Notre Dame Law Review Symposium, Negotiating IP’s Boundaries in an Evolving World

Panel One – Moderated by Joseph D. Bauer

Mark P. McKenna, Notre Dame Law School & Lucas S. Osborn, Campbell Law School, “Trademarks and Digital Goods”

BMW sued Turbosquid for hosting digital models of BMW cars—models could be used in video games or movies. Digital models are different from tangible models.  Music used to be sold in tangible form, and isn’t much any more.  Karaoke files; software used to come on floppy discs and CDs.  3D printing takes digitization to a new level.  Print out a purse, a Star Wars lamp.  Things that implicate TM law and may lead to consumer confusion.  But about what?

Insofar as these cases involve claims about use of TM w/in digital goods or to designate them, they run headlong into Dastar.  Dastar couldn’t have been clearer: “origin” means physical origin, not creative or intangible content.  Courts have struggled with the case; among other things, figuring out what counts as “goods.”  Gensler v. Strabala: Gensler sued a former employee who advertised himself as having been the designer of 5 buildings that Gensler was the architect of record for; argued that arch. works were group collaborative efforts. 7^th Cir. says this is a claim about the origin of services, and so not like Dastar: a false claim of creation of the IP, the designs—which is exactly what Dastar tried to rule out and was exactly what Fox too alleged in that case.  Cisco Tech. v. Certification Trends: Cisco created practice exams to certify people as Cisco experts; the D produced its own versions, and Cisco claimed that reuse + insertion of additional material was false designation of origin. Court thought this turned on whether D added content, and if D created a new good by adding content it would win, whereas mere repackaging/reuse of Cisco content would be actionable.  This can’t possibly be consistent w/Dastar’s rejection of “mutant copyright.”

Sleptone shows increasing challenges in digital area.  Sleptone creates karaoke tracks.  Sued restaurants for unauthorized copies.  Often claim that they made backup copies in violation of Sleptone’s policies.  Embedded in the file is the Sound Choice mark. Early on, Sleptone claimed false designation of authorship of the tracks; some courts were persuaded, in part b/c they had a hard time figuring out the goods were, and said that new copies were new goods b/c they are digital versions. Got more sophisticated over time: Sleptone argued that confusion was over origin of tangible digital file, even though that was never presented to consumers.  Seventh Circuit rejected that reasoning: intangible files are immaterial b/c consumers never interact w/them. That suggests an important doctrinal refinement: focus on tangible goods has to be layered w/Q of whether the alleged confusion relates to the tangibility of the good, rather than mere “one from column A, one from column B” approach. 

One could imagine saying that digital files are intangible, or that every file must be instantiated in memory and thus they are always tangible.  We think both approaches are too simplistic. Analysis needs to be informed by TM policy and layering approach. ID’ing tangible file is a step, but we need to ask what consumer cares about w/r/t the tangibility.  Downloading BMW model from Turbosquid, or 3D printing site: these sites say who created the file.  If it turns out to be crappy, will I blame BMW, or user 1234 or Turbosquid? That’s where source understanding applies.  [Fan fiction literature has a lot on this: understanding of origin of the particular creative work.] Doesn’t exclude digital goods altogether.  If the website looks official/officially endorsed, that could be a problem.

Role of TM law in digital world: traditional orientation was physical goods, not new to Dastar.  Not to say that TM never covers a broader view, and over time TM has figured out how to assimilate service marks, assignment and licensing.  Consumers might sometimes get valuable info about origin of intangibles, but one lesson of attempts to adapt is that when TM law has done so, it’s had real difficulty drawing lines distinguishing itself from utility patent, design patent, and ©.  Application to digital goods brings all those concerns to the fore.  TM theory: Info about the source of goods, primarily so consumers can set expectations about physical quality of goods—incentive to offer goods of consistent quality. Especially important for goods w/unobservable characteristics. That presumes that mark owner produces or stands behind production of instantiation, but digital tech could radically separate design and production.  [I’m reminded of Real Genius: would you call that a design flaw, or an implementation flaw?]

There’s a decent argument that TM doesn’t have much to say about 3D printing, if producers are now just designers w/little or no involvement in tangible production.  Not clear consumer expectations about quality of physical goods have much connection to quality of intangible content. At least that’s the Q TM should be asking. In some cases consumers might value content, and might care who created it very much; puts pressure on right of attribution. But interest in knowing who created the design isn’t the same as interest in who created the particular copy, and that matters if we want to maintain a distinction b/t TM and ©/design patent; those are the regimes charged w/dealing w/content itself, rather than source designation of content.

Perry Saidman, Saidman Design Law Group, “Design Patent Functionality”

The line b/t design & utility patent: thesis is that there should be no line.  Functionality comes into validity and infringement contexts of design patent. The statute says a design must be ornamental.  Courts used to focus on defining ornamental: Learned Hand said we must find at least a rudimentary aesthetic appeal.  Like low bar for originality in ©.  Functionality got to be the opposite side of ornamental coin, perhaps b/c they were both in the patent statute. 

Functional is either “de facto functional”—performs a function—or “de jure functional”—functional legally and not capable of being protection.  He suggests “utilitarian” for “de jure functional,” b/c now courts are constantly confusing them. 

Current rule: design patent on utilitarian features is a danger against which we should protect.  High Point Design (Fed. Cir. 2013): trial court said all major characteristics of fuzzy slipper were functional.  Fed. Cir. said that serving function was a prerequisite for design patent: design has to be for an article of manufacture, and articles of manufacture have functions. So de facto functionality/feature was required. Functionality of article isn’t the same as de jure functionality of its design.  He posits: consider underlying purpose of functionality doctrine, to prevent monopolization of utilitarian features that can only be monopolized w/utility patent.  [I think this understates the importance of defining what’s not patentable—there are useful features that are not eligible for utility patents b/c not sufficiently novel, but also that should be committed to the public domain for that reason—defines what must be free to use, not just what is appropriable.  So it’s “can only be monopolized, if at all, w/utility patent” that ought to be our standard.]

If there are alternative designs that perform substantially the same function as the patented design, the design doesn’t monopolize that function.  But then you get questions: what is the function? Claimant will define it broadly: e.g., connecting two devices.  Accused infringer: define it as making a more specific kind of connection.  Egyptian Goddess got rid of point of novelty b/c P and D would always define the point of novelty differently.  The word function is the same in his opinion.  Manipulability is limited only by the skill of the attorneys involved, rendering alternative design test meaningless.  So we should reject the idea that we need to prevent design patents from having utilitarian features. Then there’s no need for a functionality doctrine.  All utilitarian features have an associated appearance; it’s the appearance of those features, whether utilitarian or not, that matters in validity determinations, w/o need for functionality doctrine.

On infringement side, there’s been a ridiculous practice of factoring out functional features in Markman hearings, which makes no sense.  Comparing only ornamental features makes no sense.  The rationale is that you shouldn’t use design patent to protect ideas or inventions.  Design patent doesn’t protect general idea; has to look like the patented design. Richardson case involving utilitarian features, all dictated by functional purpose.  Fed. Cir. made fatal statement: DCt properly factored out functional aspects as part of claim construction. But this is wrong.  If the accused design doesn’t look like the patented design, factoring is unnecessary.  Confirmed this year in a flotation device case: functional elements of Richardson had design dictated by their functional purposes, but still had appearance aspects. Discussing Ethicon: Certain elements of a surgical device can be functional but functionality doesn’t preclude them from having a protectable appearance. Don’t entirely eliminate a structural element from the claimed design, even if de facto functional.  All claimed designs have functions, all features have appearance, and it’s the appearance that matters in validity and infringement; functionality is dead.

Pamela Samuelson, UC Berkeley School of Law, “Strategies for Discerning the Proper Boundaries of Copyright and Patent Protections”

Motivation arose in computer software, back from Whelen v. Jaslow saying that everything was protectable unless there was only one way to achieve a function.  That was the thickest scope of © ever—one idea per program. That couldn’t be right. She read the software patent cases and realized they were screwed up too.  Initial impetus: you don’t need to protect all w/© b/c patent can play a role, but she showed how incoherent the court of appeals was in software patent space.

90s cases started to recognize role for patent and for ©.  Expression v. process/functionality, but those are hard to separate. Some courts rejected © claims where they perceived attempt to get patent like protection. Fed. Cir. in Oracle v. Google endorsed complete overlap b/t patent and ©, invoking Mazer v. Stein dicta. 

Samuelson believes in categorical exclusivity. Art. 1, §1, cl. 8 distinguishes b/t authors and inventors.  Baker v. Selden says © protects explanations and depictions of useful art but not for useful art itself. Sec. 102(b) embodies this distinction b/c patentable subject matter includes processes, methods and systems and 102(b) excludes them from (c).  Mazer says utility patents and (c) are “mutually exclusive,” not same as design patent. Nimmer treatise since 1963 has said no problem w/overlapping subject matter; she thinks Nimmer is ignoring the real doctrine.  2016 version has changed its interpretation of Baker v. Selden, rejects election of protection theory but there’s something seriously misleading in the treatise about the case law.  Nimmer says there’s a case that suggests that (c) is available for the subject matter of an expired patent. That case stands for the opposite proposition: the only thing (c) protects is actually the pictorial elements.

Paper develops a more subtle approach for boundary line problems.  Layering/selection of elements theory is most common/promising, though election has a role.  Attractiveness of patent bargain/effectiveness in inducing disclosure depends almost entirely on the background of free competition; we worry about © giving longer rights w/o bother of application process, examination of the claim, etc.  We worry about people choosing © to get statutory damages or actual damages + disgorgement, and potential criminal liability (not there for patent).  Border control measures are another example of different treatment. 

Kohus: inventor used © to claim exclusive rights in an expired patent—a © registration for a baby swing.  Claimed it as a toy but sued people who made actual swings. CO should have cancelled the registration.  Conclusion: courts generally do a pretty good job managing the boundaries of © and utility patent and keeping them separate.  Arguments about what kinds of techniques courts can use: debunking Nimmer.

Rebecca L. Tushnet, Georgetown University Law Center, “Raising Walls Against Overlapping Rights: Copyright Preemption and the Right of Publicity”

Thanks for organizing/coming out.

I begin with the proposition that the right of publicity overlaps with trademark in its protections against false endorsement; with copyright in its supposed justifications in incentivizing performances; and with traditional privacy and defamation torts in protecting personal dignity and control over one’s own presentation of the self. Yet the right of publicity has been used to extend plaintiffs’ control over works and uses that don’t violate any of the rights with which it shares a justification.  This quicksilver nature is what makes the right of publicity so dangerous.

I will offer only one appalling recent example of the unchecked expansion of the right of publicity: In 2015, a district court allowed a right of publicity claim against a racing video game to proceed.   The theory of the case was that the game offered a visual representation of an actual racing track, including an image of a banner displaying the trademark of Virag, an Italian flooring company.  Virag is also the last name of Mirco Virag, one of Virag’s owners, and the court accepted the allegation that the Virag trademark was a “personification” of Mirco Virag.  As a result, although the First Amendment precluded a claim against the video game based on allegations of trademark infringement, a right of publicity claim survived based on the game’s alleged use of Mirco Virag’s identity—based on the same banner that triggered the First Amendment-barred confusion claim.   Even in such cases of flat-out conflict between trademark and the right of publicity’s respective First Amendment limits, courts have been unwilling to limit the right of publicity.

One reason for this expansion is that, without a coherent justification for the right of publicity, there are no obvious stopping points for its scope.  If the justification is merely about allowing a person control over unconsented uses of her image, then any uses or even reminders of her identity would violate the right, subject at most to the external constraint of others’ First Amendment rights to engage in speech about the world.   Following this logic, courts have generally only limited the right based either on statutory exclusions, which need no theories behind them,  or based on an often cramped view of the First Amendment.

What can be done?  My paper argues that recent copyright preemption cases have highlighted the potential for using preemption theories to create a sort of definition of the right of publicity by subtraction: where Congress has tried to achieve a particular aim, such as encouraging the production of creative works, in a particular way, then states should not be allowed to upset that balance by encouraging production in a different way.  Although better First Amendment treatment for the right of publicity still remains an important need where a speaker is creating a new depiction of a real person, preemption can define the boundaries of publicity rights where the challenged use is based on an existing copyrighted work in which the person voluntarily appeared.  In such cases, only an additional justification—such as consumer deception about false endorsement—can justify a separate right of publicity.

This purposive approach, applying limits to the right of publicity based on the extent to which it tries to further an acceptable purpose in the wrong way, can also guide us in finding proper First Amendment limits. The fact of the matter is that we have a long history of developing First Amendment limits to other torts based on the core justifications for the right of publicity—reputation has long been considered by the law of defamation, false endorsement by trademark law, and incentives by copyright law.  Even when there’s no copyright preemption, the First Amendment analysis should draw on those limits rather than treating the right of publicity as if it were a completely separate property right and thus entitled to be broader than the combination of those rights.

First Amendment, then, is an IP regime of its own.  It describes what may not be reduced to property: centrally, facts that have been made public.

This is not to advocate, as some cases have, case by case balancing. Categorical balancing can fit well with a purposive approach, once we define the right categories of speech.  I have argued that in general the appropriate category for the application of the right of publicity is in cases of misleading commercial advertising.  This may appear to some people to undervalue a celebrity’s dignity interests in avoiding commercialization, but I don’t think that interest is meaningfully different, for First Amendment purposes, than a dignity interest in avoiding being talked about, which celebrities and even ordinary folks don’t have.  I welcome your perhaps contrary views.

Samuelson: Saidman’s paper reminds her of effort on patent side to say patentable subject matter is too difficult, so there shouldn’t be limits on it.  She sees “ornamental” in the statute and she thinks it has to mean something. Maybe Congress should take ornamentality out and replace w/appearance, but doesn’t think the SCt would be as willing as the Fed. Cir. to ignore a word in the statute.

Saidman: Fed. Cir. has said it means the opposite of functionality.  Statutory subject matter: “original” has been ignored.  Ornamentality isn’t enough. Nobody has ever really dug into that. I’d like to think that ornamental is a minimum bar for aesthetic appeal; judges are very poor arbiters of what’s good looking. We’re supposed to protect articles of manufacture, so anything protectable by utility patent is also protectable by design patent.

RT: But I don’t think you even mean that: most clothing is not protectable by utility patent, but you think most clothing is protectable by design patent.  Sears/Compco: To some extent utility patent law must define not just what is protectable but what must be free for all to use.

Saidman: what’s not novel, what’s obvious, what’s in the public domain is unprotectable under either regime.  If novel and nonobvious, protectable by both utility and design patent.  [That would seem to me to suggest a vastly higher threshold for design patent than currently applied.  Or, I think, what he really means is that you’d then do the functionality analysis embedded into novelty/obviousness.]

Samuelson: Cts struggling w/1A as a limit on ROP.  Building into an adaptation of fair use: finding ways to limit, where commerciality, consent/lack of consent, etc. might be factors.  Courts would be more willing to play with that multifactor approach.

RT: Tried that, it’s a disaster.  Once you start covering noncommercial speech, courts say that realism is not transformative, but then news reports are anyway protected by the First Amendment. That’s not law in the sense of giving coherent reasons that have empirical or normative justifications.  © isn’t good at making these distinctions but that’s no reason to add another regime that isn’t.

McKenna: Saidman’s paper moves the question into scope rather than eliminating it; hides the fact that, at some point, it’s always a question of broad or narrow characterization.

Lemley: Back door utility patents as a risk.  The grant rate for design patent is over 99%; no real substantive examination in these cases. Should be worried about back door?

Saidman: no, b/c function is never protected, only appearance.  99% is exaggerated.  Comparison w/© and trade dress: design patents have a 15 year term, shorter than © and trade dress. Functionality in those regimes matters b/c of length of protection. 

Q: but functionality has to mean something for design patent. Industry standard of plug shape: can that be considered part of what’s protected?  That can’t be so.

Saidman: the particular connector in the actual case did have a unique appearance. Could show you 50 different nuts with threads inside that have to have certain diameter/distance; however, they all look different.  It’s that look that is protectable.  If everything was in broken lines except for the threads, I’d agree w/you, and it would be rejected, not b/c it’s functional but b/c it’s not novel/it’s obvious. Don’t need functionality to protect against that.