Yes, The Appeals Court Got Basically Everything Wrong In Deciding API's Are Covered By Copyright
from the fundamentally-not-understanding-that-apis-are-not-software dept
Copyright expert and professor Pam Samuelson, one of the most respected scholars of copyright law, has published a short paper explaining what she calls the “three fundamental flaws in CAFC’s Oracle v. Google decision.” As you may recall, that ruling was a complete disaster, overturning a lower court decision that noted that application programming interfaces (APIs) are not copyrightable, because Section 102 of the Copyright Act pretty clearly says that:
In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.
But CAFC got super confused, and basically ignored 102 while misunderstanding what an API actually is. After the White House itself got confused, the Supreme Court refused to hear the case. This means that the CAFC ruling stays in place, despite it being at odds with lots of other courts. And this might not be a huge problem, since most copyright cases won’t go to CAFC. The only reason the Oracle case went to CAFC was because it started out as a patent case, and CAFC gets all patent appeals, even if the appeal has nothing to do with patents. Except… of course, now there’s incentive to toss in a bogus patent complaint along with a questionable “interface copyright” complaint just to get it into CAFC’s jurisdiction.
Samuelson’s paper is a good read (and we’ll get to it), but I’d actually argue it’s a bit too tame, and leaves out the really fundamental flaw in the CAFC ruling and in the White House brief: these non-programmers don’t realize that an API is not software. Almost all of the mistakes stem from this simple fact. They assume that an API is software. And this is highlighted very clearly in the CAFC ruling where they quote Pam Samuelson out of context and then completely miss what she’s actually saying. Here’s from that ruling:
Google argues that ?[a]fter Sega, developers could no longer hope to protect [software] interfaces by copyright . . . Sega signaled that the only reliable means for protecting the functional requirements for achieving interoperability was by patenting them.? … (quoting Pamela Samuelson, Are Patents on Interfaces Impeding Interoperability…). And, Google relies heavily on articles written by Professor Pamela Samuelson, who has argued that ?it would be best for a commission of computer program experts to draft a new form of intellectual property law for machine-readable programs.? Pamela Samuelson, CONTU Revisited: The Case Against Copyright Protection for Computer Programs in Machine-Readable Form…. Professor Samuelson has more recently argued that ?Altai and Sega contributed to the eventual shift away from claims of copyright in program interfaces and toward reliance on patent protection. Patent protection also became more plausible and attractive as the courts became more receptive to software patents.?…
Although Google, and the authority on which it relies, seem to suggest that software is or should be entitled to protection only under patent law?not copyright law? several commentators have recently argued the exact opposite. See Technology Quarterly, Stalking Trolls, ECONOMIST, Mar. 8, 2014, http://www.economist. com/news/technology-quarterly/21598321-intellectualproperty- after-being-blamed-stymying-innovationamerica- vague (?[M]any innovators have argued that the electronics and software industries would flourish if companies trying to bring new technology (software innovations included) to market did not have to worry about being sued for infringing thousands of absurd patents at every turn. A perfectly adequate means of protecting and rewarding software developers for their ingenuity has existed for over 300 years. It is called copyright.?); Timothy B. Lee, Will the Supreme Court save us from software patents?, WASH. POST, Feb. 26, 2014, 1:13 PM, http://www.washingtonpost.com/blogs/the-switch/wp/ 2014/02/26/will-the-supreme-court-save-us-from-softwarepatents/ (?If you write a book or a song, you can get copyright protection for it. If you invent a new pill or a better mousetrap, you can get a patent on it. But for the last two decades, software has had the distinction of being potentially eligible for both copyright and patent protection. Critics say that?s a mistake. They argue that the complex and expensive patent system is a terrible fit for the fast-moving software industry. And they argue that patent protection is unnecessary because software innovators already have copyright protection available.?).
But this is just wrong. If you actually look at Samuelson’s quotes, she’s talking about interfaces not software. Notice in every quote she is not actually talking about the software itself, but “interfaces,” “functional requirements” and “program interfaces.” The absolute worst is the first quote, where Samuelson writes “interfaces” and CAFC inserts a “[software]” to imply that it’s the same thing. It’s not. The two paragraphs are not actually at odds. It is entirely reasonable to argue that interfaces shouldn’t be protected by copyright (thanks to Section 102) and that software should not be patentable.
It only looks like they’re disagreeing if you’re confused and you think that an API is the same thing as the software itself. But that’s like saying a recipe is the same as a meal or that a dictionary is the same as a novel that uses those words. It’s not the same thing.
So while Samuelson’s new paper is great, I still feel like she holds back on that key issue, which is so just blatantly wrong, and seems to underline why non-technical people (including the judges in this case) got so confused. Of course software is copyrightable. The argument is over whether or not an API necessary for interoperability is copyrightable. And, as Samuelson’s paper notes, it had been widely accepted prior to the CAFC ruling that the answer is no because they’re “procedures, processes, systems and methods” under Section 102.
A second flaw was the CAFC?s overbroad view of the extent to which the ?structure, sequence and organization? (SSO) of computer programs are protectable by copyright law. During the 1980s, some courts regarded program SSO as having a broad scope of protection under copyright law. But in the last two and a half decades, courts and commentators have recognized that the SSO concept is too imprecise and misleading to be useful in software copyright cases. The SSO concept does not help courts make appropriate distinctions between protectable and unprotectable structural elements of programs. Procedures, processes, systems, and methods of operation, almost by definition, contribute to the SSO of programs that embody them. However, this does not make those elements protectable by copyright. The design of many program structures, including APIs, is inherently functional and aimed at achieving technical goals of efficiency. This disqualifies them as protectable expression under U.S. law.
Anyway, the rest of the paper is a good read, and hopefully it means that eventually this issue will get back to the Supreme Court — and one hopes, at that time, someone can at least get through to them that an API is not software.