Quote:
Originally Posted by thirdm
If you're interested in that case, Bradley Kuhn and Karen Sandler discussed it on their recent podcast episode here: http://faif.us/cast/2014/may/13/0x44/
This being a complicated case (and if you'll excuse me saying so your post showing great confusion or indication of being influenced by severe misinformation, e.g. the copyrightable arguments by Alsop were largely rejected but fair use considerations are remanded back to the lower court, so "loss" hasn't happened yet), if you have the time I'd recommend also listening to their earlier podcast 0x35 and especially reading over Judge Alsop's decision followed by the recent circuit court decision. The pdfs of those two decisions are linked from episode 0x44. Bradley Kuhn also has a blog entry on the recent decision and there is discussion with him and others linked from there on identa.ca: http://www.ebb.org/bkuhn/blog/2014/0...le-google.html
From what I hear most technical press coverage of the recent decision has been complete crap. I'd not read any more of that and try to forget what you've already read. |
Thanks for the links. I'll check them out.
Edit: Now I'm back from the reading of the text associated with your links.
In it I see that the "loss" refers only to Google's loss of the original, favorable district court decision, because of Oracle's appeal. That's what I meant in my post (bad wording), as I did absorb that info from the news about the case. Now the case gets kicked back to the courtroom for a determination of whether or not, in the specific referenced case, the alleged infringing copy could be claimed a "fair use" copying of the code, and by direct association - (only in some people's opinions) - a green light for the use of API declarations as "fair use".
That favorable result has yet to be had. It would not be a green light, and barely a yellow one. If the new courtroom result is that the allegedly infringing usage was not "fair use" - then there is a precident for APIs being generally copyrightable, with fair use disallowed in particular cases. Subsequently, all allegedly infringing code would rely on courtroom actions, decided on a case by case basis. That is the problem with the utilization of the "fair use" defense. "Fair use" is vague to the point where it almost always has to be hashed out in a courtroom. Thus, it does not protect the general immunity (relative to copyright) of APIs. It does make more money for the legal profession, because nearly every case has to go to trial. Talk about FUD!
If one considers that "fair use" is often limited to a small number of lines, depending upon the form of expression, that does not bode well for Google. There are some 7,000 lines of declarative code (the API) and fair use has in some cases been limited to a hundred. It depends upon the percentage - so with such a large API the courtroom result could allow for a larger number of lines of code protected under fair use. Sometimes, the result in such cases has been a substantial percentage - say ten percent or so. Unfortunately, due to all the vagueness associated with the doctrine, that figure goes all over the house. Relying on fair use to protect our ability to re-implement APIs will be nothing short of a nightmare. The only way it isn't a nightmare is if "fair use" for purposes of software interoperability is precisely defined (and that is light years away from current applications of the "fair use" doctrine in courtrooms).
The ideas expressed on the sites you linked, about the court and specifically the jury not understanding the idea of "APIs" well enough to separate them from straight forward copying of code in an ordinary sense, does not help necessarily. Even if that is an error, will any subsequent actions see it as one? A precedent can be sited, error or not, and subsequent juries (and courts) could compound the issue with the same errors and lack of understanding.
The whole thing's a complete nightmare.
Again - I'm not a lawyer, so disregard everything I write.