Pam Samuelson, Functional Compilations
Commentators: Zahr Said and Steve Yelderman
Said: Framing questions: what is functionality? Does it differ in compilations v. other things? Is it the same as the useful articles doctrine? How do owners/litigants determine and assess functionality? Paper provides coherent and cohesive account of caselaw but isn’t in the business of predicting how to do it going forward. Are there tools we can offer to courts/litigants to do this kind of filtering? Tort-based questions we might try.
Act of collecting is sometimes asserted as a creative endeavor. Sometimes collecting is expensive. Cartographic mapping of caselaw that allows us to see patterns. Taking the cases at face value: you don’t move from formalist to realist modes. Incentives for creating them are typically economic, practical. Dictated by function: compilations are dictated by purpose much more than romance (or choices are more constrained in fulfilling the purpose). Merger ought to have an even higher bar in the compilation side—would need something whimsical or arbitrary, not just exercise of choice. Exercise of skill and judgment as foundation for ©? If it’s testable as good or bad—if the person exercising judgment is subject to malpractice for screwing up—that might be a reason to exclude it from ©. Almost estoppel: if you’ve asserted value in market based on truth status (that could be actionable if relied on and false), you shouldn’t be able to claim that the feature is creative. Another question: if elements improve the compilation measurably, then they shouldn’t be protectable. Look to advertising law for standards re: materiality/measurability.
Steve Yelderman: Taxonomy he perceived was somewhat different from Samuelson’s. What’s a compilation? Is a ham sandwich a compilation? Seemed to define it in the negative in contradistinction to 8 work categories in 102(a). Compilations don’t come in until 103. But then again neither do derivative works, which can be AV works etc. So why can’t compilation be literary or PGS work? True that plaintiffs often look to compilation as fallback; but usually unprotectable not b/c missed one category but b/c it wasn’t original or was functional. Often the reason we look at compilation is b/c we already have functionality concerns looming. One consequence: what’s a compilation is a bit malleable. Labels? In many cited cases, courts aren’t explicitly using that language, e.g., instruction manual cases talk about 102(b) and systems, but not so much about compilations.
Distinct limitations, sometimes together and sometimes apart: presence/absence of Feistian originality/creativity. Is the act of creation the type © intends to incentivize, as opposed to sweat of brow. That’s about origins, backward-looking. Other concern is forward-looking: the need for others to access system/process in work after the fact of creation, often under 102(b)/Baker v. Selden. That’s what he thinks of as functionality proper. Recipes have Feistian creativity, but it’s still a list of instructions about how to recreate a dish.
How do concerns about small chunks carry over to the greater whole, or not? Some border crossings are legit. Feist explicitly contemplates that lack of originality at the level of original elements can be atoned for by the judgment required in compilation. Others not so much—yoga. It’s also possible that individual elements might be creative, but assembled into a system might be unprotectable as a system—this is what Delta Dental missed. The different causes for exclusion might influence how we think about the compilation problem.
Samuelson: interested in how “compilation” served as a backup for other claims. Agrees that compilations can be one of the 8 categories—the Copyright Office says this is required for copyrightability, as w/yoga case. Drug labels/instruction manuals: they’re on the fringes of compilation, but had enough in common w/other cases to include them. Also it’s the selection and arrangement of elements that gets attention in infringement analysis, and those are compilation words. This is the small change of ©, usually invisible, b/c not Disney or Google, but borderline.
Paper reacts to Oracle v. Google, b/c the software here is functional compilations—Fed. Cir. was wrong. Showing that functionality matters outside of PGS works, contra Easterbrook.
Lemley: what would change if the paper were “functional works” v. “functional compilations”? Real work is showing functionality limits © full stop.
Samuelson: that’s larger project.
McKenna: calling them compilations highlights the extent to which they’re only marginally protected in the first place.
Said: different notion of authorship, particularly the kinds of compilations we see litigated. Term of art in statute v. artistic process—there’s a tension in those two meanings, and problematizing them is a political intervention to software as a literary work. There’s a definition of compilation in the statute—gathering and then starting to make creative choices. Literary work: the whole notion that software is literary work is unjustified artifice in copyright’s own terms, that we have to live with going forward. Keep thinking about authorship.
Samuelson: everything about a program is structure, sequence, and organization. Functionality has to limit protection of compilation elements, but it’s invisible once you adopt SSO.
Buccafusco: much of the work Altai does in filtering out still doesn’t tell us what the protectable “nugget” is.
McKenna: does © have any theory of what belongs to utility patent? Samuelson’s paper shows that © has some instincts that aren’t fully articulated. We could then ask whether it matches up w/patent law’s actual domain as defined by patent law—which he thinks is way broader than what © says is patent law. Courts have instincts about what kinds of functionality count; rejecting certain kinds of arguments, but w/o robust way of talking about.
Lemley: something other than the communicative expression achieving the results: part of the definition of functional?
Beebe: But for the TM paper we began with, papers are unified by struggle to distinguish b/t worlds of subjective and objective. Modernity has been struggling w/that for centuries. Human freedom/objective necessity. Communication of free subjects—aesthetic in nature. Functional world: world of necessary, instrumental reason, compilations that don’t express pure freedom. What Drassinower talks about in © is very different—the utterly free speaker, while Samuelson talks about someone constrained by the objective world. ©’s fate is to address both, especially b/c software consists of language. If you are committed to preserving human subjectivity through ©, patent law (ironically?) offers limits.
Samuelson: Software IP lawyers say—w/demise of patentable subject matter after Alice, some say © needs to expand. But if it’s too abstract to be protected by a patent, it’s too abstract to be protected by ©. One reason Oracle v. Google is a bad precedent is that it encourages that kind of thinking.