Commentator: Matthew Sag
Sag: what are design patents? One practitioner he knows sells them to clients as another form of TM. Risch shows how the infringement standard could be improved by applying copyright rules and by applying his economic model. Sag will criticize the use of economic models in this type of paper and how copyright informs the discussion.
Economic models are often useful to clarify one’s own thinking, but then they should be left out. The model here shows that costs are bad and that all choices have tradeoffs. Model says we should look at issue from the perspective of the time that an actor is thinking of copying. But the model doesn’t explain why this is so and he couldn’t distinguish this from the general “we should maximize social welfare.”
The paper, however, teaches a lot. Design patents are easy to get because the novelty bar is set really low, and courts still use the teaching/motivation etc. test. Image plus a display is patentable: courts are treating ephemeral displays, not always part of the object, as patentable, which is necessary for GUI patents. They protect look & feel in exactly the way of the old copyright cases, but without copyright’s filter for functionality or its various limitations and defenses. Can be used to protect function because of the way the infringement test is structured: doesn’t require nearly exact copying. If juries aren’t given strict instructions, as mostly they aren’t, they’ll be comparing general appearance including functional elements; and after Egyptian Goddess they’ll protect designs not at the point of novelty.
Apple v. Samsung: the phones are a lot alike, but they are similar. Once you start abstracting away for what’s protectable, you get into areas of function rather than the proper ambit of design. Apple’s complaint was that Samsung used features that were largely inevitable/attractive. By using abstraction, you ensure an infringement test that covers function and not just aesthetics.
History of copyright provides lessons: Courts disagree on substantial similarity and how to filter, but at least in a case like Apple they would always filter. And they take account of user demand in determining functionality, per Lotus v. Borland.
Paper should consider the static/dynamic tradeoff: Apple would argue that it deserves to be able to charge a huge amount because it took the risks/made the investment. Would also benefit from broader discussion of economics of design patents: they’re weird hybrids where you get briefly very strong copyright/TM-like protection better than either of them. If we should have these at all, why? That’s where the economic analysis should take place.
Risch: often gets the static v. dynamic question. The reason he used the delayed ex ante view is so that the court can assess efficiency of that sort at the time of the reuse. If it’s early on, at that point the court might say there hasn’t been enough time to recoup investment and it’s not a standard, and therefore it can’t be reused. But if it’s been several years, as in Apple v. Samsung, then it can be time to allow reuse. It’s problematic that Apple claimed infringement because the ideas were the same. Copyright has the tools to deal with that.
You could take a photo of this room and get a design patent on it. People are wearing colorful outfits, they have heads of different shapes—and now no one else can take a photo of people wearing colorful outfits in a room like this? That can’t be the case. Basic takeaway: when comparing, there should be filtration. Judges do this when they’re decisionmakers, but they worry it will confuse the jury so they just give the designs to the jury without more and ask “would you think the person who created X created Y?” We need to consider standards, customer switching costs, slavish copying v. nonslavish.
Q: would better written claiming identifying what’s claimed would help?
A: the claim is always the design as shown in the figure.
Q: but you never have to point to the novel part.
A: so reverse Egyptian Goddess? Egyptian Goddess endorses filtering, but doesn’t require—it’s up to the dct’s discretion to decide how to deal with prior art.
Q: the infringement standard—would the same person have made these—sounds a lot like TM. Why not aesthetic functionality?
A: yes, it’s like TM infringement, but there’s no class of goods. It’s a hybrid of substantial similarity and likely confusion.
Q: point out that Apple is an outlier—just convinced jury that Samsung was a dirty, dirty copier. Shouldn’t extrapolate—but then again it’s an early GUI case.
A: believes that if Samsung had applied for its own design patent, it would’ve issued without an office action!
Mark McKenna and Katherine Strandburg, Progress and Competition in Design
Commentator: Shubha Ghosh
Trying to explain works where function and aesthetics are integrated. Trying to fit design patents into copyright/TM scheme, where we know functional elements of design are excluded. Barton Beebe: in copyright, there is no good theory of progress. Tends to be about diversity: having something new is progress. Apply those insights to design patent—but what are these works that integrate aesthetics and function? Architecture—you have an incentive/creativity process that works to integrate the two. Bathroom fixtures—might not fit copyright or trademark, but someone has to figure out what bathroom fixtures should look like, and they should assist the user while also being aesthetic. Existing law might not deal with bathroom fixtures.
Paper shouldn’t assume that copyright/TM have it right in terms of design. PGS doctrine in copyright is not very helpful. Nothing but ad hocery. Same with TM: Taco Cabana and Wal-Mart as a series of mistakes. Subtext tends to be “why did we grant cert?”
Progress is just a question of institutional choice. Should we defer to the political process/Congress or not?
Inclination: design patent was just a compromise. If you try to explain some ideal system, why? If you’re trying to explain the law as it is, then it’s just Congressional dealmaking.
McKenna: this isn’t an attempt to explain why we have design patents, but rather a piece saying that if we have design patents, the only coherent way to think about them is as follows. Patent like rights for designs that integrate form and function: if you’re going to have a system that looks like a patent system, an indispensable component is a way to measure progress over time, even if not in any given specific case. You can’t talk about progress in design the way you can talk about progress in the speed of computers. So we need to be able to say that integration of form and function can be measured; if it can’t be measured, then the case for design patent is not good.
Strandburg: what makes a system a patent system, as opposed to something else, you need obviousness/nonobviousness and an idea of cumulative progress. Maybe the copyright system is doing a bad job; we’re not saying it is, but we are saying that what patent systems do, with obviousness, is to incentivize cumulative progress.
McKenna: this is why obviousness in design patent is presently a mess—we don’t have any idea what it would mean to be an improvement in design. The idea of integration came from what we see over the cycle of history—worry that people are using design patents as a backdoor to utility protection, and repeated attempts to solve that by creating invention/functionality doctrines that all fail because they require courts to separate form and function and that doesn’t work. All design integrates form and function, some more obviously and some better than others; that’s why courts can’t deal with functionality, because it requires them to separate the inseparable.
One thing that makes it hard to get the rules right is that we don’t know what kinds of design we want to produce. If you don’t have a way to measure the costs, competitively, of the system you run into trouble. Progress is a proxy for costs and benefits. It’s hard to imagine telling the court how to balance costs and benefits without knowing what we want to incentivize.
Strandburg: we believe that designers do care about integration, creating a space where it’s possible to have incentives.
Wendy Gordon: what if the person contesting the patent proves there’s a functional advance? Shouldn’t that exclude design patentability? But you assume we know what progress means for utility patents—if that were so, we’d know what should be in design and what should be in utility. But courts say they don’t know what will be valuable in the future—extreme agnosticism in utility.
Strandburg: there’s a distinction between the goal of the patent system and the implementation of nonobviousness. The goal, from the Constitution and elsewhere, is better technology. But we don’t always know what will be more valuable/better. So what we’ve tried to do, not totally successfully, is require something that takes a step of a certain size, a step that competition alone wouldn’t spur. The doctrine doesn’t ask “is this better,” but rather by giving patents only on things that take a big enough step, we’ll eventually end up with better. Whereas we don’t seek “better” with copyright.
Gordon: disagrees. There are good and bad shower designs.
Strandburg: but that’s functionality. Courts don’t want to say that there are good and bad plays. (RT: or rather, they want to ostentatiously announce their aesthetic judgments and then assure us that those judgments don’t affect their rulings.)
Barton: paper seems to assume that integrating form and function is aesthetic progress. What if that’s not true? One could say that this version of aesthetic advance is very historically situated. Compare Art Nouveau: one could say that such designs were much better than Apple’s minimalism. Less isn’t more; less is less—nothing to do with function. So he wants to push back on this definition of progress.
McKenna: we don’t assume integration is a form of aesthetic progress. But it’s possible that one could measure progress through integration, and that might be a goal that isn’t fulfilled well by copyright/TM etc. and therefore represents an appropriate role for design patents if this can be measured. If the only thing you can get out of design patents is new aesthetics, then let copyright take care of it.
Ghosh: every field might have its own definition of progress.
McKenna: where we are is that he doesn’t trust courts to filter out functionality. His view: avoid that by asking a different question.
Strandburg: in the market, there’s no distinction between Traffix and aesthetic functionality. So you have to think about what you want dynamically, since static analysis won’t help you at all.
Risch: how do you deal with the statute’s explicit embrace of surface ornamentation?
McKenna: that’s written into the patent statute because copyright doesn’t cover it at the time, but that doesn’t belong to his conception of what design patent could coherently be for—it should be moved. The paper is theoretical and doesn’t attempt to justify the current statute.
Strandburg is agnostic about moving to more copyright for design; McKenna worries about copyright’s duration, but we still need to figure out what kinds of investments count for what it is we’re trying to incentivize to figure out whether any system makes sense.
Strandburg: that’s also a question about what copyright is doing. Design probably is not different from everything else within copyright’s subject matter. Maybe overall the copyright system is borked and we’re focusing on design because we have the opportunity to do so.