5 ms·
A few thoughts: 1. This is a meticulously researched, marvelously analyzed, and brilliantly synthesized order done by a judge who has a keen grasp of not just
by grellas 14y ago
A few thoughts:
1. This is a meticulously researched, marvelously analyzed, and brilliantly synthesized order done by a judge who has a keen grasp of not just the facts of the case but of those that really matter. As the opinion notes, this was a case of "first impression" - meaning that no published decision has ever dealt specifically with the precise question raised of whether APIs in themselves are protectable by copyright or not.
2. The developer community needed a definitive ruling that was almost impossible to arrive at in light of the controlling precedents of the Ninth Circuit (which are binding on the judge). Why? Because (beginning with the Johnson Controls case in 1989) the precedents had held that the "structure, sequence, and organization" (SSO) of any part of a software program were potentially protectable by copyright and that the issues had to be determined case-by-case to determine whether a particular component was or was not protectable. Thus, the best a trial judge can do in such a case is to make a fact-specific conclusion about the case immediately being tried, one which would have limited impact in the next case, where the parties could argue the same issue on different facts. Yet, while doing just that and limiting his ruling to the particular facts before him, Judge Alsup has provided a definitive and logically compelling approach to how such issues are to be decided where they concern APIs and copyright and such reasoning is, in my view, destined to be widely applied throughout the court system going forward. Lower court rulings can have a powerful impact through the sheer force of their reasoning. This is just such a ruling. It is rare to get this. It could not have been better timed on a vital issue affecting interoperability in our modern era.
3. Fundamentally, Oracle had been arguing that the SSO of its 37 API packages reflected creative expression of precisely the type that the Copyright Act was intended to protect. And it is true that API design choices reflect all sorts of creative forms of expression. To deal with this issue, the judge got down to fundamentals, with the key language found at page 35 of the opinion: "Much of Oracle’s evidence at trial went to show that the design of methods in an API was a creative endeavor. Of course, that is true. Inventing a new method to deliver a new output
can be creative, even inventive, including the choices of inputs needed and outputs returned. The same is true for classes. But such inventions — at the concept and functionality level — are protectable only under the Patent Act. The Patent and Trademark Office examines such
inventions for validity and if the patent is allowed, it lasts for twenty years. Based on a single implementation, Oracle would bypass this entire patent scheme and claim ownership over any and all ways to carry out methods for 95 years — without any vetting by the Copyright Office
of the type required for patents. This order holds that, under the Copyright Act, no matter how creative or imaginative a Java method specification may be, the entire world is entitled to use the same method specification (inputs, outputs, parameters) so long as the line-by-line
implementations are different" (my emphasis). Thus, even though the judge was forced by Ninth Circuit precedents to assess the SSO based on the particular facts before him only, he did so by finding, as a matter of fact, that only a 3% layer of code dealt with the SSO of the API packages, that this consisted entirely of names that had to be identical for compatibility purposes and of tasks that could be performed in only one way in order to work, and in this way - having established the factual setting as consisting entirely of elements he concluded were unprotectable under copyright - he could make a powerful and sweeping statement of law that will undoubtedly have a huge impact on future cases.
4. The judge finally had to deal with the claim that the SSO constituted a sort of taxonomy that has been held protectable under copyright in other circuit courts. Here, too, he addressed the issue based on fundamentals: assuming that the API design structure here constituted a taxonomy, he nonetheless held that it was above all a command structure that was unprotectable based on 17 U.S.C. section 102(b) (which categorically excludes ideas, concepts, etc. from copyright protection). Thus (at page 39): "In our circuit, the structure, sequence and organization of a computer program may (or may not) qualify as a protectable element depending on the 'particular facts of each case' and always subject to exclusion of unprotectable elements. Johnson Controls v. Phoenix Control Sys., 886 F.2d 1173, 1175 (9th Cir. 1989). Contrary to Oracle, Johnson Controls did not hold
that all structure, sequence and organization in all computer programs are within the protection of a copyright." This is another way of saying that what might otherwise constitute protecable expression (e.g., the creative design choices made in developing the SSO of APIs) is nonetheless not protectable if it functionally operates within a computer program as a command structure. This, of course, is what APIs do and this means that his conclusion can be used as a powerful guide in all future API copyright cases. As a trial court decision, this ruling is not binding on other courts, yet it is compelling and persuasive, which can amount to the same thing (including in its impact on the Ninth Circuit when it considers this case on appeal).
5. Oracle here is like the Black Knight in Monty Python: as each limb of this case was lopped off, it dismissively would say "a mere flesh wound." Now that its case has been reduced to a final stump, it can continue to declaim but who will listen? Maybe a long shot on appeal but I wouldn't hold my breath. This is an unbelievable outcome that illustrates that great good can come even from lousy things that people do. The computing world owes a great vote of thanks to Judge Alsup: the cause of interoperability has won a huge victory.
- jhund 14y agoThanks to Judge Alsup for helping the cause of interoperability and thanks to grellas for a great writeup with Monty Python analogy!
- gbhn 14y agoseconded
- Natsu 14y agoI remember you saying that this was clearly a judge who did his homework. I'm glad to see that he took his time and came up with the well-reasoned judgement I was hoping for. Oracle's case has been coming undone for some time now. I couldn't believe the way their expert witness was rebuked in the denial of Oracle's JMOL motion the other day, either. They have, what? One claim for copyright infringement via those test files? I sincerely doubt that even the maximum penalty there would permit them to recoup the cost of this litigation. And I think you've said as much before.
- fpgeek 14y agoThe statutory penalty for the test files and rangeCheck is $150k per incident or $300k (which is nowhere near the $50 million estimated cost of the litigation for each party to date). However, Oracle has elected to go after "infringer's profits" instead of taking statutory damages, so it is very possible that they will end up with less. It's very unlikely that they'd end up with more as the judge understands how trivial both parts are (and Oracle and Google have agreed to let him decide damages).
- bornhuetter 14y agoWow, great analysis. Thanks.
- scott_s 14y agoPlease seriously consider starting a blog. Even if you post just once every two months, after major rulings, your analysis is always so illuminating. I come into legal threads on HN specifically to read what I know will be your top-rated comment.