Thursday, January 16, 2014

API copyrightability to be confirmed soon -- and that's great news for software developers

The most important smartphone IP ruling in the months ahead is going to be the opinion of the United States Court of Appeals for the Federal Circuit in the Oracle v. Google Android-Java copyright appeal. The appellate hearing took place on December 4, 2013, making March and April the most likely months for the issuance of the appellate decision.

The second-most interesting smartphone IP decision in the near term will also be handed down by the Federal Circuit. That ruling will relate to the Apple-Motorola "Posner case", involving FRAND issues and non-standard-essential patents such as the "Steve Jobs patent". That decision is almost overdue by now (more than four months after the appellate hearing), and will most likely strengthen Apple's position vis-à-vis Google.

Of course, a decision is only a decision when it issues. I know that some people are generally skeptical of conclusions drawn from what judges say at a hearing. While it's true that changes of mind sometimes occur after hearings, there are many cases in which a court indicates its inclination clearly enough that it's perfectly reasonable to predict the outcome (always recognizing the uncertainty that inevitably remains) on that basis. As far as the copyrightability of Oracle's Java API declaring code is concerned, I would go even further: it's practically inconceivable that the district court's non-copyrightability holding will be upheld considering that the circuit judges made perfectly clear that District Judge Alsup confused "fair use" and copyrightability issues and that Google's whole non-copyrightability theory, which Judge Alsup adopted in its entirety, rests on a complete misreading of two key cases (Sega and Sony). Simply put, just based on what the circuit judges said at the hearing, Google's (and Judge Alsup's) non-copyrightability theory is totally in ruins, and the rubble is being cleared away now as the appeals court is writing its decision. (Only with respect to "fair use", the outcome is impossible to predict.)

A settlement between now and the Federal Circuit decision would make a lot of sense. It would be out of character for Google not to take its chances, yet this result is more realistic than affirmance of the fundamentally erroneous non-copyrightability conclusion.

Even if the Federal Circuit ruling never came down due to a prior settlement, no infringer of API copyrights would be able to get any mileage whatsoever out of the Northern District of California ruling. The December 4, 2013 appellate hearing was recorded and transcribed. Any right holder who has to defend the copyrightability of API declaring code (provided that it's truly creative and original, as is the Java code) will cite to what the circuit judges said at the hearing. The only way this could change (apart from an unforeseeable about-face by the court) would be a Supreme Court ruling, and while I fully expect Google to ask the Supreme Court for writ of certiorari, I don't think a finding in favor of copyrightability -- unlike one to the contrary -- is "certworthy". Affirmance of the Alsup decision would have created a conflict with Supreme Court precedent (for example, the rule that even a modicum of creativity is sufficient to justify copyrightability). But why should the SCOTUS have any problem with what the Federal Circuit is apparently going to do? Google can't point to any SCOTUS case in which a similar quality and quantity of original expression was deemed uncopyrightable. Google's best case will be something like the affirmance by an equally divided SCOTUS in Lotus v. Borland -- and the Oracle case is easily distinguishable from that one (even Google's counsel conceded at the Federal Circuit hearing that what was at issue in Lotus was "less developed than Java"). Google will presumably claim that Sega and Sony established an interoperability exception to the scope of copyrightable works, but I really doubt that the SCOTUS would view those cases differently from the Federal Circuit: those are just "fair use" cases that allowed only very limited copying for the purpose of writing compatible programs.

Now, in light of this situation, we are in for a total doomsday scenario if we want to believe what the EFF, CCIA and other Google-funded, Google-aligned organizations warned against in their public campaign on Google's behalf as well as the amicus curiae briefs they submitted and orchestrated. The fact of the matter is that, unless you're Google or close to Google, there's no reason at all to be concerned. On the contrary, it would really have been a threat to software development if the appellate hearing had worked out more favorably for Google because this would have made it very easy for large companies to hijack APIs developed by small, innovative players.

Google is one of a very, very, very few entities to hijack third-party APIs this way -- by hijacking I mean unauthorized use of API code (with or without modifications) in a new (typically competing) platform. I take issue with its unauthorized distribution of more than 7,000 lines of (on balance) exceptionally creative declaring code, and I disagree with its assertion that the Linux API declaring code it also integrated into Android was devoid of copyright protection in the first place ("This header was automatically generated from a Linux kernel header of the same name [...] It contains only constants, structures, and macros generated from the original header, and thus, contains no copyrightable information."). Linus Torvalds may be on Google's side, but he didn't author Linux single-handedly and does not have legislative authority over the question of copyrightability.

The way all software developers -- including yours truly (as I mentioned in a couple of posts last year, I'm working on a smartphone and tablet computer app, which is why I've recently limited my consulting on IP issues to two days per week) -- use APIs sooner or later is that we write software for a platform and need APIs to make our software run on (or with, if it's about network-based communication) a given platform. This type of API use is simply not at issue in Oracle v. Google. If, hypothetically speaking, Google had developed a Google Maps client for some Java-based Nokia or BlackBerry devices (which used to be popular at some point, but nowadays Java has been almost completely displaced by Android on mobile devices), Oracle wouldn't have sued over infringement of declaring API code. Had Oracle done so (in an alternative universe), then the likes of EFF and CCIA would have been right and there would have been a serious issue -- which would not have been a copyrightability issue but more of an estoppel and fair use issue or maybe a matter of interpretation of license agreements.

It's just not a possibility that someone could publish an API, encourage third parties to write code for it (because that's what makes an API successful), and later turn around and sue them for doing so. It's commercially illogical, but above all, it would be a legal non-starter. Any software developer could choose among various defenses. There could be a contract-based defense. There would always be equitable estoppel, promissory estoppel, and fair use. Legitimate API use is fair use. Hijacking APIs the Android way is unfair, but using them the way they are intended to be used is above board.

Given that someone trying to go after legitimate API developers this way has no prayer, there wouldn't even be a point for a "troll" in acquiring API copyrights for a litigation campaign like this. It would be a money-losing proposition. No one in his right mind would do this (or fund this).

Some zealots and lobbyists supporting Google have over the last couple of years propagated the allegation that Oracle's position on API copyrightable is somehow extreme and extends the scope of copyrightable subject matter into, if not beyond, the realm of patents. In reality, it's Google's position that is extreme, and Oracle's position is balanced and moderate. The two extremes would be a scope of copyrightability without meaningful boundaries and requirements (the perfect scenario for abusers) and another one in which even non-API program code could be denied copyrightability, despite being creative (a fact Google never contested in the Android-Java case), because of some functional purpose it serves, in which event software developers would be able to protect their works only through patents and trade secrets. Google's position is, unfortunately, very close to the latter -- in its cross-appeal it also disputes the copyrightability of a certain (not huge, but still significant) amount of non-API code. Oracle, however, would never claim copyrightability for a single line of very simple API declaring code such as the header of a maximum value function (two parameters, result is the greater of the two numbers) in its own right. Oracle's case here is based on an enormously creative, large-scale body of code and its structure, sequence and organization.

As long as there is a sufficient creativity threshold, honest software developers are protected, not threatened, by copyright law. They are not threatened because if such code is reasonably creative, none of us will write the same code independently by happenstance: infringement will, in practical terms, require willfulness. They are, however, protected with respect to the code they author independently and can prevent others from infringing on it.

A couple of paragraphs ago I mentioned that Google and its allies misportrayed Oracle's position as an extreme one. This is not just my opinion. Circuit Judge Kathleen O'Malley also set the record straight on this question when Google's counsel made an attempt to mischaracterize Oracle's copyrightability position at the December 4 hearing. The relevant part of the official hearing recording starts at 46:50, and this is what it's about:

Google's counsel Robert van Nest referred to § 102 a and b of U.S. copyright law. In a nutshell, $102 a giveth (defines what is copyrightable) and §102 b taketh away (defines what copyright does not extend to), but not in the sense of a straightforward subtraction: it's more like §102 a defining what types of works are copyrightable and §102 b ensuring that enforcement doesn't reach further beyond. For example, you get a copyright on a cookbook under §102 a, but §102 b ensures that you can't use that copyright and sue everyone who cooks a meal according to your book. For software, it means creative code (including declaring code) is copyrightable, but you can't assert broad patent-like monopoly rights later over methods of operation.

Google's counsel misrepresented Oracle's position on copyrightability by saying that "the basic structure of the copyright statute is you have (a) and (b), and they're proceeding as though (b) doesn't even exist -- they're saying 'if it's creative under (a), boom, you're home'". Circuit Judge O'Malley contradicted immediately and firmly: "I don't think that's what they're saying at all. No. They're just saying that a method of operation [which copyright does not extend to] should be looked at at a more abstract level." She got this so right: contrary to Google's counsel's representations, Oracle does not want to use copyright to monopolize abstract concepts such as the idea of having a function that determines the greater of two numbers -- Oracle just claims copyright in a body of work that involves in this case many thousands of lines of highly creative declaring code, which is concrete and not abstract. It's about specific expression, not about high-level concepts and algorithms.

When I listened to the hearing recording for the first time, the passage I just quoted was one of the most important ones to me. It indicated to me that the copyrightability part of the case was strategically lost for Google, and that all of the FUD submitted by Google and its amici curiae failed to manipulate this sophisticated appeals court.

Google will have to respect Oracle's and other companies' and people's copyright in API declaring code, and it will have to take licenses and/or make the necessary modifications to Android to address those issues. But in the long run, even Google may very well find itself in situations in which it benefits from copyright protection of APIs. I already wrote about a year ago that Google might one day use API copyrights to prevent Samsung from making Android apps run on Tizen.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: