Friday, August 08, 2014

IPSC part 7: copyright limitations

Fifth Breakout Session
Copyright Limitations

Inventing Around Copyright
Dan Burk

Recurring pattern in ©: Napster/Grokster, where Napster was held liable for having centralized architecture/ability to know what’s on the system and thus being contributorily/vicariously liable. Not surprisingly, next generation P2P got rid of centralized architecture.  Initially successful in that, but the SCt invented a new kind of liability and held them responsible for that instead.  Similar result in Aereo. Judge Chin saw Aereo as Rube Goldberg contrivance to get around Copyright Act, and SCt basically agreed.  When a company tries to design something that would be legal, that’s usually not culpable, and we try to encourage it in the patent area.  Long tradition of inventing around in patent, and Federal Circuit has often said this is a positive feature of the patent system: trying to avoid infringement encourages innovation.  Innovators become more muscular/innovative.

Why is this supposedly good in patent but bad in copyright?  You can look at inventing around in patent as a kind of bargaining breakdown—will invent around if licensing isn’t cheaper—patentee has incentive to license if the alternative is a competitor who’s invented around.  Failure to agree = mismatch in valuation between parties.

But in patent law, this isn’t avoidance/evasion—not a loophole, but engaging in socially beneficial activity.  Also, Aereo/Napster/Grokster are not trying to create substitute for copyrighted work.  Trying to develop an alternative delivery system.  So the copyright owner and innovator aren’t racing in the same space. 

Might justify more generous view of inventing around, because in inventing I’m up against a range of rightsholders and not just a single rightsholder (breakdown in bargaining). 

Is Judge Chin correct that this is just socially wasteful redesign?  Netflix is looking at Grokster style distribution to deliver films; might be worthwhile.

Q: aren’t there successful examples of inventing around in copyright?  They don’t call it designing around.

A: yes; it’s common but that makes it interesting that we haven’t talked about it much.  Some of the examples will be more in Fishman’s area of creating around, as opposed to inventing around.

Gibson: Is Grokster really an example? The SCt decision wasn’t really about the architecture (indeed Bittorrent remains extant) which is ok if you don’t say the magic words.

A: was about the architecture in a negative sense, in the sense that it was avoiding Napster liability. SCt would’ve loved to have found justification in existing doctrine for finding control, but had to borrow anew from patent.

Q: is there a constitutional argument against giving copyright owners veto power over non-Writing innovations?

A: Sympathetic to the notion, but surprised if there were a constitutional problem because both in patent and copyright the secondary liability doctrines extend outside the scope of the right granted.  Something that’s specially adapted to infringe that’s outside the patent still is subject to the patentee’s control. 

Q: copyright owner doesn’t have much influence over text once created, while patent owner has control over scope of claims, so it’s not just encouraging production but about shaping what incentives different people have to make claims in the first place.

A: doctrine of equivalents forces you to invent way around—so patentee has some control, but not all.

Beyond the Readymade: Michael Asher’s Skulptur Projekte Münster Caravan and the Challenges for Copyright Law
Shane Burke

This is a trailer placed around town as an art project.  Is it an artwork?  Lucasfilm standard in the UK—definition of sculpture.  Having another use doesn’t disqualify something from being a sculpture; purpose is important and artistic purpose can be discerned by looking to communities/reception.  A pile of bricks temporarily on display in a museum is different—it can be a sculpture—whereas a pile of bricks dumped on a driveway in preparation for repairs isn’t.

Traditional definitions wouldn’t call Asher’s trailer a sculpture, but harmonization may force UK standard to be more flexible.  US standard: industrial product v. creative product.  Blanch v. Koons demonstrated more progressive attitude towards appropriation; Cariou adopted reasonable observer and deferred to institutional status of Prince as recognized artist—which may also matter in UK.  Conceptual separability: problematic on a number of levels.

Possibility of protection in the UK exists under an institutional theory.  Probably not in the US.

Gilden: Why is it important to protect this work under copyright?  What would infringement be, if it’s selection, coordination and arrangement?

A: infringement is a different issue—he is separating it out.  (Wouldn’t photos of the work at least if it were in a gallery infringe if there’s copyright? Or recreating a readymade?)

Q: different forms of art may be eligible for different types of IP protection.  Fluxus—a movement intended to be ephemeral. Would artists even want copyright protection?  Property rights—land on which the jetty sits for a famous conceptual artwork.  Problems of attribution: under American moral rights, the idea that attribution/integrity are important is hard to reconcile with some of these works.

A: yes, different solutions for different works.  Ascher was a Marxist into institutional critique of art world, so he might be turning in his grave.

Ochoa: should this be copyrightable?

A: moved from institutionalism away from it—he thinks © would be against the nature of the work.  Respect the artist’s own perspective.

Ochoa: if he wanted copyright you’d feel differently?

A: yes.

RT: What about the other part of authorship, attributing the thing to the author?  He’s definitely the author of an idea I’m willing to call art, but he’s not an author of the trailer.

A: comes from context, and the fabrication requirement has been relaxed.

RT: but there’s relaxation and then there’s not having a requirement at all.  I can’t understand a regime in which the maker of the trailer, or of toilets, could become an infringer if it continued to make trailers/toilets after Ascher/Duchamp.

Lemley: but only in museum context.

RT: makes no sense to me. We have a reproduction right for a reason.  That’d be a reproduction, unless the artistry is the idea, which isn’t protectable/isn’t copied by manufacturing.

Creating Around Copyright
Joseph Fishman

Constraint is getting a bad rap in copyright discourse.  We want the copyright system to promote creativity, and the usual view is that © affects upstream creative output incentivized minus downstream creative output suppressed.  Sum of that equation is argued to be a negative number.  (RT: I don’t think the people he cites, including me, argue that in the form presented.  We are arguing about marginal effects of increased protection on the curve, which we think we’re already too far down.)

Constraint is in fact often good for creativity, adding a third variable: downstream creative output stimulated by having to work around copyright restrictions. Our goal shouldn’t be to minimize or maximize constraint, but to optimize it.  Project: Preliminary steps to that.

Creativity to him means what psychologists have converged around: generation of products/ideas that are original and appropriate to whatever the task happens to be, defined according to consensus of particular community/audience.  This tracks real world assessments of creativity. Constraint ends up fueling creativity—your own recent scholarship may have involved coming up with a new idea/improving an idea by having to distinguish your claims from those before. Inability to license protected content leads to generation of some other creative output.

Star Wars began when George Lucas wanted to remake Flash Gordon, but couldn’t get the rights and set out to translate what he likes about the world of Flash Gordon into a new creative setting. Similar stories about birth of Mario, b/c Nintendo wanted to make a Popeye game but couldn’t.  DJs who, to get around restrictions on sampling, pushing sampling methods to more intricate territory.

Experimental evidence: more constraint can be better than less.  Appropriate exemplars become harder and harder to find—problem solving under constraint becomes more abstract and exploratory. People use analogies to build a new solution from the ground up. We tend to get a more diverse set of original products.  Removing constraints means that some other creations are never going to be made, and there’s no reason to assume that creations born of more choice will be better than creations born of less. That doesn’t mean we should ramp up constraints indiscriminately, b/c not all constraints have the same effect on creativity. We ask who is imposing the constraints, how broadly, when during the creative process it’s imposed. These are levers that can be pushed and pulled. 

Paper develops taxonomy and argue that © does better on some than others.  Where it’s on the right track: the target of the constraint.  The object that’s restricted.  Budget = target is money; deadline = target is time. ©’s constraint is different—constraint is helpful when it makes familiar exemplars harder to access. 

Still, copying can be used creatively too. Fair use emphasis on transformativeness sends a signal to downstream creators that they can copy if they transform.  Fair use functions as a constraint on downstream creators by telling them to stay on the right side.  Having to comply w/constraints of fair use ended up improving one poet’s poetry in the long run. 

His argument is currently theoretical, and can’t tell us whether current law helps or hurts on net, and even if we knew that it would be hard to parse out which aspects of constraint were helpful because there are so many different moving parts. Now designing controlled experiments, looking to see whether/how creative production fares under similar constraints to ©.

Dan Burk: Fuzziness of constraints—if you get too close that’s infringement—ensures greater distance?  Your theory assumes that the follow on creator has a number of choices. Interesting questions for merger doctrine.  Would support your argument.

A: substantial similarity/central claiming does tend to force outcomes with less reinvention of the wheel.  Does want cake/eat it too in that substantial similarity is unclear and clarity for downstream creators is valuable.  Headway is being made in fair use/becoming more predictable, not so much in substantial similarity. Clearer rule would help.

RT: [This theory is about how second-comers react, not about how pure copiers react.  You therefore have to have a theory by which Rosencrantz & Guildenstern are Dead is “less creative” than Star Wars.  I don’t think that theory can consistently be fit into copyright as we know it.]

“Constraint is good when it makes exemplars harder to access” equivocates on the meaning of “access.”  Copyright law can’t stop people from being familiar with existing designs, which is what happens in the experiments on which you rely. Instead an example given to them and they’re instructed by the law not to copy too closely—which is, incidentally, similar to the instruction that failed in at least some experiments.  Related to the problem that unlike the subjects in the experiments cited in the paper, the lazy creator actually has a whole menu of possible existing works both in the public domain (Clueless) and not (Outlander) that it can use or license if Game of Thrones is unavailable.  The make or buy decision, in other words, is quite different than the make or copy decision at least subconsciously faced in the creativity studies.  (I think this is related to the differences between patent and copyright and the fact that, as Barton Beebe has so persuasively argued, we don’t have an idea of what it would mean to have “Progress” in copyright, whereas we have a much better idea of what progress in the cellphone market means.)  Someone will license a sword and sorcery fantasy to you at a price that at least matches the cost of creating around Game of Thrones—other optioned properties, for example, include Marion Zimmer Bradley’s Darkover.  (Conclusion: someone who prefers to make a derivative work can do so if they are merely profit-maximizing and don’t care so much which work they ultimately make; copyright’s constraints don’t have the same punch for inventing around as they do for controlling pure reproduction.)

A: There are conflicting results on how much you get out of making something available for participants to look at in these experiments; other experiments find that you can give an exemplar and then tell people not to copy and they obey. [This is a standard problem of reproducibility, which also shows up in the work on verbal overshadowing I love so much; the studies used identical instructions.]  This is one thing we plan to look at in the experiment we’re running—a trade secret like condition in which participants never see the works they could copy from and another where they see it but are instructed not to copy and see what that leads to. Doesn’t have a great answer.  [I don’t see how that’s relevant to the copyright environment.  Wouldn’t you want a condition in which both groups saw the exemplar, but for one the instructions were that you could do anything but copy exactly versus a condition in which you couldn’t create something substantially similar to the exemplar?]

Wendy Gordon: good to look at areas where there may never be a full doctrinal/empirical payoff.  Bloom’s creative misprision may be of use.  Quote: “creative people are prisoners. That is to say they get captured and the only way out is to beat a path out …”  These constraints are “as functional as any jail cell ever constructed.” 

Forgotten and Abandoned: Section 113’s Limitations on Copyright in Works Depicting Useful Articles
Tyler Ochoa

Reproducing a useful article is not infringement even if it’s depicted in a copyrightable drawing or even a model.  We don’t want you circumventing prohibition on protecting useful articles by creating drawings of those useful articles.  Drawing of a useful article—e.g., traffic approach to bridge—doesn’t give you exclusive right to make the bridge.

Case law has gone off the rails.  Former employee of plaintiff copied blueprints of tire designs for underground mining vehicles and produced a range of tires based on the designs: Tire Eng’g & Dist v. Shandong Linglong Rubber (4th Cir. 2012). Jury awards $26 million in damages based on sale of tires.  Jury has no business awarding damages for the manufacture of tires at all--§113 precludes this, but it isn’t recognized or even raised. That is just bad lawyering in one sense, but §113 isn’t about protecting Ps or Ds—it’s intended to protect the public interest in useful articles, and should’ve been raised sua sponte.

Kohus v. Mariol (6th Cir. 2003): the work is a protable children’s playyard inclduing drawings for a latch locking the upper rails in place. P’s drawing allegedly infrined by patent drawings submitted by D.  Court says the latch is unlike others b/c it has different features—throughout opinion equate copying latch w/copying drawings of latch.  Case endorses abstraction/filtration/comparison, but their abstraction isn’t the drawing, it is the latch—elements dictated by the function of the latch, standard industry practices for constructing latches. Wrong! You need to filter out the latch!

Most recently, the Batmobile: DC Comics v. Towle (S.D. Cal. 2013): © owner doesn’t own © in movie, but in original comic series—claim Batmobile is derivative work.  But a 3D useful article is covered by §113, and merely depicted in the comics, though the logo might be different.

Q: these cases all seem to be “really” other kinds of IP cases—trade secret, patent, TM. Would these problems be solved if courts used the proper IP regime?

A: Yeah, that’s the whole point of §113—separating industrial design from copyright. But we never enacted separate protection for industrial design, other than design patent, people try to get that protection some other way.

Lemley: what does it mean to reproduce the copyrightable portions of the drawing in a useful article? There are a bunch of useful features in a Batmobile, but he takes it that those aren’t the copyrightable elements in the drawing of the Batmobile anyway—the wheels/chassis weren’t what made it creative.  Don’t we have a separability problem?  I couldn’t be liable for making a car, but why couldn’t I be liable for adding to the car the elements that aren’t useful or functional?

A: separability inquiry is necessary.  But there are several problems w/that in Batmobile case, b/c Batmobiles built for TV series didn’t look anything like the Batmobile in the comics.  Some features like the logo may be separable, but many of these features aren’t.

Lemley: if so, what is §113 doing for us that isn’t done by Brandir etc.?

A: even in a case w/no separability, like the mining tire case, we see courts awarded damages for making the tires.

Lemley: but wouldn’t that be true w/o §113?

A: codifies that you don’t get protection for useful articles just b/c you designed or drew them.

Gordon: separability isn’t a criterion for infringement.  You need §113 to do extra work. 

Toward Principled Justifications for Copyright Limitations and Exceptions
Pamela Samuelson

Comparative: Netherlands has 75 specific exceptions but no general fair use. Interested in taxonomy of exceptions for quite some time.  We didn’t have them for a long time, but then we had very limited rights conditioned on formalities for a very long time.  Exhaustion came up first. Fair use didn’t become meaningful until 20th century; instead it was first sale, compulsory license for musical works, and outright exemption for coin operated music players.  Even under 1909 Act the exclusive rights were sufficiently cabined that we didn’t need that many.  Now, you can’t read the Copyright Act all the way through—exceptions almost half of the statute, and way too long.

Bernt Hugenholtz has a taxonomy of exemptions.  Her project: working through the US exemptions and figuring out what and why.  Berne Convention’s mandatory right of quotation—promoting the next generation of authorship.  Other countries have fair quotation, parody, other things that here are mostly done through fair use to promote ongoing authorship.  Fostering public interest in free expression/access to information is another goal, and many countries have exceptions for news reporting, current political events and the like—again fair use does most of that.  A cluster of exceptions: those that protect privacy, autonomy, and ownership of consumers—fair use does some work, exhaustion does more, and §120 allows you to remodel your house even under architectural works protection. 

Most exceptions in national copyright laws are aimed to fulfill social policy goals—libraries, access for disabled.  Often have a remuneration obligation from the user community, present here in AHRA.  We have a number of exemptions that enable public institutions to functions—police investigation, etc.  Many countries have specific exceptions to allow that. There are an astonishing number of litigation fair use cases here.  Other countries have exceptions for things like reverse engineering.

Incidental use: fair use; ephemeral copies for broadcasters in §112 also found in other countries’ laws.  Euro. InfoSoc Directive has mandatory exemption for incidental uses with no independent economic significance.  Most interesting exemptions: those designed to cure market failures—compulsory licenses in §115 etc.  Starting to think about whether we have exemptions in US law because we don’t have collecting societies.  We exempt certain things and leave everything else to fair use instead, which is why educational use is still so controversial.

Many politically expedient exemptions—agricultural fairs, Shriners’ conventions, etc.  What about flexibility for adaptation in times of rapid change?  Some exemptions have more than one justification: §§110-111 is partly for innovation/software, but also partly for people who watch movies in the way they want to watch them.  Her categorizations seem to work for other countries too.  But: Dutch exemption for funerals—for the little people?

Core interest: what exemptions should exist in Next Great Copyright Act?

Mike Carroll: Doesn’t think the burden is that high for limitations and exemptions, because Act is designed to be overbroad and uniform across all forms of creativity, knowing that in fact our policies should not be that undifferentiated. Different means of tailoring exist—legislative as well as judicial (Altai).  Interesting problem is to do a policy audit, of which this is a version.  When you’re overbroad, you get a signal when people start to feel the pinch. When you say all of this is inefficient rentseeking, that’s not true, but some of it clearly is. Which is the right response v. people who just don’t want to pay.

A: yes, you see examples where there’s outrage over enforcement—WSJ’s anger about ASCAP’s efforts to get Girl Scouts to pay royalties for singing at camp.  Girl Scouts weren’t represented at the revision of the Copyright Act and couldn’t conceive that it would apply to them.

Q: exceptions for public institutions—isn’t that part of sovereign immunity?

A: not how other countries have conceptualized it.

Madison: you have a hodgepodge of categories.  Can you distill a smaller number?  Some are specifically designed to advance ©’s policy goals w/r/t creativity, and others designed to avoid spillover harms to adjacent regimes/public practices (privacy, avoiding conflict w/patent).

A: one of the risks of a project like this: looking at what we had from the bottom up. Unsurprisingly the overarching themes are evolving.

Josh Sarnoff: if existing subsidies of © are so excessive, how do we figure out where it’s really too much and cut back.

A: 1909 Act did try to figure out what works were protected and what was normal exploitation, and design rights to protect normal exploitation. 1976 Act: effort to Europeanize, grant broad rights w/exceptions, and they weren’t very systematic.

My unasked Q: Do other countries (than the Netherlands) have these weird politically expedient exemptions for the equivalent of the Shriners?


  1. I think you are missing the point about what makes this installation art. It's not the trailer itself that is copyrightable anymore than the paint with which an artist paints is copyrightable. The trailer is only copyright worthy in that space at that time - that is it's fixation. Or the embodiment of the idea. I think a better right - also elusive in the US would be akin to performer's performance right...

  2. I don't disagree that it's art at all! I don't, however, think that all art has to be copyrightable.

    I can't see the placement as fixation, and I also can't distinguish that "fixation" from the pile of bricks that languishes for a week on my driveway before it's incorporated into my new addition. What makes the placement art is the idea, but that's not copyrightable.

    Further, I think undesirable policy consequences would follow from finding such a work to be copyrightable, particularly with respect to photographs of the world around us, which would risk infringing if such installations (and anything else that was deliberately arranged in space) were considered copyrightable.