End of a fishing expedition

The result of the Oracle-Google case blocks an inappropriate extension of copyright.

Oracle has come up empty-handed in its fishing expedition for a big damages settlement against Google. The jury has ruled against patent infringement, a charge that was far-fetched, to say the least. There may be some limited damages for copyright violation of a short bounds-checking method, but the judge recognized that any competent programmer could have written this method in a few minutes. As Jury Foreman Greg Thompson said, “The more tech-savvy a person is, the more difficult it is to persuade them about what limitations should be placed on technology.” Indeed. As I’ve frequently said, I invented linked lists and hash tables when I was in high school, but fortunately for the industry, software patents didn’t exist back then. I don’t claim to be unique; I’m sure many clever 17-year-old programmers have invented hash tables. Just about everything in this industry has been invented many times, by almost everyone who is vaguely clueful. Once you understand the way computing works, you fairly quickly understand why software patents should be extremely rare, if they exist at all. There’s prior art everywhere, and almost every invention is “obvious” when looked at from the right perspective.

The final significant issue, whether an API can be copyrighted, was resolved Friday afternoon. That’s an extremely important issue, possibly the most important issue of the trial. In addition to being a monstrous extension of copyright law into completely new territory, copyrightable APIs would kill much open source software, and a lot more besides. Linux couldn’t exist if APIs were copyrightable, since Linux implements many of the APIs from BSD and System V Unix. The nasty copyright infringement lawsuits that SCO brought against various Linux vendors aren’t that far in the past (only ending in 2007). Imagine how much of a mess that would have been if copyright applied to APIs. Imagine if the many Unix vendors of the early ’90s had the legal tools to shut down the upstart Linux. And it’s not just open source that would be hurting. The early Windows TCP/IP implementations copied the BSD UNIX networking APIs. (I haven’t looked, but I’d expect that the APIs are still substantially identical, even in Windows 7.) Who knows what might have happened if the University of California had been hungry for some quick litigation-based income? While the BSD source code is covered by a license that allows source redistribution, what about the APIs themselves, as distinct from the source code? In the absence of copyright violation itself, Oracle’s claim was essentially that you can violate an API copyright even if you haven’t violated the license for the source code. That is interesting, and horrifying, territory for legal exploration. The stakes were extremely high.

Judge William Alsup’s ruling states an API is “a utilitarian and functional set of symbols.” It’s analogous to a table of contents, which is primarily a utilitarian set of symbols linking to a book’s internal structure. As anyone in publishing knows, only a “work” can be copyrighted: not a title, not an outline, not an organization. (Copyright is not plagiarism; plagiarism is an ethical offense, not a legal one, and has to do with giving credit for ideas, not with copying itself.) Thus, when software is copyrighted, the only work protected by copyright is the code itself, not the organization of the code, not the list of titles (method names) in the code, but the code itself. Specifically:

“Contrary to Oracle, copyright law does not confer ownership over any and all ways to implement a function or specification, no matter how creative the copyrighted implementation or specification may be. The Act confers ownership only over the specific way in which the author wrote out his version. Others are free to write their own implementation to accomplish the identical function, for, importantly, ideas, concepts and functions cannot be monopolized by copyright.”

An API copyright is tantamount to copyrighting all possible implementations of that API, not just the implementation created by the copyright owner. That’s clearly unacceptable. Copyright is not about abstractions, it’s about concrete texts and implementations.

In an era where abusive copyright and patent trolling is victimizing many companies, from startups to giants, it’s important to see an important case establishing a limit to trolling. And it’s important that this limit protects the many companies and projects that could (and would) be victimized by an inappropriate extension of copyright law far beyond its original intent.

And more than that, it’s important to see that we will get proper rulings on copyrights and patents when judges and jurors understand the technology that they are ruling on. Judges and jurors are not computing professionals, nor should they be. However, it’s as nonsensical for a judge who doesn’t understand programming to preside over a software intellectual property case as it would be for an illiterate judge to preside over a literary copyright case. Judge Alsup’s decision starts with an elementary Java tutorial that discusses the key language concepts that were important to his decision. It’s gratifying to see that he did his homework, and really understood not just the language, but what an API means to a software developer. He understood classes, packages, public and private methods, and how they work together to form a standard specification. He understood that implementation of a specification is necessary for interoperability, and that interoperability is essential to modern programming.

The winner in the case isn’t just Google; it’s all software developers, who don’t have to worry as much about creative interpretations of copyright law, and are free to develop compatible implementations of an API. It’s the entire intellectual property system, as it has proven that it can work, given intelligent and well-informed judges. We can only hope that this decision raises the bar for future trials. If this ruling discourages patent and copyright trolls, we will all be winners; trolling imposes a huge cost on innovation, both in legal fees and in dampened creativity. As I said two weeks ago, Judge Alsup is a modern hero.

Related:

tags: , , ,
  • D. Joe

    When you write “Linux couldn’t exist if software was copyrightable” I think you mean “Linux couldn’t exist if APIs were copyrightable” right?

    Software packages as a whole can, of course, be covered by copyright.

    • http://radar.oreilly.com/mikel Mike Loukides

      Thanks for the correction!

  • AP

    ‘almost every invention is “obvious” when looked at from the right perspective’

    And that sentence, alone, explains why you blather about inventions as opposed to doing. The doers out there have more complex feelings about patents. The talkers instead find everything trivial.