Why? It wasn’t about whether APIs are under copyright (they are), but whether this particular case constitutes fair use. I think the effect of this decision was way overblown.
> whether this particular case constitutes fair use
The English common law system relies heavily upon judicial precedent. A ruling in this particular case that the copying of an API constitutes fair use will inform decisions in future cases of similar copying.
Justice Thomas clarifies why this case has wide-reaching impact in his dissent.
"Congress rejected categorical distinctions between declaring and implementing code. But the majority creates just such a distinction. The result of this distorting analysis is an opinion that makes it difficult to imagine any circumstance in which declaring code will remain protected by copyright."
I concur with Thomas (not in the connotation - he thinks this is a bad thing, I do not - but in the denotation). This ruling is such strong precedent that almost any API cannot be bound by copyright that no company is going to be willing to spend the money to raise the question.
I think Thomas makes a mistake in disregarding the reason for introducing the distinction between declaring and implementing code. The law (as I understand it - IANAL) doesn't exhaustively enumerate what does and does not constitute fair use. Therefore, any specific instance of fair use is bound to introduce new distinctions that are not specifically mentioned in the law itself.
In this particular case, Google's copying of interfaces was ruled to be fair use because its purpose was to allow Java developers to reuse their skills. This reasoning only holds for the declaring parts of the copyrighted work though. And that's why the distinction becomes necessary even if Congress rejects it as a primary distinction between copyrightable and non-copyrightable work.
That's also why I think the ruling is perhaps not quite as broadly applicable as some (including Thomas) are thinking right now. It may not cover code that isn't used by a significant number of developers. I haven't read the entire ruling, so I'm not sure whether it covers non-public interfaces that are nevertheless useful for interoperability. Such interfaces were at the center of past disputes (I think Microsoft's SMB protocol was one such case).
Wine would be a much better example of fair use. It is explicitly made to allow users of linux to run programs written against win APIs. It is not so clear in the Oracle vs Google case, because it is questionable whether Google’s case was really fair use (google pretty much didn’t want to pay Sun for the mobile license and the whole thing was done as cost saving) - hence the very long timeline of the case.
The dissent explains how Android fails three of the four fair use tests. I'm not sure what the majority was smoking when they wrote this, but far better examples of fair use exist, especially uses that are noncommercial and especially are noncompetitive with the original.
Java was a mobile OS before Android literally obliterated the market after copying Java. Arguably, if Android was fair use, everything is fair use and all software copyright is fundamentally defeated by this ruling.
