Oracle v. Google Has Reached The Worst Result Possible

For those who don't know, Oracle v. Google was centered around Google's Dalvik- which is a version of the Java Virtual Machine (JVM) for Android.

Now, when I say a "version" of the JVM, Google re-implemented everything from scratch. The only thing that was unchanged- and it was verbatim- was the Application Programming Interface (API). The case and subsequent appeal seemed to focus on whether an API was copyrightable in the first place.

What's an API?

Without getting too confused about what an "API" is, let's focus on "Interface". It's an interface- that is, it's basically a set of controls that you can work with to perform some software function. "Application Programming" meaning it's an interface that's exclusively used by software- not human beings. So an API is really just a definition of how two pieces of software interact with each other.

For example: Say I want to open a new window in a Java application. I would interact with a separate package that's responsible for the user interface, using its Application Programming Interface (API) to direct it to create windows. That, in turn, will work with, for example, the Windows or Mac or other API that allows you to create new windows on a running session.

When you write a piece of software that you want to expose some functionality to a group of tools, you write an API for it. Much in the same way that you would design a screw head to interact with specific tools.

Useful Articles

The problem is that, for about 150 years, it's been settled law that you can't copyright useful articles. This has been interpreted over the years such that "useful articles" may refer to "features" of a work. This is partially based on jurisprudence in Feist Publications, Inc., v. Rural Telephone Service Co., 499 U.S. 340 (1991) and other cases.

The basic idea is that whether something is "copyrightable" means examining the elements in a work to see if they qualify as minimally "original", and if so, whether they're disqualified for other reasons. One disqualification is that a "useful" article whose design serves a purely utilitarian purpose isn't copyrightable. Instead, they say, seek a patent.

The description of the art in a book, though entitled to the benefit of copyright, lays no foundation for an exclusive claim to the art itself. The object of the one is explanation; the object of the other is use. The former may be secured by copyright. The latter can only be secured, if it can be secured at all, by letters patent.

Baker v. Selden, 101 U.S. 99 (1879)

Which, for any practicing patent attorney, makes intuitive sense. If you develop some new design for a screw head that's been constructed so only your brand of screwdriver will fit into the notch, that is the purview of patent law. This isn't just musing, this is actually done.

Similarly, if you wanted to prevent others from using your API, then I think you should have to file a patent on it. But then, unlike Copyright law, you actually have to prove that it's novel, useful, and non-obvious (and eligible subject matter, since it's a software patent!).

You don't have that API protected for the life of the author plus 70 years either- only 20 years give or take term adjustments. I don't think just because you're a software company, that you're entitled to be given what amounts to unexamined, nearly indefinite "super patents".

Okay, But Google Won

Google won this time because they managed to convince a jury that even though the API was copyrightable, their use of it was fair use.

So yes, they won, but they won for the wrong reasons. Here's the thing- the fair use defense will probably stick on appeal, because it was decided by a jury and isn't a question of law at this point- questions of fact are granted strong deference on appeal. But recognize how they got to the point of finding fair use: at the end of a full trial to verdict. For a case the size of Oracle v. Google, "millions" in legal fees may cover the process up to the beginning of trial.

This doesn't stop the fact that the federal circuit just created an entirely new class of IP troll, the API troll, who can now file copyright infringement cases based on their "ownership" of an unexamined copyright on their API and force competing companies and startups to either pay up or go to trial.

So the message here is: "APIs are copyrightable, and you can sue people for implementing your API, unless they're a large corporation with lots of clever copyright lawyers on retainer."

How to Cite Republished Information

Do not use this information as legal advice.
If you need advice, call or email a licensed attorney.
No attorney-client relationship is created by your use of this site, including any communication with the author in comments.