Thursday, August 05, 2021

IPSC Panel 9 – Crosscutting IP

Derek E. Bambauer, Everything You Want: The Paradox of Tailored IP Regimes

Customized IP has benefits (avoids lowest common denominator [or highest] problem), but also costs: manipulation to shift from one regime to another; colleciton of information by decisionmakers.

Standard debate assumes semi omniscience of designer neutrally concerned with social welfare. What if rules are largely written by the regulated parties? Bespoke systems are often underutilized by their own designers and outmoded. The paradox of getting your way and finding it unsatisfying.
Examples: Vessel Hull Design Protection Act. Industry pushed very hard against cheap copying, and yet as of 2019 there were only 538 registrations in 20 years versus hundreds of thousands of utility patents. Was more heavily used 1999-2003. It seems that boat designers/manufacturers used C&Ds under this regime, but there were 1100 utility patents on boat hulls and 20 design patents since 2013, which was the last boat hull registration. Sharp shift to regular regime. Fed. Cir. found in one key case that boat hulls didn’t infringe if the decks were different. Registration is fast—takes a month—versus patent, but term of protection is only 10 years.

Computer chips: Semiconductor mask works, 1984. Through 2012, only about 1000 registered. Technical reason: as chips have grown in size and specialization, reverse engineering is more expensive than designing them from scratch.

Audio Home recording Act: Excluded general purpose computers, so whoops. Diamond Rio case: excluded MP3 players. Customers didn’t like paying more for DAT; CDs won.

It’s hard to write and get passed exactly what you want—industry insiders may compete. Also, as a tailored regime begins to flow closer to a generalized regime, it falls down the gravity well of the general field and is overtaken. Maybe small changes are easier to pass.

Industries design for current needs and not for what it may evolve to because innovation is difficult to predict (like the future), especially if generated by upstart or edge firms or driven by wider tech changes. Even if we assumed incumbents have better info, their own internal pressures exist to align legal rules w/existing business models. Give and take of legislative process may be better when it harnesses information from more stakeholders than those who are generating the IP. Tailored regimes may also be tightly coupled and fragile—subject to disruption/irrelevance from one unfavorable judicial ruling, like boat hulls.

Generalized regimes are more adaptable, adjustable by more institutional stakeholders (judges, admin agencies, the bar). Overton window: you get your specialized regime; it’s difficult to revisit that after something changes because the legislature thinks it’s done. Copyright Term Extension Act: blatant rent extraction generates or increases opposition. When Big 3 automakers requested their own tailored design regime to exclude repair parts, Congress didn’t want that.

Bespoke regimes may divert innovation—music industry was successful in delaying and killing off DAT, but that allowed the rise of the CD, and MP3s, and then Napster and P2P.

IP industries have historically been terrible at prognostication, as w/the VCR. Maybe we should be less worried about attempts at incumbent protection/rent extraction via IP.

Even if my thesis is right, it will still be an irresistable lure/strong optimism bias, which will cause incumbents to concentrate on legislative activity instead of innovation under the standard IP system.

Jake Linford: are there overlaps with patents and boat hulls suggesting two bites at apple.

Bruce Boyden: another reason is that crafting a bespoke system takes years, making narrow legislation more likely to be out of date.

Mark McKenna: Another explanation was maybe these bespoke regimes weren’t addressing actual problems; overrepresentation of people w/specific litigation interests.

Betsy Rosenblatt: If they solve problems, general systems might be much better at solving specific problems and not advantaging opportunists—a narrow protection might be harder to manipulate.

RT: Before coming to a conclusion about comparative advantage, consider what creates trolls in more general system: different kinds of exploitability of larger systems. Consider compulsory licensing as a midway point here as well. Also, the history of DMCA was that 512 was the price (which the © industries thought was small) for 1201; 1201 turned out not to be worth it, which supports your argument, but also indicates that matters can be more complex if there is horsetrading across regimes. 

Mala Chatterjee, Understanding Intellectual Property: Expression, Function, and Individuation

© and patent differ in breadth and type of rights (copying requirement v. independent invention). Treated as distinct and scholars warn against overlapping rights. But what are the subject matters and how are they to be distinguished?

Argues that defining difference b/t creative works and inventions is the way in which they are individuated. Creative works are author-individuated while inventive works are structure-individuated. Two acts of authorship can’t result in the same creative work, only structurally identical works, while two acts of inventorship can result in the same invention.

An author who has made something has said something. Sets aside what counts as expressive work, but the defining feature of the category is speech. Expression is the kind of thing for which where it comes from makes a difference to what it is. Author’s work is uniquely hers even if others might also be connected to it and might make things that look alike. Pierre Menard; appropriation art where the point is that it is structurally identical to something else—Warhol’s Brillo Boxes and Pettibone’s Brillo Boxes, and this matters to their aesthetic properties.

Inventions are importantly different from creative works in being tools: instrumentally valuable for some specified end. This isn’t to say that it doesn’t matter at all who invents: it could be historically relevant that both Liebnitz and Newton invented calculus, but that doesn’t make calculus different.

Vindicates core features of the structure—originality as © requirement—must come from author, not must be novel. Independent creation is likewise justified. For patent, this makes sense of utility, novelty, and lack of independent creation defense. Ownership is of linguistic description of structural properties of invention.

For ©, standards of proof are wrong if they allow inference of copying from structural similarity alone; theory also refutes rejections of transformativeness where clear meaning to audiences is different even w/o structural change. Also relevant to some patent doctrines (sorry).

Jeremy Sheff: Paul Goldstein has told this story several times—quotes a colleague saying if Shakespeare had died as a child, we would never have had Hamlet, but if Newton had died as a child, we would still have calculus. He sees this as something about teleology of patent and ©: what they are and what ends they’re for. Do you care about ends?

A: don’t know what it means for design of legal systems unless paired with theory about aims of legal systems.

Linford: is this reputational? Richard Prince is understood to have a different meaning/authorship of his works because of his reputation. That might have implications for what counts as authorship in cases like Garcia. Is this bringing in concepts from TM/ROP about branding? Could a patent turn on the personality of the creator?

A: recognizing that creative works are more tied to authors generally than invention is. Need to look for authorial intent, but part of what tells us is whether that results in successful communication is context surrounding the actor. But not saying that authorial intent wholly constrains the work.

Mark McKenna: sounds like Abraham Drassinower’s theory—he says he’s explicating a view inherent in © as a system not a metaphysical view that there are things that are inventions out there in nature. Are you making an argument that these are not just legal constructs but things that are true about the world?

A: aligned with Drassinower but one important difference is that she is starting not just with legal system but with works and practices surrounding them and arguing that they are plausibly real things in the world.

Betsy Rosenblatt: pace Amy Adler, doesn’t your approach kill all of © if it’s about individuated expression? © is at odds with that at least for the sorts of works that are personally expressive. We don’t care about authenticity of an air conditioner manual, but it's still in ©. Also urge you to include recipient/dialogical creation of meaning, which helps you—an invention is structurally the same to every recipient and expression isn’t.

A: not quite all of ©, though it does suggest limits. 

Dilip Sharma, Arbitrability of Intellectual Property Disputes

Indian cases have long pendencies, making arbitration more attractive. WIPO offers mediation and arbitration services. Number is rising, and expected to rise in India. Court: Booz Allen v. Hamilton—only disputes involving rights in personam are arbitrable and IP isn’t. Delhi High Court also held that © infringement can only be dealt with by courts because the statutory remedies are all there is. But Eros v. Telemax allowed arbitration: law said that every © suit or civil proceeding should be instituted in a district court; but interpreting that to bar arbitration would be too broad where the IP dispute arises out of a commercial contract. Another case: patent disputes about infringement were not about in personam rights and were arbitrable.

Other nations: much more pro arbitration. Countries may require registration of arbitral award before it’s enforceable. India needs legislative support.

RT: Does India recognize UDRP results? It’s mandatory for registrants to agree. Possible model?

A: to the extent required by international rules, yes, but not much support.

Ana Alba: consider enforceability of international arbitrations—that is also required in some cases.

Mauritz Kop, Quantum Technology: Waiving or Pledging IP?

General principles of quantum mechanics: physics of very small + engineering. Superposition, entanglement, tunneling: defies laws of physics on the macro level. Applications: computing, communication, sensing, simulation, basic research, and AI. Not yet many use cases beyond cybersecurity, finance, and defense; consequences remain mostly unknown. Dual military/civilian uses are possible necessitating export controls and shared tech transfer policies. Foresees heavy regulation and chilled innovation given the very high risks. Key principle: equal access to benefits. Similar motivation to open COVID pledge, but key quantum tech is now controlled by a handful of multinationals, universities, and gov’ts. Could we temporarily override rights in quantum tech to repair market power problems?

Should build on adjacent fields like CRISPR, nanotech, which also require huge initial investment. Looking for mechanisms beyond social solidarity. Right now waiving IP would have no effect without sharing knowhow, since patents don’t disclose that right now; there is no skilled workforce available; clean rooms are very expensive. Pledging IP looks promising given the current limited number of stakeholders. Lack of consensus about whether a special regime is needed.

Sheff: questions about resource limits being more important than IP

Rosenblatt: given those resource requirements, maybe history of nuclear regulation is more important than history of AI regulation


No comments: