(To rehash what I said on Twitter...)
I don't think this is as bad as the headlines make it sound.
Here's a direct link to the brief in case you can't find it in the article (I couldn't):
http://computemagazine.com/wp-content/uploads/2015/05/Google...
While the DoJ does fully agree with Oracle on the question of copyrightability, the brief goes into some length about its belief that interoperability issues would be appropriate and relevant to a fair use argument - fair use being the question that was never decided in the original trial due to a hung jury, and AFAIK would be subject to appeals regarding the legal standard anyway. For example, it states: "Interoperability and lock-in concerns like those raised by petitioner can appropriately be considered as part of fair-use analysis."
Now, if you're FOSS Patents, you take the one sentence where the DoJ uncritically repeats a claim about interoperability that's been made several times in the record and never really contested, as part of its restatement of the facts, and you extrapolate it into something "damning" for Google's entire fair use argument:
http://www.fosspatents.com/2015/05/us-dept-of-justice-finds-...
The claim is: "The Android platform uses the Java programming language, but petitioner purposely designed Android not to be compatible with the Java platform or interoperable with Java programs." I'll get to that in a bit.
Mr. Mueller argues that the discussion on the relevance of fair use is just window dressing and that "the U.S. government has also concluded that Google can't make a 'fair use' argument based on compatibility or interoperability because of the specifics of this case". Based, as far as I can tell, only on that sentence (and a repetition later), never mind that the brief also explicitly mentions that Google's argument about programmer fluency (i.e. even if Android isn't compatible with Java software, it avoids lock-in by letting programmers use their existing skills) is relevant to fair use... or the fact that the issue which is supposedly pivotal is only mentioned in passing... or that the appeals judgement which Google is trying to get overruled already states, and the brief quotes, that there are not yet "sufficient factual findings" regarding the questions involved in fair use - for the DoJ to actually be rejecting Google's fair use argument, it would have to be supposing that there are already sufficient findings, without actually saying so.
So much for that.
But let's get back to that statement, about Android not being compatible with Java programs. If you're like me, you found it absurd on its face when it first showed up in the appeals judgement, because while Android is not compatible with entire JME applications, it is compatible with many Java libraries and with non-UI code, which in reality is extremely important if you're trying to port your Java codebase to Android. I think (correct me if I'm wrong) that Google can't really start an argument over it, because Oracle made these contentions at trial and Google did not contest them then. I fail to understand how Google's lawyers allowed such a misleading claim to reach a jury when it did - maybe, as non-programmers, they didn't fully understand the issue either. But in any case, if the Supreme Court rejects Google's case and there is a new trial on fair use, Google will have a chance to start fresh with the whole line of reasoning.
It will arguably be better for all of us if the Supreme Court accepts the case anyway and decides Oracle has indeed failed at the first hurdle - that Google has not copied something copyrightable in the first place - because the more uncertainty there is around clean room reimplementations in general, the more people will be discouraged from working on such projects, decreasing interoperability. Also, fair use is partially a question of fact (as opposed to law) and thus for a jury, and while I wouldn't trust judges these days to truly understand the facts and actual practices around software issues, as opposed to making analogies that act like a funhouse mirror... I trust randomly selected juries even less. But in reality, the applicability of copyrightability vs. fair use is a relatively arcane question that will be decided in part by however the statute happens to be worded, and Oracle getting to the next hurdle instead is not the end of the world.
Also, even if the lack of compatibility with entire applications is (implicitly or explicitly) found critical in such an argument, projects like Wine which are more principled about compatibility and do run entire existing applications will have some wiggle room. So don't despair...