As most of you will be aware the Oracle vs Google court case has final begun (they failed to negotiate a deal outside court).

Oracle are making their arguments first and have posted their opening slides here (makes for an interesting read).

Should be interesting to see who comes out on top.

I know it makes sense, but it is sad that nobody cares that all developers prefer having android being java-based .

Business ruins developments :/.

What I'm wondering is how Oracle got these apparently secret e-mails

Groklaw has really good coverage of the case. The highlight from yesterday's hearing being Oracles Larry Ellison taking the stand and being questioned by Google's lawyers.

I just come from spacee, what is the issue?

And rightly so in this case. When you sign a contract it will state that all the stuff you're producing for your employer is your own work (which you're signing over all rights to typically) and is free from any legal/licensing issues unless you've informed management.  I surely wouldn't want the shadow of a lawsuit against me from both Google and Oracle.  Back when I was doing custom video/audio & speech codecs I had clients ask me (on the phone of course)..."Hey!  Can't we just pretend we don't know about patents X,Y,Z?"  Man, screw that.

Basically that Google more than willing went stomping all over patented and copyrighted materials.

James Gosling lays out his position here.

The sports term for this would be "rooting for injuries."

I'd still like to see Oracle get the worst of it, otherwise you can look forward to paying for Java on a per-CPU basis.

Obviously Google are in the wrong, but in this case I simply don't care.

This might strike at the very heart of copyright law itself, but I don't feel the law is serving it's intended purpose.


Sun's repeated failures at creating a viable consumer platform for Java development are evidence enough that penalizing Google's success does not serve scientific progress.

It's the dichotomy of where to draw the line between the word of the law Vs the spirit of the law.

The slides seemed to only focus on the copyright infringement of the API design and implementation.

On a side-note this slide puzzles me:

(Oracle's Slide 76)

How does requiring a "clean room" implementation reduce fragmentation?

100 monkeys writing 100 independent implementations of any given method will result in 100(+) different bugs.
J2ME is a perfect example of this; it's API might have been consistent across the many hundreds of different handsets, but each implementation was so riddled with unique bugs that the platform itself became nonviable due to fragmentation.

Perhaps this is a failing of the compatibility tests (or Sun's historically lackluster enforcement of them), rather than a problem with clean room implementations per se.
Though if that's the case, I think it's naive to expect even the most thorough of compatibility tests to catch all problems.

Indeed, but as @Abuse mentions, the fact that a de facto standard implementation exists that everyone actually codes to in practice means that a clean room implementation, well, simply introduced more fragmentation and bugs than reduced them. It was with hindsight and the best of intentions, a bad idea. The long and the short of it is, Sun should have fully opened the java.* packages and plonked an appropriate license on them such that anyone could use them for anything, and left the real licensing behind the sun.* packages and the actual Hotspot native binaries.

Cas

The verdict of the first half of the Oracle vs Google trial is out (second half will cover patent infringement). The jury has sided with Oracle by finding that Google partially infringed Oracles copyright with Android, there is still a chance that Google might be able to argue a 'fair use' defence but not looking good for them, full info here.

Update: another take on the outcome here, from which it doesn't seem so bad for Google.

The only solid "infringement" is nine lines of code, from which Oracle is unlikely to show actual damages.  Otherwise the jury found that Android copied the API, but that the judge is inclined to rule that the API is not copyrightable.

Even if some novel decision came out that an API is copyrightable, Google could certainly make a claim that Sun waived any claims long ago by having a program for certifying java implementations, then again by GPL'ing the codebase.  They're certainly going to make that claim for the patent parts.

You'll probably find lots better analysis of the case on Groklaw, assuming you can stand the screechy sanctimonious vindictive paranoid bitch who runs the place, or her legion of sychophantic fans.

I'm glad that I'm doing most of my stuff with ECMAScript these days. What Oracle is doing there is really horrible for Java. Android was the best thing which ever happened to Java.

>Obviously Google are in the wrong

How so? Sun was fine with it.

Besides, Oracle stole SQL's API.

Source? Everything I've seen looks like a clean-room re-implementation. The 'copied' code is so trivial it's not surprising it's the same given the api contract it has to adhere to.