Several months after requesting an extension to file a petition for writ of certiorari (request for Supreme Court review), Google today announced the filing of that petition today (one day before the extended deadline), and published the document (PDF).
As I suspected, Google is seeking a review of both the Federal Circuit's 2014 copyrightability finding and the same appeals court's 2018 holding that the way Google incorporated many thousands of lines of Java API declaring code does not constitute fair use.
My positions on the issues have not changed. However, I do keep my fingers crossed for Google's petition, and here's why:
At the cert stage, it's not about whether a petitioner's theories have merit. It's about whether the issues raised are important enough to warrant review because clarification is needed.
I want Google to win at the cert stage, even on both counts (copyrightability and fair use), but then I hope (and think it's very likely) that the Supreme Court of the United States will affirm the Federal Circuit's well-reasoned opinions, which are good for innovation.
No one can blame Oracle for preferring that the Supreme Court deny the petition: then the case, filed in 2010, could finally proceed to the remedies stage, where it would have been a long time ago if not for totally outrageous decisions by Judge William H. Alsup in San Francisco.
But Oracle's interest in saving time, no matter how understandable and legitimate, doesn't make the issues any less important.
If the Federal Circuit's decisions had been handed down by the United States Court of Appeals for the Ninth Circuit, I'd still prefer a ruling by the top U.S. court, but then the benefit of a denial of cert would at least be legal certainty in the most important circuit.
Here, however, the case went to the Federal Circuit because Oracle had originally also asserted software patents, which went nowhere but made this a case that must be appealed to the Federal Circuit and not the regional circuit (here, the Ninth Circuit).
In such a situation, the Federal Circuit applies the regional circuit's law. But it doesn't make or modify another circuit's case law. As a result, those two beautiful Oracle v. Google appellate opinions--which I welcomed as an app developer and trivia content author (the latter is also a field where the fact-expression dichotomy plays a role)--aren't really binding on anyone. In theory, even the Federal Circuit, next time it applies Ninth Circuit law, could easily take a new position (though it probably wouldn't).
This means that if the Supreme Court didn't agree to hear the matter, all the people I disagree with (respectfully in most cases, though admittedly not in all) would keep running around telling everybody that the Oracle v. Google appellate rulings aren't binding.
I have confidence in the Supreme Court, especially in its conservative majority, protecting us software developers. Oracle's appellate counsel, Orrick Herrington Sutcliffe's Joshua Rosenkranz, and his team are so great they can do it again.
But even if, contrary to what I expect, the Supreme Court sided with Google on one or both issues, I'd rather have clarity on that basis than a situation where some are still going to point to Ninth Circuit rulings that I believe don't support those other people's positions at a closer look.
I want strong software copyright, and I view software as something where it's very hard to imagine "fair use" when significant amounts of code (not just a dozen bytes) are distributed in large numbers (as opposed to a few private copies). So I hope the Supreme Court will grant Google's petitions on both counts and then affirm the Federal Circuit 100%.
Neither copyrightability nor fair use are, in my view, the appropriate level at which to ensure software interoperability. Trying to protect interoperability at those levels would throw out the baby with the bath water and do too much harm to software copyright, which someone who's invested and keeps investing in software development doesn't want.
If, when and where there is a case where an API (Application Programming Interface) really must be available to other parties even without the copyright holder being prepared to extend a license, the solution is a compulsory license under antitrust law on FRAND (fair, reasonable and non-discriminatory) terms. There are standard-essential patents; in Germany and possibly some other jurisdictions, they additionally have standard-essential utility models; and there's no reason why the same FRAND rules wouldn't and couldn't apply just as well to standard-essential copyrighted works.
Share with other professionals via LinkedIn: