proposed Category Scheme." Nupedia. (http://web.archive.org/web/200101050928/http://www.nupedia.com/category. shtml, retrieved June 1, 2012). Nupedia. 2001c. "Nupedia.Com Editorial Policy Guidelines, Version 4 (June 23, 2000)." Nupedia (http://web.archive.org/web/20010607080354/http://www.nupedia.com/policy. shtml, retrieved June 1, 2012). Nupedia. 2003. "Nupedia: Newest Articles." Nupedia. (http://web.archive.org/web/20030810153103/http://www.nupedia.com/newest .phtml, retrieved June 1, 2012). Parks, Rose. 2001. "Wiktionary." Wikipedia-l Mailing List. (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-April/thread.html, retrieved April 9, 2011). 273 Rybo, Stefan. 2001. "Policy Content and Its Navigation." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-June/thread.html, retrieved April 9, 2011). Sanger, Larry. 2000a. "Visula Arts....What about Photography, Drawing etc?" Nupedia-l Mailing List (http://web.archive.org/web/20030325121343/http://www.nupedia.com/piperm ail/nupedia-l/2000-June/000526.html, retrieved June 7, 2011). Sanger, Larry. 2000b. "Please Post This Flyer." Nupedia-l Mailing List (http://web.archive.org/web/20030515104119/http://www.nupedia.com/piperm ail/nupedia-l/2000-May/000303.html, retrieved June 7, 2011). Sanger, Larry. 2000c. "Two Big Changes." Nupedia-l Mailing List (http://web.archive.org/web/20030325114458/http://www.nupedia.com/piperm ail/nupedia-l/2000-August/000558.html, retrieved June 7, 2011). Sanger, Larry. 2000d. "Changes Summarized." Nupedia-l Mailing List (http://web.archive.org/web/20030325115949/http://www.nupedia.com/piperm ail/nupedia-l/2000-August/000561.html, retrieved June 7, 2011). Sanger, Larry. 2000e. "A New Nupedia Project." Nupedia-l Mailing List (http://web.archive.org/web/20030325122718/http://www.nupedia.com/piperm ail/nupedia-l/2000-December/000646.html, retrieved June 7, 2011). Sanger, Larry. 2000f. "Let's Make a Wiki!" Nupedia-l Mailing List (http://web.archive.org/web/20030414014355/http://www.nupedia.com/piperm ail/nupedia-l/2001-January/000676.html, retrieved June 7, 2011). Sanger, Larry. 2000g. "Need Help with These Topics." General-l Mailing List (http://web.archive.org/web/20020129183408/http://nupedia.com/pipermail/ge neral-l/2000-November/000001.html, retrieved June 7, 2011). Sanger, Larry. 2001a. "The Nupedia Chalkboard is up--Two Important Requests." Nupedia-l Mailing List (http://web.archive.org/web/20030518145848/http://www.nupedia.com/piperm ail/nupedia-l/2001-July/000936.html, retrieved June 7, 2011). Sanger, Larry. 2001b. "Peer Review Project for Wikipedia." Nupedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-January/thread.htm, retrieved June 7, 2011). Sanger, Larry. 2001c. "A Few Comments about Wikipedia." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-January/thread.html, retrieved April 9, 2011). Sanger, Larry. 2001d. "Web Links Subpage." Wikipedia-l Mailing (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-May/thread.html. retrieved April 9, 2011. 274 List Sanger, Larry. 2001e. "Wiktionary." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-April/thread.html, retrieved April 9, 2011). Sanger, Larry. 2001f. "Policy Content and Its Navigation." Wikipedial-l (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-June/thread.html, retrieved April 9, 2011). Sanger, Larry. 2001g. "Wikipedia Is Wide Open. Why Is It Growing So Fast? Why Isn't It Full of Nonsense?" Kuro5hin.com. (http://www.kuro5hin.org/story/2001/9/24/43858/2479, retrieved April 9, 2011). Sanger, Larry. 2002a. "Announcement about My Involvement in Wikipedia and Nupedia—Larry Sanger." Meta-Wiki (http://meta.wikimedia.org/w/index.php?title=Announcement_about_my_invol vement_in_Wikipedia_and_Nupedia--Larry_Sanger&oldid=1173, retrieved April 9, 2011). Sanger, Larry. 2002b. "My Resignation--Larry Sanger." Meta-Wiki (http://meta.wikimedia.org/w/index.php?title=My_resignation--Larry Sanger&oldid=23899, retrieved April 9, 2011). Sanger, Larry. 2005. "The Early History of Nupedia and Wikipedia: A Memoir." Slashdot. (http://features.slashdot.org/article.pl?sid=05/04/18/164213, retrieved January 15, 2008). Shell, Tim. 2001. "A Few Comments about Wikipedia." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-January/thread.html, retrieved April 9, 2011). Wales, Jimmy. 2000a. "Reading Level/s and Hypertext." Nupedia-l Mailing List (http://web.archive.org/web/20030328164222/http://www.nupedia.com/piperm ail/nupedia-l/2000-May/000172.html. (retrieved April 9, 2011). Wales, Jimmy. 2000b. "Two Big Changes." Nupedia-l Mailing List (http://web.archive.org/web/20030325115114/http://www.nupedia.com/piperm ail/nupedia-l/2000-August/000559.html. (retrieved April 9, 2011). Wales, Jimmy. 2001a. "A Question." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-October/000657.html. (retrieved April 9, 2011). Wales, Jimmy. 2001b. "Re: [Wikipedia-l] A Proposal for the New Software." Wikipedia-l Mailing List (http://marc.info/?l=Wikipedial&m=104216623606384&w=2. (retrieved April 9, 2011). Wales, Jimmy. 2001c. "Admins (I suspect LMS) Permanently Deleting Things without Reason." Wikipedia-l Mailing List. (http://marc.info/?l=Wikipedia&m=104216623606508&w=2, retrieved April 9, 2011). 275 Wales, Jimmy. 2001d. "Statement of Principles." Wikipedia (http://nostalgia.Wikipedia.org/w/index.php?title=Jimbo_Wales/Statement_of_ principles&oldid=75340, retrieved April 9, 2011). Wales, Jimmy. 2001e. "Refactoring Policy." Wikipedia Backups Archive Edit of january 30 by user 111, timestamp: 980815498. Wales, Jimmy. 2001f. "What's This List For." Wikipedia-l Mailing List. (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-January/thread.html, retrieved April 9, 2011). Wales, Jimmy. 2001g. "A Few Comments about Wikipedia." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001January/thread.html, retrieved April 9, 2011). Wales, Jimmy. 2001h. "Wiktionary." Wikipedia-l Mailing List. (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-April/thread.html, retrieved April 9, 2011). Wales, Jimmy. 2001i. "Britannica and Free Content." Slashdot. (http://slashdot.org/story/01/07/26/0312258/britannica-and-free-content, retrieved April 9, 2011). Wales, Jimmy. 2001j. "Human Vs. Animal." Wikipedia-l Mailing List (http://lists.wikimedia.org/pipermail/Wikipedia-l/2001-April/000097.html, retrieved April 9, 2011). Wales, Jimmy. 2005a. "Re: Wikipedia, Emergence, and the Wisdom of Crowds." Wikipedia-l Mailing List. (http://lists.wikimedia.org/pipermail/Wikipedial/2005-May/021764.html, retrieved April 20, 2011). Wales, Jimmy. 2005b. "Re: Sanger's Memoirs." Wikipedia-I Mailing List. (http://lists.wikimedia.org/pipermail/Wikipedia-l/2005-April/021463.html, Retrived November 1, 2011). Wayback machine. 2011a. (http://wayback.archive.org/web/*/http:/nupedia.com, 2011). retrieved Nuepedia, April 11, Wayback machine. 2011b. Wikipedia (http://wayback.archive.org/web/*/http:/Wikipedia.com, retrieved April 11, 2011). Wikimedia. 2005. "April 2005 Archives by Thread." Wikipedia-I Mailing List. (http://lists.Wikimedia.org/pipermail/Wikipedia-l/2005-April/thread.html, Retrived November 1, 2011). Wikimedia. 2009. "Licensing Update/Result." Wikimedia Wiki. (http://meta.wikimedia.org/w/index.php?title=Licensing_update/Result&oldid =1776040, retrieved November 10, 2011). 276 Wikimedia. 2011."Wikimedia Report Card." Wikimedia Wiki. (http://stats.wikimedia.org/reportcard/RC 2011 09 synopsis.html, Retrived November 1, 2011). Wikimedia. 2012. "Wikipedia Statistics." Wikimedia (http://stats.wikimedia.org/EN/Sitemap.htm, retrieved June 1, 2012) Wikipedia. 2001a. "Wikipedia Announcements." (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Milestones 2001&oldid=357968644, retrieved April 11, 2011). wiki. Wikipedia. Wikipedia. 2001b. "Larry's Text." Wikipedia Backups Archive (edit of March 6, timestamp: 983919610). Wikipedia. 2001c. "Atlas Shrugged." Wikipedia Backups Archive (edit of February 27, timestamp: 983243272). Wikipedia. 2001d. "September 11 Attacks." (http://en.Wikipedia.org/w/index.php?title=September attacks&oldid=331585895, retrieved April 9 2012). Wikipedia. 11 Wikipedia. 2001e. "The Wikipedia Militia/Talk." Wikipedia. (http://nostalgia.Wikipedia.org/wiki/The Wikipedia Militia/Talk, retrieved January 18 2012). Wikipedia. 2001f. "Neutral Point of View/Talk." Wikipedia. (http://nostalgia.Wikipedia.org/w/index.php?title=Neutral_point_of_view&old id=32469, retrieved April 11, 2011). Wikipedia. 2001g. "Wikipedia is not a Dictionary/Talk." Wikipedia. (http://en.Wikipedia.org/wiki/Wikipedia_talk:Wikipedia_is_not_a_dictionary/ Archive_1, retrieved April 11, 2011). Wikipedia. 2001h. "Homepage." Wikipedia Backups Archive (edit of January 17, timestamp: 979692991). Wikipedia. 2001i. "Wikipedia Process." Wikipedia Backups Archive (edit of January 21, timestamp: 980068786.) Wikipedia. 2001j. "Patent Nonsense." Wikipedia Backups Archive (edit of January 21, timestamp: 980532222). Wikipedia. 2001k. "Bad Jokes and Other Deleted Nonsense." Wikipedia Backups Archive (edit of January 21, timestamp: 9805321). Wikipedia. 2001l. "Be Bold in Updating Articles." Wikipedia Backups Archive (edit of January 17, timestamp: 979691400). Wikipedia. 2001m. "Chiristianity." Wikipedia Backups Archive (edit of January 26, timestamp: 980547425). 277 Wikipedia. 2001n. "Chiristianity/Talk." Wikipedia Backups Archive (edit of January 26, timestamp: 980549300). Wikipedia. 2001o. "Refactoring Policy." Wikipedia Backups Archive (edit of January 26, timestamp: 981229327). Wikipedia. 2001p. "Wikipedia Policy." Wikipedia Backups Archive (edit of March 8, timestamp: 984017596). Wikipedia. 2001q. "Bush talk." Wikipedia Backups Archive (edit of February 4, timestamp: 980623844). Wikipedia. 2001r. "History of United States Talk." Wikipedia Backups Archive (edit of January 30, timestamp: 980815498). Wikipedia. 2001s. "Partake in Discussion/Talk." Wikipedia Backups Archive (edit of February 12, timestamp: 982022383). Wikipedia. 2001t. "History of United States Discussion." Wikipedia Backups Archive (edit of January 30, timestamp: 980815181). Wikipedia. 2001u. "Altruism/Talk." Wikipedia Backups Archive (edit of March 2, timestamp: 983567914). Wikipedia. 2001v. "Altruism." Wikipedia Backups Archive (edit of March 2, timestamp: 980118691). Wikipedia. 2001w. "History of United States." Wikipedia Backups Archive (edit of january 29, timestamp: 980754765). Wikipedia. 2001x. "Libertarianism." Wikipedia Backups Archive (edit of january 27, timestamp: 980555271). Wikipedia. 2001y. "Anarchy." Wikipedia Backups Archive (edit of February 2, timestamp: 981082841). Wikipedia. 2001z. "Traditional Anarchism." Wikipedia Backups Archive (edit of February 13, timestamp: 982045929). Wikipedia. 2001aa. "Democracy." Wikipedia Backups Archive (edit of February 11, timestamp: 981909946). Wikipedia. 2001ab. "George W. Bush." Wikipedia Backups Archive (edit of January 21, timestamp: 980371261). Wikipedia. 2001ac. "History of United States." Wikipedia Backups Archive (edit of january 29, timestamp: 980888645). Wikipedia. 2001ad. "Rules to Consider." Wikipedia Backups Archive (edit of February 4, timestamp: 981308293). 278 Wikipedia. 2001ae. "Football." Wikipedia Backups Archive (edit of January 24, timestamp: 980355564). Wikipedia. 2001af. "Capitalism." Wikipedia Backups Archive (edit of February 2, timestamp: 981084054). Wikipedia. 2001ag. "American Football." Wikipedia Backups Archive (edit of February 8, timestamp: 981674606). Wikipedia. 2001ah. "Anarchocapitalism." Wikipedia Backups Archive (edit of February 13, timestamp: 982045747). Wikipedia. 2001ai. "Anarchy/Talk." Wikipedia Backups Archive (edit of February 13, timestamp: 982046534). Wikipedia. 2001aj. "Traditional Anarchism/Talk." Wikipedia Backups Archive (edit of February 15, timestamp: 982280760). Wikipedia. 2001ak. "Neutral Point of View." Wikipedia Backups Archive (edit of February 16, timestamp: 982358834). Wikipedia. 2001al. "Creationism." Wikipedia Backups Archive (edit of April 4 16, timestamp: 986411388). Wikipedia. 2001am. "Creationism/Talk." Wikipedia Backups Archive (edit of April 4, timestamp: 986426281). Wikipedia. 2001an. "Creationism/Talk." Wikipedia Backups Archive (edit of April 5, timestamp: 986513055). Wikipedia. 2001ao. "Tim Sehll." Wikipedia Backups Archive (edit of January 28, timestamp: 980653562). Wikipedia. 2001ap. "Larry Sanger/Columns." Wikipedia Backups Archive (edit of June 10, timestamp: 992171611). Wikipedia. 2001aq. "Wiki is not an Encyclopedia." Wikipedia Backups Archive (edit of March 19, timestamp: 985033332). Wikipedia. 2001ar. "Wiki is not Paper." Wikipedia Backups Archive (edit of March 3, timestamp: 985041337). Wikipedia. 2001as. "Wikipedia is not a Dictionary." Wikipedia Backups Archive (edit of May 21, timestamp: 990449593). Wikipedia. 2001at. "Public Domain Resources." Wikipedia Backups Archive (edit of March 28, timestamp: 985793997). Wikipedia. 2001au. "Public Domain Resources/Talk." Wikipedia Backups Archive (edit of March 28, timestamp: 985801520). 279 Wikipedia. 2001av. "Drunk." Wikipedia Backups Archive (edit of April 17, timestamp: 987467255). Wikipedia. 2001aw. "Drunk/Talk." Wikipedia Backups Archive (edit of April 17, timestamp: 987488749). Wikipedia. 2001ax. "Encyclopedia and or Versus Dictionary." Wikipedia Backups Archive (edit of April 17, timestamp: 987531878) Wikipedia. 2001ay. "Monarch." Wikipedia Backups Archive (edit of May 10, timestamp: 989485414). Wikipedia. 2001az. "Ship." Wikipedia Backups Archive (edit of May 10, timestamp: 989511423). Wikipedia. 2001ba. "Philosopher." Wikipedia Backups Archive (edit of May 10, timestamp: 989515365). Wikipedia. 2001bb. "Comic-strip." Wikipedia Backups Archive (edit of May 21, timestamp: 990438237). Wikipedia. 2001bc. "WIPO." Wikipedia Backups Archive (edit of May 17, timestamp: 990119789). Wikipedia. 2001bd. "University of Oxford." Wikipedia Backups Archive (edit of May 19, timestamp: 990313235). Wikipedia. 2001be. "Wikipedia." Wikipedia Backups Archive (edit of May 16, timestamp: 990012645). Wikipedia. 2001bf. "Wikipedia Bugs." Wikipedia Backups Archive (edit of May 16, timestamp: 990015686). Wikipedia. 2001bg. "LA2." Wikipedia Backups Archive (edit of May 10, timestamp: 989510096). Wikipedia. 2001bh. "Open Source Initiative." Wikipedia Backups Archive (edit of May 16, timestamp: 990014026). Wikipedia. 2001bi. "Rules to Consider." Wikipedia Backups Archive (edit of May 17, timestamp: 990096406). Wikipedia. 2001bj. "Trojan Horse." Wikipedia Backups Archive (edit of May 18, timestamp: 990214042). Wikipedia. 2001bk. "Hacking." Wikipedia Backups Archive (edit of May 20, timestamp: 990333626). Wikipedia. 2001bl. "Boston." Wikipedia Backups Archive (edit of May 20, timestamp: 990334165). 280 Wikipedia. 2001bm. "Larry Sanger." Wikipedia Backups Archive (edit of May 20, timestamp: 990335720). Wikipedia. 2001bn. "Scheme." Wikipedia Backups Archive (edit of May 21, timestamp: 990443645). Wikipedia. 2001bo. "Scheme/Talk." Wikipedia Backups Archive (edit of May 21, timestamp: 990449079). Wikipedia. 2001bp. "Am." Wikipedia Backups Archive (edit of May 21, timestamp: 990448977). Wikipedia. 2001bq. "LA2." Wikipedia Backups Archive (edit of May 21, timestamp: 990449323). Wikipedia. 2001br. "Wikipedia is a Dictionary." Wikipedia Backups Archive (edit of May 21, timestamp: 990450747). Wikipedia. 2001bs. "Larry Sanger/Why a List of the Senses of a Word Is Not anEncyclopedia Article." Wikipedia Backups Archive (edit of June 10, timestamp: 992171922). Wikipedia. 2001bt. "Disease." Wikipedia Backups Archive (edit of June 8, timestamp: 992003426). Wikipedia. 2001bu. "Of." Wikipedia Backups Archive (edit of June 9, timestamp: 992107877). Wikipedia. 2001bv. "Basis." Wikipedia Backups Archive (edit of June 30, timestamp: 993921669). Wikipedia. 2001bw. "Simon Flexner." Wikipedia Backups Archive (edit of June 29, timestamp: 993807262). Wikipedia. 2001bx. "Larry Sanger/The Perfect Stub Article." Wikipedia Backups Archive (edit of July 2, timestamp: 994057581). Wikipedia. 2001by. "Welcome, Newcomers." Wikipedia Backups Archive (edit of July 28, timestamp: 996306037). Wikipedia. 2001bz. "Wikipedia FAQ." Wikipedia Backups Archive (edit of July 26, timestamp: 996121421). Wikipedia. 2001ca. "Wikipedia Policy." Wikipedia Backups Archive (edit of July 28, timestamp: 996307590). Wikipedia. 2001cb. "What Wikipedia is not." Wikipedia Backups Archive (edit of July 28, timestamp: 996307233). Wikipedia. 2001cc. "How To Destroy Wikipedia." Wikipedia (http://en.Wikipedia.org/w/index.php?title=User:The_Cunctator/How_to_destr oy_Wikipedia&oldid=471104385, Retrived April 9, 2011). 281 Wikipedia. 2001cd. "Wikipedia Utilities: find or fix a stub." Wikipedia (http://nostalgia.Wikipedia.org/wiki/Wikipedia_utilities/find_or_fix_a_stub, retrieved January 18 2012). Wikipedia. 2001ce. "Wikipedia Utilities/Pages Needing Attention." Wikipedia (http://nostalgia.Wikipedia.org/wiki/Wikipedia_utilities/Pages_needing_attenti on, retrieved January 18 2012). Wikipedia. 2001cf. "Countries of the World." Wikipedia Backups Archive (edit of June 5, timestamp: 991701864). Wikipedia. 2001cg. "Philosophy." Wikipedia Backups Archive (edit of June 4, timestamp: 991617660). Wikipedia. 2001ch. "Music." Wikipedia Backups Archive (edit of June 7, timestamp: 991940145). Wikipedia. 2001ci. "Sport." Wikipedia Backups Archive (edit of June 6, timestamp: 991833397). Wikipedia. 2001cj. "Psychiatry." Wikipedia Backups Archive (edit of June 11, timestamp: 992254064). Wikipedia. 2001ck. "Naming Conventions/Talk." Wikipedia Backups Archive (edit of June 9, timestamp: 992107672). Wikipedia. 2001cl. "Dewey Decimal System." Wikipedia Backups Archive (edit of June 14, timestamp: 992505188). Wikipedia. 2001cm. "Wikipedia NEWS." Wikipedia Backups Archive (edit of June 22, timestamp: 993181960). Wikipedia. 2001cn. "Wikipedia Help Desk." Wikipedia Backups Archive (edit of June 29, timestamp: 993806785). Wikipedia. 2001co. "Larry Sanger/List-o-Links." Wikipedia Backups Archive (edit of July 7, timestamp: 994328588). Wikipedia. 2001cp. "Unamerican Activities/Talk." Wikipedia Backups Archive (edit of July 21, timestamp: 993136172). Wikipedia. 2001cq. "Abortion/Talk." Wikipedia Backups Archive (edit of July 5, timestamp: 994318498). Wikipedia. 2001cr. "Literary Quotations/Talk." Wikipedia Backups Archive (edit of July 3, timestamp: 994163741). Wikipedia. 2001cs. "Rules to Consider." Wikipedia Backups Archive (edit of June 29, timestamp: 993814566). Wikipedia. 2001ct. "Computer." Wikipedia Backups Archive (edit of July 7, timestamp: 994340062). 282 Wikipedia. 2001cu. "Requested Articles." Wikipedia (http://nostalgia.Wikipedia.org/wiki/Requested_articles, retrieved January 18 2012). Wikipedia. 2001cv. "Nazi Germany." Wikipedia Backups Archive (edit of June 1, timestamp: 991434648). Wikipedia. 2001cw. "Nazi Germany/Related Terms." Wikipedia Backups Archive (edit of July 23, timestamp: 995915052). Wikipedia. 2001cx. "Nazi Germany/Talk." Wikipedia Backups Archive (edit of July 24, timestamp: 995961275). Wikipedia. 2001cy. "W00t." Wikipedia Backups Archive (edit of July 28, timestamp: 996306649). Wikipedia. 2001cz. "Woot/Talk." Wikipedia Backups Archive (edit of July 28, timestamp: 996306898). Wikipedia. 2001da. "Two Main Theories of History." Wikipedia Backups Archive (edit of July 27, timestamp: 996236645). Wikipedia. 2001db. "Wikipedia Commentary." Wikipedia Backups Archive (edit of March 26, timestamp: 985633655). Wikipedia. 2001dc. "Wikipedia Commentary." Wikipedia Backups Archive (edit of July 27, timestamp: 996236953). Wikipedia. 2001dd. "Yet Another." Wikipedia Backups Archive (edit of July 30, timestamp: 996529973). Wikipedia. 2001de. "What WIkipedia is not." Wikipedia. (http://nostalgia.Wikipedia.org/w/index.php?title=What_Wikipedia_is_not&ol did=55404, retrieved January 18 2012). Wikipedia. 2001df. "Give the Author a Chance Debate." Wikipedia Backups Archive (edit of February 16, timestamp: 982367290). Wikipedia. 2001dg. "How to Start a Page." Wikipedia Backups Archive (edit of April 12, timestamp: 987117723). Wikipedia. 2001dh. "Welcome, Newcomers." Wikipedia Backups Archive (edit of February 16, timestamp: 992408267). Wikipedia. 2001di. "Abortion/Talk." Wikipedia Backups Archive (edit of July 5, timestamp: 994318498). Wikipedia. 2001dj. "Scientology/Talk." Wikipedia Backups Archive (edit of April 5, timestamp: 986490351). Wikipedia. 2001dk. "Racism/Talk." Wikipedia Backups Archive (edit of May 21, timestamp: 990433597). 283 Wikipedia. 2001dl. "John Stuart Mill/Talk." Wikipedia Backups Archive (edit of June 10, timestamp: 992174329). Wikipedia. 2001dm. "Unamerican Activities/Talk." Wikipedia Backups Archive (edit of June 21, timestamp: 993136093). Wikipedia. 2002a. "Wikipedia Announcements." (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Milestones 2002&oldid=277387993, retrieved April 11, 2011). Wikipedia. Wikipedia. 2002b. "Wikipedia:Traffic." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Traffic&oldid=192123, retrieved April 15, 2011). Wikipedia. 2003. "Wikipedia:What Wikipedia is not: Difference Between Revisions." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:What Wikipedia is not&diff=next&oldid=715637, retrieved April 9, 2011). Wikipedia. 2006. "User:Raul654/Raul's Laws." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=User:Raul654/Raul%27s laws&oldid=44834502, Retrived April 9, 2011). Wikipedia. 2012a. "Ignore all Rules." (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Ignore rules&oldid=467661261, Retrived January 18, 2012). Wikipedia. all Wikpedia. 2012b. "Wikipedia: Editing Policy." Wikipedia. (http://en.Wikipedia.org/wiki/Wikipedia:IMPERFECT, Retrived January 18, 2012). Wikipedia. 2012c. "Arbitration Committee (English Wikipedia)." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Arbitration Committee %28English Wikipedia%29&oldid=445392876, retrieved January 18 2012). Wikipedia. 2012d. "History of Wikipedia." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=History_of_Wikipedia&oldid=471 890325, retrieved January 18 2012). Wikipedia. 2012e. "Ben Kovitz." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=User:BenKovitz&oldid=45619583 4, Retrived January 18, 2012). Wikipedia. 2012f. "Wikipedia: Neutral Point of View." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Neutral point of view&oldid=481531601, retrieved April 9, 2012). Wikipedia. 2012g. "Wikipedia: What Wikipedia is not." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:What Wikipedia is not&oldid=485526418, retrieved April 9, 2012). 284 Wikipedia. 2012h. "Wikipedia: What Wikipedia is not: Revision History." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:What_Wikipedia_is_not &dir=prev&limit=500&action=history, retrieved April 9, 2012). Wikipedia. 2012i. "Wikipedia:Verifiability." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Verifiability&oldid=479 894582, retrieved April 9, 2012). Wikipedia. 2012j. "Wikipedia: No Original Research." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:No_original_research& oldid=483422922, retrieved April 9, 2012). Wikipedia. 2012k. "Wikipedia: Core Content Policies." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Core_content_policies& oldid=466952275, retrieved April 9, 2012). Wikipedia. 2012l. "Wikipedia: Identifying Reliable Sources." Wikipedia. (http://en.Wikipedia.org/w/index.php?title=Wikipedia:Identifying_reliable_sou rces&oldid=480005796, retrieved April 9, 2012). 285 ‫תקציר‬ ‫מחקר זה עוסק בתהליכי התהוות וההתמסדות של ויקיפדיה כזירה מרכזית שבה מתכוננות‬ ‫פרקטיקות ואמות מידה ליצירה והערכה של ידע בתקופתנו‪ .‬ויקיפדיה היא אנציקלופדיה מקוונת‬ ‫רב‪-‬לשונית שתכניה מיוצרים על ידי המוני מתנדבים מכל רחבי העולם‪ ,‬וניתן להשתמש בהם או‬ ‫לערוך אותם כמעט ללא מגבלות‪ .‬במהלך העשור האחרון‪ ,‬מתנדבים אלו צברו וארגנו כמויות‬ ‫חסרות תקדים של תוכן אנציקלופדי והפכו את ויקיפדיה למקור ידע מרכזי בחברת המידע‬ ‫הגלובאלית‪ .‬יחד עם זאת‪ ,‬הסתמכותה על עבודה התנדבותית וחובבנית ללא בקרה מקדימה‪,‬‬ ‫מוסיפה לעורר ספקות רחבים לגבי מהימנותה והלגיטימיות שלה‪ .‬במסגרת עבודה זו אציע ניתוח‬ ‫היסטורי ופרגמטי של המנגנונים והפרקטיקות שעורכיה ומשתמשיה של ויקיפדיה פיתחו על מנת‬ ‫להתמודד עם בעיה זו של לגיטימציה‪.‬‬ ‫מחקר זה בוחן את מיסודה והתקבלותה של ויקיפדיה כצורה ייחודית וחדשנית של‬ ‫תרבות אפיסטמית – כלומר‪ ,‬כמערך של פרקטיקות ואמות מידה העומדות בבסיס היצירה‬ ‫והלגיטימציה של ידע בהקשר חברתי מסוים )‪ .(Knorr-Cetina, 1999‬ככזאת‪ ,‬היא מהווה מקרה‬ ‫אסטרטגי המאפשר לנו לבחון כיצד תרבויות אפיסטמיות חדשות מכוננות‪ ,‬נידונות וממוסדות‬ ‫בפועל‪ .‬תרומתו העיקרית של מחקר זה טמונה ביכולתו לפצות על שתי חולשות בולטות בגוף‬ ‫המחקר המתפתח במהירות אודות ויקיפדיה‪ .‬הראשונה מביניהן היא הנטיה להסביר את הצלחתה‬ ‫כתוצאה ממאפיינים מסוימים‪ ,‬הנתפשים כמהותיים לויקיפדיה )כגון הבסיס הטכנולוגי שלה‪,‬‬ ‫ההנחות התרבותיות שבבסיסה‪ ,‬ארגונה החברתי או צורת הממשל שלה(‪ ,‬תוך הזנחת היבטיהם‬ ‫ההיסטוריים והדינמיים‪ ,‬והתעלמות מכך שהם מהווים הישגים שיש להסבירם‪ .‬השניה היא‬ ‫ההזנחה היחסית של אופן ההתקבלות של תכניה ודרכי השימוש בהם‪ ,‬הגורמת להתעלמות‬ ‫מהאופן שבו בעיית הלגיטימציה שלה משפיעה על צריכתה בפועל‪.‬‬ ‫בניגוד לנטיה הא‪-‬היסטורית השוררת במחקר הקיים‪ ,‬מחקר זה מפנה את תשומת הלב‬ ‫לתהליכים המותנים ותלויי ההקשר אשר ליוו את יצירתם של מנגנונים וקווי מדיניות המאפיינים‬ ‫את ויקיפדיה עד היום‪ .‬בנוסף‪ ,‬מחקר זה מוסיף על הספרות הקיימת על ידי בחינת האופן שבו‬ ‫קבוצה מסוימת של מומחי ידע‪ ,‬עיתונאים‪ ,‬שופטים את התוצרים של תהליכים אלו ומשתמשים‬ ‫בהם בפועל‪.‬‬ ‫ברמה התיאורטית‪ ,‬עבודה זו נסמכת בראש ובראשונה על המסורת הפרגמטיסטית‪ ,‬המדגישה את‬ ‫החשיבות של ניתוח פרקטיקות ואובייקטים בזמן ובמרחב קונקרטיים על חשבון אמונות או‬ ‫א‬ ‫ערכים מופשטים; את האופן שבו אובייקט הידע והכלים המשמשים לחקירתו מעוצבים בהדרגה‬ ‫תוך שהם משנים זה את אופיו של זה; ואת הקיום של פאזות פעולה שונות‪ ,‬אשר לעתים מונחות‬ ‫על ידי הרגלים ממוסדים ולעתים על ידי הצורך למצוא פתרונות יצירתיים לבעיות קונקרטיות‬ ‫)‪ .(Joas, 1993, 1996; Gross, 2007‬גישת האינטראקציה הסימבולית‪ ,‬ובפרט הגותו של אנסלם‬ ‫שטראוס )‪ (Strauss, 1978‬הקרובה ברוחה למסורת הפרגמטיסטית‪ ,‬מפנים את תשומת הלב‬ ‫לאופן שבו מפגש בין עולמות חברתיים במסגרת של זירות מפגש דורש עבודת גבול‪ :‬דיון ועימות‬ ‫סביב נושאים במחלוקת בין נציגים של עולמות חברתיים שונים‪ ,‬אשר כרוך במאבקים על כוח‪,‬‬ ‫סמכות ולגיטימיות‪ .‬הקשר תיאורטי רלוונטי נוסף הינו התחום של לימודי מדע‬ ‫וטכנולוגיה )‪ ,(STS‬שם כולל למחקרים היסטוריים וסוציולוגיים השואפים להסביר תהליכים של‬ ‫ייצור ידע "מבפנים" – כלומר‪ ,‬תוך מעקב מדוקדק אחר פרקטיקות ייצור הידע באתרים בהם הוא‬ ‫נוצר‪ .‬חשוב במיוחד בהקשר זה הדיון במושגים הבאים‪ :‬המושג תרבות אפיסטמית )‪Knorr-‬‬ ‫‪ ,(Cetina, 1999‬שממקד את תשומת הלב בהיבטים המקומיים והייחודיים של אתרי ייצור ידע;‬ ‫המושג נקודות מעבר הכרחיות )‪ (Callon, 1986 ,Obligatory Passage Points‬המתייחס לצמתים‬ ‫המהווים אמצעים לייצור סמכות ולגיטימציה באתרים של ייצור ידע; ו‪-‬חפצי גבול ) ‪Boundary‬‬ ‫‪ ,objects‬ראו ‪ ,(Star and Griesemer, 1989‬מושג שמציין אמצעים לתיאום בין צרכי המידע של‬ ‫עולמות חברתיים שונים בנקודות המפגש ביניהם‪ .‬לבסוף‪ ,‬הסוציולוגיה הפרגמטית הצרפתית‬ ‫)‪ (Boltanski and Thévenot, 2006; Thévenot, 2001a‬מוסיפה על הגישות שנסקרו את‬ ‫החשיבות של היבטים תרבותיים ומוסריים במצבים קריטיים בהם אין עקרון מארגן ברור‬ ‫ומוסכם מראש לשיפוט הפעולות של שחקנים‪ .‬גורמים אלו באים לידי ביטוי בתהליכים של‬ ‫השקעה צורנית )‪ – (form investment, Thévenot, 1984‬כלומר יצירת הקבלות בין אובייקטים‪,‬‬ ‫פרקטיקות ואנשים המעורבים במצב‪ ,‬לבין אחרים המרוחקים ממנו בזמן ובמרחב‪ ,‬תהליכים אשר‬ ‫מאפשרים תיאום ושיתוף פעולה בין השחקנים במצבים שאינם מוגדרים באופן ברור‪.‬‬ ‫בנוסף‪ ,‬המחקר נוגע לשני נושאים תיאורטיים נוספים‪ :‬הסוציולוגיה של הנתינה ושאלת יחסה של‬ ‫ויקיפדיה למודרניות‪ .‬בהקשר הראשון‪ ,‬הספרות על ויקיפדיה מעלה מתח בלתי פתור‪ :‬בעוד חלק‬ ‫מהחוקרים מדגישים את ההסתמכות של ויקיפדיה על תוכן שנתרם ללא ציפייה לתמורה או‬ ‫להכרה‪ ,‬חוקרים אחרים מדגישים את החשיבות של מערכת המוניטין הקהילתית המבוססת על‬ ‫הכרה הדדית בנתינה לפרוייקט‪ .‬סקירת הספרות בנושא מראה כי מתח זה משקף מחלוקות‬ ‫ב‬ ‫רחבות יותר בספרות אודות נתינה ומעלה את הצורך לבחון כיצד העירוב בין סוגי נתינה שונים‬ ‫משפיע על ויקיפדיה‪ .‬בהקשר השני‪ ,‬אני מנסה לבחון כיצד ניתן למקם את התרבות האפיסטמית‬ ‫של ויקיפדיה במתח בין תיאוריות של מודרניות מאוחרת‪ ,‬המדגישות את ההמשכיות בין תפיסות‬ ‫הידע העכשוויות לבין אלו המודרניות‪ ,‬ובין תיאוריות פוסטמודרניות המדגישות את הנתק‬ ‫ביניהם‪.‬‬ ‫בפרק השני מתוארות שיטות המחקר‪ .‬עבודה זו נסמכת על שני בסיסי נתונים מובחנים‪ ,‬שנאספו‬ ‫בשנים ‪ :2009-2011‬הראשון מבניהם שימש כדי לבחון את תהליכי היווצרותם ומיסודם של‬ ‫מאפייניה של ויקיפדיה ושל קוי המדיניות המרכזיים שלה בהקשרם ההיסטורי‪ ,‬והשני כדי לבחון‬ ‫כיצד התרבות האפיסטמית שנוצרה במסגרתה משתלבת במסגרת חיי היומיום של צרכניה‪,‬‬ ‫ולאמוד את מידת ואופן השפעתה החברתית‪ .‬בסיס הנתונים הראשון נבנה על ידי בחינה של‬ ‫ארכיבי רשימות התפוצה של ויקיפדיה )וקודמתה‪ ,‬ניופדיה – ראו להלן( ושל כלל העריכות שנעשו‬ ‫בה בחודשים הראשונים לקיומה‪ ,‬כמו גם עריכות נבחרות מתקופות מאוחרות יותר וכן תכנים‬ ‫אינטרנטיים נוספים שנקשרו לתהליכי המיסוד הרלוונטיים‪ .‬בסיס הנתונים השני כלל עשרים‬ ‫ראיונות מובנים למחצה עם עיתונאים אודות יחסם לוויקיפדיה והאופן בו הם משתמשים בה‬ ‫בחיי היומיום ובעבודה שלהם‪.‬‬ ‫ניתוח בסיס הנתונים הראשון נעשה בעיקר בשיטת "מיפוי המחלוקות" ) ‪"cartography of‬‬ ‫‪ ,(controversies", Venturini, 2009‬הקשורה לגישת ‪ .ANT‬על פי שיטה זו‪ ,‬על החוקר למפות את‬ ‫העמדות ורשתות השחקנים המעורבות במחלוקות בדבר ידע‪ ,‬מדע וטכנולוגיה בטרם נושא‬ ‫המחלוקת מתייצב ונסגר‪ ,‬על מנת לבחון את התופעה הנחקרת בתהליך ההיווצרות שלה‪ ,‬בטרם‬ ‫היא הופכת למובנת מאליה‪ .‬כמו כן‪ ,‬גישה זו מדגישה את הפרקטיקות ואת האובייקטים‬ ‫המעורבים בתהליכים אלו על פני הפנומנולוגיה של השחקנים האנושיים המעורבים בהם‪ .‬בהתאם‬ ‫לשיטה זו עקבתי אחרי "חוט אריאדנה" )‪ (Latour, 1993, 121‬בדמות שרשראות ההפניות‬ ‫)‪ ,(chains of references‬קדימה ואחורה בזמן‪ ,‬דבר שדרש את שילובם של היבטים ומודלים‬ ‫היסטוריים בתיאור תהליך ההתמסדות של ויקיפדיה‪.‬‬ ‫בסיס הנתונים השני שימש כדי לשחזר את הדרכים שבהן עיתונאים מחפשים אחר מידע‬ ‫באינטרנט ואת האופן שבו ויקיפדיה משתלבת בתהליך זה‪ .‬ברוחם של מחקרים איכותניים אודות‬ ‫האינטראקציה אדם‪/‬מחשב‪ ,‬ממצאי הראיונות נותחו על מנת להבין כיצד אנשים )עיתונאים‪,‬‬ ‫במקרה זה( מבנים את הטכנולוגיות בהן הם משתמשים באמצעות הפרקטיקות היומיומיות‬ ‫ג‬ ‫שלהם וכיצד הם באים באינטראקציה איתן‪ ,‬באופן שמשנה הן את הפרקטיקות והזהות שלהם‬ ‫עצמם והן את הטכנולוגיה הרלוונטית באופנים יצירתיים ובלתי צפויים‪ .‬כבר בשלבי איסוף המידע‬ ‫המוקדמים התברר שישנו דמיון גדול מאוד בין התיאורים שסיפקו המרואיינים‪ ,‬ולא נמצאו‬ ‫הבדלים שיטתיים ביניהם‪ .‬לפיכך‪ ,‬נעשה שימוש בתהליך של אינדוקציה אנליטית ) ‪Robinson,‬‬ ‫‪ (1951; Denzin, 2007‬על מנת לבנות מודל מותנה בזמן )‪ (time-bound‬של דפוס הפעולות‬ ‫האופייני המתרחש בזמן השימוש של עיתונאים בוויקיפדיה‪ ,‬שעודן ככל שהתוסף מידע מראיונות‬ ‫נוספים‪.‬‬ ‫הפרק האמפירי הראשון )פרק ‪ (3‬בוחן את מקורותיה של ויקיפדיה בפרויקט אנציקלופדי‬ ‫קודם בשם ניופדיה‪ ,‬שהוקם על ידי אותם המייסדים‪ ,‬לארי סאנגר )‪ (Larry Sanger‬וג'ימי ויילס‬ ‫)‪ ,(Jimmy Wales‬תוך התמקדות בשני נושאים מרכזיים‪ :‬אופן ייצור הידע הנהוג בה‪ ,‬והאופן שבו‬ ‫הומשג בה תוכן אנציקלופדי ראוי לפרסום‪ .‬ניופדיה תוכננה כאנציקלופדיית אינטרנט פתוחה אך‬ ‫מנוהלת ומפוקחת על ידי מומחים‪ ,‬שהיו אמורים לערוך ולשפוט את תכניה‪ .‬מייסדיה ניסו באופן‬ ‫מודע ומכוון להעניק לגיטימציה לפרויקט על ידי ארגונו באופן המבוסס על צורות תרבותיות‬ ‫ממוסדות ועל מנגנונים מוכרים חברתית לייצור ידע‪ :‬צורתם של תכניה והמבנה שלה התבססו‬ ‫באופן ישיר על מהדורת הדפוס של האנציקלופדיה בריטניקה‪ ,‬התרומה לה הותנתה בחשיפת‬ ‫הזהות האמיתית של המשתתפים‪ ,‬ייצור הידע במסגרתה התבסס בעיקרו על תהליך שיפוט‪-‬‬ ‫עמיתים )‪ (peer-review‬למאמרים הדומה מאוד לזה הנהוג בכתבי עת אקדמיים‪ ,‬ומדרג הסמכות‬ ‫בה התבסס ככלל על היררכיה של תארים אקדמיים‪ .‬נסיון זה נכשל בעיקרו‪ ,‬מכיוון שלאחר שנה‬ ‫של פעילות ייצרו משתתפיה של ניופדיה רק תריסר ערכים שנמצאו ראויים לפרסום‪ .‬כישלון זה‬ ‫יוחס על ידי מייסדיה )ובעיקר על ידי ג'ימי ויילס( לשמרנות של ההנחות האפיסטמיות ושיטות‬ ‫ייצור הידע שלה‪ .‬לפיכך‪ ,‬לאחר מספר נסיונות כושלים להגמיש שיטות אלו ולהאיץ את קצב‬ ‫הייצור‪ ,‬החליטו סאנגר וויילס ליצור פרויקט אנציקלופדי משלים השונה באופיו במידה קיצונית‪.‬‬ ‫הבסיס הטכנולוגי של פרויקט זה היה מערכת ניהול תוכן אינטרנטית בשם "ויקי"‪ ,‬אשר באותה‬ ‫התקופה זכתה לפופולאריות מסוימת בחוגים החברתיים של מומחי מחשבים רדיקאליים‬ ‫)"‪ ("extreme programmers‬בחוף המערבי‪ .‬לפיכך‪ ,‬סאנגר העניק לו את השם "ויקיפדיה"‪ ,‬אותו‬ ‫הגדיר מאוחר יותר כ"שם מטופש למה שהיה בראשיתו פרויקט מטופש מאוד"‪.‬‬ ‫הפרק הבא )פרק ‪ (4‬מתאר בקצרה את ההיסטוריה של ויקיפדיה בשנתה הראשונה‪ ,‬כרקע‬ ‫לפרקים הבאים‪ ,‬ואת הקשרים בינה ובין ניופדיה‪ .‬בפרק מתוארים התהליכים שהובילו להחלטה‬ ‫ד‬ ‫על הקמתה של ויקיפדיה; העלייה הדרמטית במספר המשתתפים בה בעקבות מספר כתבות‬ ‫שהתפרסמו באתרי אינטרנט בולטים; ושינויים במבנה הסמכות‪ ,‬בקוי המדיניות‪ ,‬בכלים‬ ‫הטכנולוגיים ובתהליך ייצור התוכן שהתרחשו בתקופה זאת‪ .‬כפי שעולה מתיאור זה‪ ,‬ויקיפדיה‬ ‫עמדה בראשית דרכה במערכת יחסים מורכבת מול ניופדיה‪ .‬בהקשרים מסויימים‪ ,‬היא התעצבה‬ ‫בניגוד מוחלט לשאיפות למכובדות וללגיטימציה שאפיינו את ניופדיה‪ :‬יוצריה הציעו אופן ייצור‬ ‫חדש המבוסס על מודל של השתתפות ושיתוף פעולה רדיקאליים‪ ,‬שבו כל אחד יכול לכתוב על כל‬ ‫נושא‪ ,‬וכל משתתף יכול לערוך את עבודתם של משתתפים אחרים‪ .‬הסטנדרטים וקוי המדיניות של‬ ‫ניופדיה נזנחו‪ ,‬ונדרשו זמן ומאמץ ניכרים על מנת לעצב חדשים‪ .‬יחד עם זאת‪ ,‬סכמות ומשאבים‬ ‫חשובים ביותר שפותחו בניופדיה עברו לויקיפדיה‪ .‬אלו כללו‪ ,‬בין השאר‪ ,‬התעקשות על דמותה‬ ‫המסורתית של האנציקלופדיה; עמדה חזקה ביחס להטיות בתכני ערכים; משתתפים בעלי‬ ‫משאבים המחויבים ליצירת אנציקלופדיה אינטרנטית פתוחה; ומימון הפרויקט בעיקרו על ידי‬ ‫ג'ימי ויילס )באמצעות בומיס‪ ,‬חברה בבעלותו(‪.‬‬ ‫שני הפרקים הבאים )‪ ,(5-6‬מתארים בהרחבה את האופן שבו קוי המדיניות ואמות‬ ‫המידה של ויקיפדיה נוצרו והתמסדו בתקופה זו תוך כדי סדרה של עימותים ומחלוקות‪ .‬הראשון‬ ‫מביניהם מתאר את מיסודו של אופן ייצור התוכן בויקיפדיה‪ ,‬ואילו השני בוחן את האופן שבו‬ ‫נוצרו והתמסדו במסגרתה סטנדרטים להבחנה בין תכנים ראויים ולא ראויים‪.‬‬ ‫קוי המדיניות והמנגנונים הקשורים לאופן ייצור התוכן בויקיפדיה התפתחו כתגובה‬ ‫לבעיות שנבעו מכך שתוכן זה נכתב על ידי מתנדבים באופן שיתופי‪ .‬כבר בימים הראשונים‬ ‫לקיומה נאלצו משתתפיה ליצור עקרונות וקוים מנחים שיאפשרו שיתוף פעולה במצבי מחלוקת‬ ‫מבלי לפגוע ברעיון המקורי של הגבלה מינימאלית על תרומה לפרויקט‪ .‬המחלוקת הראשונה נגעה‬ ‫למיקומם של דיונים אודות ערכי האנציקלופדיה‪ :‬בימיה הראשונים‪ ,‬דיונים וויכוחים בין‬ ‫משתתפים אודות תכניו של ערך מסוים נכתבו בראשו או בתחתיתו של דף הערך‪ .‬פרקטיקה זאת‬ ‫היתה חלק בלתי נפרד מאופן השימוש המקובל באתרי ויקי באותו הזמן‪ ,‬אולם עמדה בניגוד‬ ‫לצורתם המסורתית של ערכי אנציקלופדיה‪ .‬לפיכך‪ ,‬המשתתפים בפרויקט נאלצו להכריע האם‬ ‫להעדיף את צורתו המקורית של הערך האנציקלופדי או את אופן השימוש המקובל בפלטפורמה‬ ‫הטכנולוגית שבבסיס הפרויקט‪ .‬הדיון בנושא‪ ,‬שהתקיים בתחילה ברשימת התפוצה של הפרויקט‪,‬‬ ‫עבר לאחר מכן לדפי דיון מיוחדים בויקי עצמו שהיוו בסיס לרשת‪-‬ליבה של דפי מדיניות שממקדת‬ ‫עד היום את הפעילות בפרויקט‪ .‬המחלוקת עצמה הוכרעה לבסוף באופן חד‪-‬צדדי על ידי קבוצת‬ ‫משתתפים – ובראשה לארי סנגר – אשר חבריה יצרו "דפי שיחה" המופרדים מ"דפי הערכים"‪,‬‬ ‫ה‬ ‫והעבירו את תכני הדיונים אליהם‪ ,‬למרות שלא הושגה הסכמה לדפוס פעולה זה‪ .‬הכרעה זו‬ ‫התמסדה עם הזמן‪ ,‬למרות ההתנגדות הראשונית‪ ,‬ושולבה מאוחר יותר בתוכנה שמפעילה את‬ ‫אתר ויקיפדיה‪ .‬בכך הפכה ממוסד חברתי הדורש תהליכי חברות ואכיפה מתמידים על מנת‬ ‫לשמרו לחלק בלתי נפרד מהמרקם הטכנולוגי של הפרויקט‪.‬‬ ‫המחלוקת השניה עסקה באופן שבו יש לייצג חילוקי דעות ועמדות מנוגדות במסגרת ערכי‬ ‫האנציקלופדיה‪ .‬מהרגע שבו החלו משתתפי הפרויקט ליצור במשותף ערכים בנושאים שנויים‬ ‫במחלוקת‪ ,‬החלו להופיע מחלוקות וקונפליקטים עזים המלווים בהאשמות הדדיות בדבר יצירת‬ ‫תוכן מוטה‪ .‬לאחר שנוצרו פתרונות מקומיים לבעיות מסוג זה‪ ,‬יצר ג'ימי ויילס מדיניות בשם‬ ‫"נקודת מבט נייטראלית"‪ ,‬אשר איגדה חלק מהפתרונות שפותחו בעבר ונתנה להם ניסוחים‬ ‫כלליים והצדקות‪ .‬תוכנה של מדיניות זו‪ ,‬שקבעה שיש להצהיר על כל תוכן שנוי במחלוקת ככזה‬ ‫ולייצג בהגינות את כל הצדדים בעימות‪ ,‬התפתח עם הזמן על מנת לכלול סוגים שונים של‬ ‫מחלוקות ולפתור או להכיל התנגדויות‪ .‬סביב מדיניות זו נבנו מנגנונים חברתיים שמטרתם‬ ‫להתריע על קיומם של ערכים בעייתיים‪ ,‬ולאפשר דיון והכרעה במצבי קונפליקט‪.‬‬ ‫הפרק השישי מתמקד בהתמסדות הסטנדרטים המסדירים את ההכלה או ההדרה של‬ ‫תכנים ויקיפדיה‪ .‬כבר בימים הראשונים נקבע שלכל אחד ממשתתפי הפרויקט שמורה הזכות‬ ‫)והחובה( למחוק תוכן לא מועיל שאחרים כתבו – אולם ההגדרה של תוכן מסוג זה היתה‬ ‫מצמצמת מאוד וכללה רק "שטויות מוחלטות" )‪ ,(patent nonsense‬כך שלפחות בחודשים‬ ‫הראשונים לקיומה של ויקיפדיה היא הכילה תכנים מגוונים ביותר שחלקם לא אנציקלופדי‬ ‫בעליל‪ .‬רק בשלב מאוחר יותר‪ ,‬לאחר מיסודם של קוי המדיניות שנסקרו בפרק הקודם‬ ‫והתייצבותו היחסית של הפרויקט‪ ,‬התעורר הצורך לתרגם את המודל האנציקלופדי כך שיתאים‬ ‫למאפייניה הייחודיים של ויקיפדיה‪ .‬תהליך זה‪ ,‬שדמה בצורתו לאלו שתוארו בפרק הקודם‪ ,‬היה‬ ‫קונפליקטואלי ביסודו ונכרך במאבקים על כוח וסמכות בפרויקט‪.‬‬ ‫המחלוקת הראשונה שהובילה לעיבוד ופרשנות של המודל האנציקלופדי ולברירה של‬ ‫מאפיינים רלוונטיים להקשר החדשני של ויקיפדיה‪ ,‬נבעה מעימות סביב השאלה האם ויקיפדיה‬ ‫צריכה להכיל ערכים מילוניים‪ .‬בעקבות עימות זה התמסדה מדיניות הקובעת שויקיפדיה אינה‬ ‫מילון‪ ,‬ושמטרתה היא יצירת אנציקלופדיה במובן מצומצם של המילה‪ ,‬המניח מאפיינים‬ ‫מסויימים שמקורם בהיסטוריה ובמוסכמות של האנציקלופדיה המודפסת‪ .‬ניסוח זה סלל את‬ ‫הדרך ליצירת סטנדרטים נוספים אשר נועדו להתמודד עם סוגי תוכן שונים שאינם רווחים‬ ‫באנציקלופדיות מסורתיות‪ .‬סטנדרטים מסוג זה‪ ,‬שקבעו איזה תוכן אינו ראוי לויקיפדיה אוגדו‬ ‫ו‬ ‫בעמוד מדיניות אחד שכותרתו "מה ויקיפדיה אינה"‪ .‬עמוד זה הפך לכלי מרכזי בהדרכת עורכים‬ ‫חדשים ולא מנוסים בדבר מדיניות התוכן בויקיפדיה‪ .‬עם הזמן‪ ,‬סוגי תוכן נוספים הלכו והתוספו‬ ‫אליו‪ ,‬לאחר שתכנים קונקרטיים בעלי מעמד מפוקפק שנוספו לויקיפדיה העלו את השאלה בדבר‬ ‫לגיטימיות שלהם‪.‬‬ ‫הדיונים וההכרעות במחלוקות שנסקרו בפרקים החמישי והשישי הובילו לגיבושם של‬ ‫מספר עקרונות של קואורדינציה שמלווים את ויקיפדיה עד היום‪ ,‬כגון הדגשת האופי הדיאלוגי‬ ‫של ייצור הידע המשותף בפרויקט; הסתמכות על הפניות וקישורים לטקסטים חיצוניים כערובה‬ ‫לידע לגיטימי‪ ,‬ולא על מומחיות או מחקר מקורי; השאיפה לחקות את הצורה המסורתית של‬ ‫ערכים אנציקלופדיים כפי שהיא באה לידי ביטוי באנציקלופדיות דפוס; עמדה ליבראלית ביחס‬ ‫לטכנולוגיה וייצור תוכן המעודדת יזמות ואלתור; זניחת האידיאל של ייצור תוכן מקורי‬ ‫והסתמכות נרחבת על העתקה או עיבוד של תוכן קיים; והסתמכות על "קונסנזוס בפועל" בין‬ ‫המשתתפים ככלי לקביעת מדיניות‪ .‬בנוסף‪ ,‬פרקים אלו מתעכבים על האופן שבו נוצרו בתקופה זו‬ ‫"אמצעי גיוס" )‪ (conscription devices‬ו"מנגנונים סקופיים" )‪ (scopic mechanisms‬טכנולוגיים‬ ‫וחברתיים כאחד כדי לתמוך בעקרונות אלו ולפקח על יישומם‪ .‬בין מנגנונים אלו‪ ,‬שרבים מהם‬ ‫קיימים בויקיפדיה עד היום‪ ,‬ניתן למנות את דפי המדיניות של ויקיפדיה‪ ,‬דפי מעקב אחרי עריכות‬ ‫מסוגים שונים‪ ,‬ותכונות שנוספו לתוכנת הויקי שעומדת בבסיס הפרויקט‪ ,‬כגון הפרדה בין סוגי‬ ‫עמודים‪ ,‬תבניות אזהרה‪ ,‬רשימות מעקב ועוד‪.‬‬ ‫הפרק האמפירי האחרון )פרק ‪ (7‬בוחן כיצד מומחי ידע משתמשים בויקיפדיה ומעריכים‬ ‫את מהימנות הידע המופיע בה‪ ,‬בהתבסס על ראיונות מובנים למחצה עם עשרים עיתונאים‬ ‫ישראלים‪ .‬ממצאים חשובים שעלו מראיונות אלה הינם השימוש השגרתי בויקיפדיה בעברית‬ ‫ובאנגלית כאחד בידי עיתונאים )תוך העדפה של הגרסה האנגלית(‪ ,‬ותהליך החיברות אותו עוברים‬ ‫עיתונאים חדשים הנועד להדריך אותם כיצד להשתמש בה כהלכה‪ .‬ממצאים אלו מעידים על כך‬ ‫שויקיפדיה‪ ,‬לפחות בהקשר זה‪ ,‬אכן הפכה למקור ידע מרכזי בחברת המידע העכשווית‪ .‬שימוש זה‬ ‫מותנה ומעוצב על ידי הסביבה והפרקטיקות המקובלות בעולם העיתונאי‪ ,‬ובפרט אילוצי הזמן‬ ‫הפרקטיים בהם נתונים השחקנים במסגרתו‪ ,‬אולם נראה כי הוא חלק בלתי נפרד משגרת העבודה‬ ‫העיתונאית כיום‪ .‬יחד עם זאת‪ ,‬ישנו טאבו חמור האוסר על אזכורה כמקור‪ ,‬ועיתונאים נדרשים‬ ‫למחוק כל עדות לשימוש שלהם בויקיפדיה מהטקסט אותו הם מייצרים‪.‬‬ ‫על מנת לבחון כיצד עיתונאים מתמודדים עם מתח זה בפועל בניתי מודל של האופן שבו‬ ‫עיתונאים משתמשים בויקיפדיה‪ ,‬הכולל חמישה שלבים‪ :‬בעיה ← חיפוש גוגל ←היוועצות בערך‬ ‫ז‬ ‫ויקיפדיה ← חיפוש אחר מקורות נוספים ← ניסוח מחדש‪ .‬בתהליך זה‪ ,‬העיתונאית עוקבת אחר‬ ‫שרשרת הפניות המובילה מהצורך הפרקטי המיידי שלה )ומהאילוצים אותם מפעילה עליה‬ ‫סביבתה( לכיוונו של מקור מידע משוער‪ ,‬עד אשר המשך המעקב הופך לבלתי אפשרי‪ ,‬או לא‬ ‫פרקטי או שהיא בטוחה מספיק בכך ששרשרת ההפניות מוצקה דיה וראויה לאמון גם בהמשך‪.‬‬ ‫במסגרת התהליך‪ ,‬ויקיפדיה מתפקדת כקיצור דרך מועיל אל ידע אשר מיוצר ומתוקף במקום‬ ‫אחר‪ .‬עם זאת‪ ,‬ויקיפדיה נוטה להפוך בסבירות הולכת וגבוהה לנקודת הסיום המעשית של‬ ‫שרשרת החיפוש‪ ,‬מכיוון שמשתמשיה נוטים להסתמך יותר ויותר על תוקפו של המידע הנמצא בה‪,‬‬ ‫ולשפוט אותו רק באמצעות עדויות מתקפות הנמצאות בערך עצמו‪ .‬לפיכך‪ ,‬השימוש של עיתונאים‬ ‫בויקיפדיה נוטה יותר ויותר לערב פרקטיקות שיפוטיות וביקורתיות שמטרתן לבסס את תוקפו‬ ‫של המידע הנמצא בה‪ ,‬ולסייע להם להחליט האם להשתמש בו בפועל או לחפש מקור אחר‪.‬‬ ‫מהראיונות עולה כי העיתונאים משתמשים באלמנטים בויקיפדיה המעודדים צריכה סלקטיבית‬ ‫וביקורתית מסוג זה‪ ,‬כגון דפי שיחה‪ ,‬הפניות למקורות שבאמצעותם נכתב הערך ותבניות אזהרה‬ ‫המתריעות בפני המשתמשים על בעיות בו‪.‬‬ ‫פרק הסיכום מוקדש לדיון בשאלות שעומדות בבסיס עבודה זו לאור הניתוח שהוצע‬ ‫בפרקי המחקר‪ :‬כיצד התמסדה התרבות האפיסטמית של ויקיפדיה‪ ,‬וכיצד היא מתקבלת על ידי‬ ‫עיתונאים‪ ,‬קטגוריה ייחודית של משתמשים שעיסוקם המקצועי כרוך בשיפוט והערכה של ידע‪.‬‬ ‫בניגוד לנטיה הרווחת לראות את מאפייניה של ויקיפדיה כנגזרים ישירות ובאופן בלתי‬ ‫בעייתי מתופעות טכנולוגיות או תרבותיות רחבות הקשורות בה )כגון הפתיחות של טכנולוגיית‬ ‫האינטרנט‪ ,‬השיתופיות של תנועת הקוד הפתוח או השטחיות של "הערצת החובבן"(‪ ,‬מהמחקר‬ ‫עולה כי אלו נוצרו באופן יצירתי והדרגתי על ידי מייסדיה ומשתתפיה‪ ,‬תוך תגובה לבעיות‬ ‫קונקרטיות ולכישלונות עבר‪ .‬בתקופה בה מתמקד המחקר נוצרו דפוסים כלליים של מיסוד‬ ‫והכרעת סכסוכים שעיצבו במידה רבה את אופייה‪ (1 :‬יצירת פתרונות מקומיים לבעיות שבבסיס‬ ‫המחלוקות בידי משתתפים בעלי השפעה ויוקרה; ‪ (2‬תרגום של פתרונות אלו לדפוסי פעולה‬ ‫שגרתיים ומסגור שלהם‪ ,‬לעתים תוך צירוף של פתרונות שונים לבעיות שונות ממספר הקשרים‬ ‫שונים והיטשטשות של מקורותיהם הפרוביזוריים; ‪ (3‬תרגום דפוסי הפעולה השגרתיים‬ ‫לפרקטיקות לגיטימיות באמצעות השקעה צורנית )‪ (Thévenot, 1984‬באובייקטים ומודלים‬ ‫תרבותיים מוכרים )כגון הצורה האנציקלופדית המסורתית‪ ,‬תרבות ויקי וכו'(; ‪ (4‬שזירתן של‬ ‫פרקטיקות אלו אל תוך מרקם הפרויקט באמצעות מנגנונים חברתיים ו‪/‬או טכנולוגיים‪ ,‬תוך‬ ‫התגברות על אפשרויות פעולה אחרות‪.‬‬ ‫ח‬ ‫בעוד חלק מקווי המדיניות שנוצרו כך הפכו לאחר מכן לנקודות מעבר מחייבות‬ ‫)‪ (Obligatory Passage Points, Callon, 1986‬המתנות את הפעילות של כל המשתתפים‬ ‫בפרויקט‪ ,‬נראה שחלקם שומרים על אופיים כאמצעי קואורדינציה וולונטריים או גמישים יותר‪.‬‬ ‫קווי מדיניות אלו כיניתי בשם נקודות מעבר מועדפות )‪ ,(Privileged Passage Points‬המתאמות‬ ‫את הפעילות החברתית ברוב המקרים אך משאירות מרווח וחופש פעולה להבדלים בין מקרה‬ ‫למקרה וניתנות לשינוי בעצמן‪ .‬נקודות מעבר מועדפות אלו הינן פחות יציבות ודומיננטיות‬ ‫מנקודות מעבר הכרחיות‪ ,‬ויש להן שליטה חלשה יותר בסביבתן‪ ,‬אך הן בו זמנית יותר גמישות‬ ‫וסתגלניות‪ .‬מכיוון שויקיפדיה נסמכת על תרומה וולונטרית של ידע בידי עורכים אנונימיים‪,‬‬ ‫גמישות כזו מאפשרת לה לשמר את הזרימה של מתנות הידע שלהם‪ ,‬גם אם במחיר של הגבלת‬ ‫הפיקוח עליהן‪ .‬גמישות זו גם תורמת ליכולתה של ויקיפדיה לתפקד כאובייקט גבול ) ‪Boundary‬‬ ‫‪ ,objects‬ראו ‪ :(Star and Griesemer, 1989‬אובייקט המאפשר לתורמים בעלי תפיסות שונות של‬ ‫ידע ושל אמת המגיעים מעולמות חברתיים שונים מאוד‪ ,‬ולעתים עוינים‪ ,‬לשתף פעולה‪.‬‬ ‫יחד עם זאת‪ ,‬כפי שמדגים אופן השימוש של עיתונאים בויקיפדיה‪ ,‬הגמישות והשליטה‬ ‫המוגבלת בתכנים האופייניות לתרבות האפיסטמית של ויקיפדיה יוצרות חוסר ודאות לגבי‬ ‫מעמדו של הידע המופיע בה‪ .‬מהראיונות עולה כי המתח בין חוסר הודאות ביחס לידע המופיע‬ ‫בויקיפדיה לבין השימושיות של תכניה מתבטא ביחס אותו כיניתי לגיטימציה פרקטית‪ ,‬סוג‬ ‫ייחודי וחלקי של לגיטימציה תלוית‪-‬הקשר המשוקע בפרקטיקות של המשתמשים‪ ,‬ולא בהצדקות‬ ‫עקרוניות‪ .‬לגיטימציה מסוג זה מעניקה לידע סמכות מוגבלת‪ ,‬המותנית בקשירתו למקורות ידע‬ ‫נוספים על ידי רצף מובנה של פעולות‪ ,‬ובכך מבטאת את מגבלות האמון בו‪ .‬סוג זה של לגיטימציה‬ ‫מביא בחשבון בו‪-‬זמנית את שימושיותה של ויקיפדיה ואת סכנותיה‪ ,‬את הלגיטימיות שלה בפועל‬ ‫ואת היעדר הלגיטימיות העקרונית שלה‪ .‬למרות שסוג מוגבל זה של לגיטימציה מספיק על מנת‬ ‫להפוך את ויקיפדיה לחלק בלתי נפרד מארגז הכלים של עיתונאים‪ ,‬הוא מונע ממנה את ההכרה‬ ‫הציבורית המתבקשת מהשימוש הנרחב בה‪.‬‬ ‫על בסיס הממצאים שתוארו‪ ,‬אני מצביע על מספר מגבלות וכיווני מחקר עתידיים‪ .‬מגבלה‬ ‫מרכזית של המחקר נובעת מהתמקדותו בתקופת זמן מוגבלת למדי שבה התמסדו מאפיינים‬ ‫מרכזיים שלה‪ .‬מחקר המשך יאפשר לבחון כיצד ועד כמה גמישותן של נקודות המעבר‬ ‫המועדפות בויקיפדיה אובדת עם הזמן‪ .‬בהקשר זה‪ ,‬מחקר נוסף נחוץ גם כדי לבחון האם ניתן‬ ‫לאתר תופעות דומות לאלו שתוארו בעבודה זו גם בפרויקטים מדעיים‪ ,‬או בייצור של מוצרי‬ ‫ט‬ ‫תרבות שאינם קשורים ישירות לידע‪ ,‬ולמפות את התנאים בהם הן יטו להופיע כאמצעי תיאום‬ ‫חברתי‪ ,‬או להיעלם‪ .‬מגבלה מרכזית נוספת נוגעת לאוכלוסייה המתוחמת מאוד של משתמשים‬ ‫אותם חקרתי‪ .‬על מנת לאשש את הרעיון בדבר הלגיטימציה הפרקטית של ויקיפדיה יש צורך‬ ‫להשוות בין משתמשים ברמות מומחיות ובתחומי עיסוק נוספים‪ .‬בנוסף‪ ,‬רצוי לבחון האם‬ ‫לגיטימציה פרקטית מאפיינת פרויקטים דיגיטאליים מבוזרים נוספים‪ ,‬ובכך מרמזת על תופעה‬ ‫עכשווית רחבה יותר‪ ,‬וכן האם תופעות דומות קיימות בפרויקטים חדשניים אשר הסביבה‬ ‫והטכנולוגיה שלהם שונות מאוד מאלו של ויקיפדיה‪.‬‬ ‫לבסוף‪ ,‬אני דן בהשלכות האפשריות של עבודה זו על חקר הנתינה‪ ,‬וכן באופן שבו‬ ‫התרבות האפיסטמית של ויקיפדיה מתייחסת לקישור בין ידע ומומחיות במודרניות‪ .‬בהקשר‬ ‫הראשון‪ ,‬נראה כי ויקיפדיה ממשיכה את המסורת המערבית בכלל והמודרנית בפרט הקושרת בין‬ ‫ידע ו‪/‬או חיפוש האמת ובין נתינה ללא כוונת רווח‪ .‬יחד עם זאת‪ ,‬היא מסתמכת על רפרטואר של‬ ‫נתינה הכולל סוגי נתינה שונים המתקיימים במקביל ומנגנונים חדשניים התומכים בקיומם הבו‪-‬‬ ‫זמני‪ .‬בהקשר השני‪ ,‬נראה כי למרות שויקיפדיה יוצרת תרבות אפיסטמית חדשה וקוראת תגר על‬ ‫מרכיבים חשובים בתפיסת הידע המודרנית‪ ,‬היא משלבת אלמנטים של תפיסה זו יותר מאשר היא‬ ‫דוחה אותם‪ .‬בכך היא מצביעה על הצורך בהמשגות מורכבות ורבות רבדים יותר של השינויים‬ ‫ביחס לידע בתקופתנו‪ .‬המשגות מסוג זה אינן דורשות את הביטול של מושג המודרניות ) ‪Latour,‬‬ ‫‪ (1993‬או את הבנתו כמיושן ולא רלוונטי )‪ .(Lyotard, 1984‬במקום זאת‪ ,‬אני מציע לוותר על‬ ‫הניסיונות להמשיג את המודרניות באופן כוללני וחד משמעי ולהתייחס אליה כאל רפרטואר‬ ‫תרבותי מורכב ורווי מתחים אשר היבטים שונים שלו באים לידי ביטוי – או נדחים במפורש –‬ ‫באופן מקומי ותלוי הקשר‪.‬‬ ‫י‬ ‫תוכן העניינים‬ ‫תקציר ‪i‬‬ ‫פרק ‪ :1‬מבוא ‪1......................................................................................................‬‬ ‫‪ 1.1‬עלייתה של ויקיפדיה‬ ‫‪ 1.2‬סקירת הספרות‬ ‫‪ 1.3‬המסגרת התיאורטית‬ ‫‪ 1.4‬מטרות המחקר‬ ‫‪4........................................................................‬‬ ‫‪10 .......................................................................‬‬ ‫‪29.........................................................................‬‬ ‫‪49 .............................................................................‬‬ ‫פרק ‪ :2‬מתודולוגיה ‪53 ............................................................................................‬‬ ‫‪ 2.1‬בסיס נתונים ראשון‪ :‬ארכיונים אינטרנטיים ‪57 .....................................................‬‬ ‫‪61.....................................................................‬‬ ‫‪2.2‬בסיס נתונים שני‪ :‬ראיונות‬ ‫פרק ‪ :3‬ניופדיה ‪66.................................................................................................‬‬ ‫‪67..........................................................................‬‬ ‫‪ 3.1‬רקע היסטורי‬ ‫‪71.......................................................................‬‬ ‫‪ 3.2‬המודל האנציקלופדי‬ ‫‪77........................................................................‬‬ ‫‪ 3.3‬אופן הייצור‬ ‫‪85 .................................‬‬ ‫‪ 3.4‬מחלוקת‪ :‬שינויים בשיטות הייצור‬ ‫‪90 .........................................................................................‬‬ ‫‪ 3.5‬סיכום‬ ‫פרק ‪ :4‬התקופה המכוננת של ויקיפדיה ‪95 ..................................................................‬‬ ‫פרק ‪ :5‬מויקי לאנציקלופדיה ‪104 ...............................................................................‬‬ ‫‪109 .......................................................‬‬ ‫‪ 5.1‬שלב ‪ :1‬יצירת כללי בסיס‬ ‫‪116.....................‬‬ ‫‪ 5.2‬שלב ‪ :2‬הטיהור של דפי ערכים‬ ‫‪124......................................................‬‬ ‫‪ 5.3‬שלב ‪ :3‬נקודת המבט הנייטראלית‬ ‫‪142 .......................................................................................‬‬ ‫‪ 5.4‬סיכום‬ ‫פרק ‪ :6‬מאנציקלופדיה לויקיפדיה ‪146........................................................................‬‬ ‫‪149 ..................................................................‬‬ ‫‪ 6.1‬שלב ‪ :1‬ויקי אינו נייר‬ ‫‪152 .................................................‬‬ ‫‪ 6.2‬שלב ‪ :2‬ויקיפדיה אינה מילון‬ ‫‪176 ..........................................................‬‬ ‫‪ 6.3‬שלב ‪ :3‬מה ויקיפדיה איננה‬ ‫‪191 .......................................................................................‬‬ ‫‪ 6.4‬סיכום‬ ‫פרק ‪ :7‬ויקיפדיה בשדה העיתונות הישראלי ‪197 ..........................................................‬‬ ‫‪200 ...........................................‬‬ ‫‪ 7.1‬ויקיפדיה בסביבת העבודה העיתונאית‬ ‫‪204.........................................................‬‬ ‫‪ 7.2‬ויקיפדיה בפרקטיקה העיתונאית‬ ‫‪220...............................................‬‬ ‫‪ 7.3‬השימוש בעיתונאים ישראלים כמרואיינים‬ ‫‪221........................................................................................‬‬ ‫‪ 7.4‬סיכום‬ ‫פרק ‪ :8‬סיכום ודיון ‪225 ...........................................................................................‬‬ ‫‪231......................................................................‬‬ ‫‪ 8.1‬השלכות תיאורטיות‬ ‫‪239.......................................................................‬‬ ‫‪ 8.2‬מתנת הידע‬ ‫‪244....................................‬‬ ‫‪ 8.3‬ויקיפדיה והתרבות האפיסטמית העכשווית‬ ‫ביבליוגרפיה‪248 ....................................................................................................‬‬ ‫נספח‪ :‬מקורות ראשוניים ‪272 ...................................................................................‬‬ ‫תקציר בעברית ‪ ...................................................................................................‬א‬ ‫עבודה זו נעשתה בהדרכתה של פרופ' אילנה סילבר‬ ‫מהמחלקה לסוציולוגיה ואנתרופולוגיה‪ ,‬אוניברסיטת בר‪-‬אילן‪.‬‬ ‫תרבות אפיסטמית חדשה?‬ ‫ויקפדיה כאתר של ייצור ידע במודרניות המאוחרת‬ ‫חיבור לשם קבלת התואר "דוקטור לפילוסופיה"‬ ‫מאת‪:‬‬ ‫ליאור גלרנטר‬ ‫המחלקה לסוציולוגיה ואנתרופולוגיה‬ ‫הוגש לסנט של אוניברסיטת בר‪-‬אילן‬ ‫רמת גן‬ ‫אייר תשע"ג‬

RELATED PAPERS

RELATED TOPICS

Log In

A New Epistemic Culture? Wikipedia as an Arena for the Production of Knowledge in Late Modernity

A New Epistemic Culture? Wikipedia as an Arena for the Production of Knowledge in Late Modernity

A New Epistemic Culture? Wikipedia as an Arena for the Production of Knowledge in Late Modernity

A New Epistemic Culture? Wikipedia as an Arena for the Production of Knowledge in Late Modernity

A New Epistemic Culture? Wikipedia as an Arena for the Production of Knowledge in Late Modernity

Related Papers

RELATED PAPERS

RELATED TOPICS