[Discuss] can you copyright an API?

[blu at nedharvey.com (Edward Ned Harvey)]

> From: discuss-bounces+blu=nedharvey.com at blu.org [mailto:discuss-
> bounces+blu=nedharvey.com at blu.org] On Behalf Of Tom Metro
> 
> Mostly is seems this is coming down to the issue of whether you can
> patent an API - the names of classes, methods, and their arguments.

Not patent an API.  Copyright the API.  Which Sun did, but Google doesn't
feel is legal to do.

I loves me some android, so I don't want to see any harm come to it, but in
this case, I think google f**ked up.  There were lots of ways they could
have avoided all this mess... For one, they could have started with the GPL
openjdk.  Even if they threw out and rewrote 99% of the code in there, as
long as it started with code that Sun released under GPL, and they continue
to develop all their modifications under GPL, then oracle wouldn't have a
case against google...  But google didn't decide to do that.  

For two, why java?  If you're going to the effort of writing the whole
language from scratch, why not rename the classes and stuff, so you're
basically copying an existing language, "inspired by" some language, but not
really copying it?  Actually, I rather hope that google is thinking about
this, and I rather hope they're able to use this reverse-logic as the basis
of part of their argument.  By merely substituting different names for
classes & methods, they could avoid the problem...  Which means...  Yes you
can copy an API as long as you transliterate words...  Which is ridiculous.

Well, unfortunately, this is copyright law, not patent law.  Which means
using the same name is the problem.  By simply substituting a different
word, you're not infringing on copyright...  

It would be a very different case, if the API were patented.

And, exactly as you say, I have some concerns about what implications this
may have for the likes of mono.  As I understand, originally .Net is/was a
commercial product, that MS released open source (or the compiler) but the
open source version isn't/wasn't as good as the commercial version, in terms
of performance / optimization...  Basically MS was trying to make open
source look bad by comparison.  And then mono reverse engineered it, threw
it all out except the API, reimplemented it all, to be competitive with the
commercial offering.  But I don't know if they started from their own
codebase (under the same or different license terms from what MS
released)...  Or if they started from the open source MS released and then
threw away a lot of it...  I don't know...