Oracle case crippled after judge rules APIs can’t be copyrighted Google has won a major victory in its legal fight with Oracle over the use of Java in Android after the presiding judge ruled that the APIs under dispute can't be copyrighted. "So long as the specific code used to implement a method is different, anyone is free under the Copyright Act to write his or her own code to carry out …
No, the BIG started long ago - the so called LEGO precedent where LEGO tried to sue other brickmongers. This is just an application of the same rules to the computer domain. Expecting anything different was frankly beyond optimistic.
It has some interesting side effects.
The long standing practice by the Open Source community to "protect" against interfacing to GPL2 components through the gratuitous application of GPL headers to include files which define API has just been ruled to have no protection. GPL, MPL, etc work will still be protected against theft, however the so called element of "virality" has been removed for a lot of the possible use cases.
I'd like to join you in your optimism, unfortunately I know our legal system and the players too well.
I expect Larry "I have more money than God" Ellison will charge his lawyers with finding a way to appeal the decision and have the judge's ruling overturned. I don't expect it to end until SCOTUS has ruled. I'd quote a famous Brit about what this is, but I won't offend you by getting it wrong.
This post has been deleted by its author
I think it's a tad disingenuous to try to liken this case to the SCO debacle. Oracle isn't going to destroy itself by persuing a fruitless lawsuit to the detriment of their entire business like SCO did. They'll just appeal and drag the case out until they can't any more and *when* they lose they'll be really butthurt about it and the entire tech community will take great satisfaction in watching them moan and bitch.
We all know that Oracle are conscienceless dicks. It's kinda heartwarming to see them getting fucked for once.
Like software patents, like the stupid "slide to unlock" anybody could have thought of that idea, and anyone (within the field) could have reproduced it via any API using any number of different methods.
I'm actually hoping somebody does try to use that as a defense at some point, if ti works it could destroy software patents... probably not.
--rant incoming--
All I'm trying to get at on the software patent is that well, If I were to create a piece of software using the above example of slide to unlock and patent it, nobody would be able to slide to unlock using any form of software. That is how software parents are defined these days.
If I were to go back in time and create a car engine and patent it, anyone else could still go out there and create their own version of the car engine, so long as the underlying construction is sufficiently different from my patent.
This is why I dislike the software patent rules, and I'm glad that this ruling has been made. It actually gives a foot in from the bottom level if an API cannot be patented for the aforementioned reasons, IE method names the same (outer appearance) but the underliying construction is different, then why can features within software (slide to unlock best example I can think of) be patented based on their external functionality when the underlying construction is vastly different.
I can agree 100% on software patents which pioneer new software technology like a new encryption algorythm, a new way of identifying images of people etc, but I cannot agree on the stupid patents like using a touch screen interface to predictively dial a number. Its making use of common technology which has existed in different fields since computers with GUIs arrived. But just because they add the fact its on a mobile technology its suddenly a whole new patent.
Gah, sorry I just hate software patents, I REALLY hope that eventually somebody with half a brain makes it so they're more clearly defined and protect innovation (new alogrythms and programming paradigms etc ) rather than stifling others.
Patents were meant to protect you, not to punish your competition.
I'd also love it if companies who take somebody to court for patent infringement when no such infringement occured, they had to pay the legal costs for both sides. Might make them think twice before trolling since most of the cases seem to lose.
I made the mistake of trying to read his response. He implies Google have pulled an OJ Simpson:
Google's decision to defend its position at a trial has paid off: it gets away unscathed, at least for another couple of years, with what it's done.
Damn you Google! Damn you for being innocent of the bogus allegations!
He goes on to criticise how the judge ran his court room and then quacks about how he wasn't wrong anyway. I gave up at this point, I have better things to do. Like driving nails into my head.
So I wonder if Andrew Orlowski will run a write-up on this result and maybe eat some humble pie? After all the register did post the ORACLE V GOOGLE ROUND-UP: THE SHOW SO FAR but then didn't post any more of the "show" after just pointing out the Oracle case - despite the Reg's DrewC saying "Note: "show so far" Google's turn will come next week." and Diodesign stating "Re: Re: article fail
But my point is that there should be no suggestion that we're omitting or censoring Google's defence.
C." apart from the fact that Goggle's opening statement or it's patent win was never reported on.
Of course it couldn't be the Florian effect that creeps into The Register's Articles from some of the lesser author's? As DrewC stated "Florian Mueller has become a bit of a player. You know what his POV is and you know who is paying him - because he tells you what is point of view is and he tells you who is paying him. This does not make him wrong or corrupt."
Well Saying that this case is worth Billions, That Oracle would win, that there was hang out's for the Jury in the minority on Google's side, that the API copyrights would be deemed copyrightable and telling his NYSE clients that Oracle was a good bet (Last one speculation" does make him wrong, the same as his SCO predictions and most other's. You don't know who is paying FM, the Oracle deal was month's old and he provides no disclosure on his post or his Bio.
Just to clarify, I meant the "show" as in the show shown by the Lawyers - Andrew Orlowski did pipe up with the little gem of Java jury finds Google guilty of infringement: where he stated "The jury found Google to be unequivocally guilty of copyright infringement on the major charge" without realising that the judge had told them to assume it was copyrighted and as the fair use was hung in Googles favour they were Not Guilty at all (you aren't guilty if there is fair use).
Also stating "For its part, Oracle has presented damning and unequivocal evidence that Google knew it needed a licence if it was to build Android on Java"
Damning? Not; Unequivocal? Really, why do they need a licence again when they haven't done anything wrong?; knew? They never did and they are right
Then mentioned " embodied in the catastrophic appearance of former Sun CEO Jonathan Schwartz at the trial" - well he was one person who spoke the truth in raw terms, some good for Google, some for Oracle. Why was his appearence "catastrophic", ah because it ruined the Oracle argument - only works if the piece was written in a pro Oracle slant though.
But of course the heading "Google Guilty of Infringement" Not "Oracle loses all but a tiny fraction of their Copyright claim"
As the trial went on and Oracle looked to lose everything the reporting stopped....
A Judge who understands the difference between the *name* of the thing, *what* the thing does and *how* that thing is made/implemented?
Priceless.
Copyright should only ever apply to the "how". Achieving the same result via a different method (the "what") is not in the purview of copyright, neither is the name given to the thing... although that last one may be Trademarked, which is a different kettle of fish.
> A Judge who understands the difference between the *name* of the thing, > *what* the thing does and *how* that thing is made/implemented?
During the trial it transpired that Judge Alsup was himself a bit of a hacker. Apparently he used to write software, in FORTRAN. There's a reference to this in one of the articles at Groklaw.
The thing is, that since this is only the _first_ judgement, we all know that it will be overturned on appeal.
Which it certainly should be - this was shameless copying by Google - because they needed something that was identical, the argument that it is only a name is not valid.
" the argument that it is only a name is not valid."
Let's imagine I come round your house for dinner and you serve, say, Goulash.
I enjoy the dinner and when I get home I try to remember the flavours, knock a few ingredients together and then gradually fine tune ingredients and method until I have a dish that tastes identical to yours.
Have I stolen your recipe?
If you take home a sample, analyse it scientifically, then write a recipe from scratch that "just happens" to be identical to mine, then yes you have copied my recipe.
If you do it for commercial reasons, because your studies show that "Goulash" sells ten times better than "Hungarian-style meat stew", then your hands are decidedly unclean.
If you take home a sample, analyse it scientifically, then write a recipe from scratch that "just happens" to be identical to mine, then yes you have copied my recipe.
Remarkably wrong. I'm impressed; it takes a lot of effort to achieve that degree of error.
Under the (implausible[1]) scenario you describe, I'd've duplicated your recipe. I would not have copied it, if I didn't have the original to hand. That's what "copying" entails.
Furthermore, under the applicable law here there is no simple prohibition against "copying". There are copyrights and patents (and other forms of IP protection that don't apply here). Creating a duplicate from an analysis of function (which is the process you're describing here), without reference to the original, is very much not a violation of copyright. It's what patent, not copyright, protects; and this decision is about copyright.
If you do it for commercial reasons, because your studies show that "Goulash" sells ten times better than "Hungarian-style meat stew", then your hands are decidedly unclean.
Utter rot, and I'd like to see you defend that position with any reasonable ethical argument. What happens here is simple competition. Your protection here would fall under patent law, which does not prohibit duplication for profit.
[1] I'm well aware that food chemists have done just this sort of analysis, and indeed that there are commercial outfits that will do it on arbitrary samples for a reasonable fee. The implausible bit is the "identical" recipe. A recipe is more than a list of ingredients and their proportions.
Where is Oracel's major business - Databases right?
What language/standard do they use, SQL right?
Should Oracle give 50% of all profits ever made to IBM for shameless copying of the SQL spec? No of course not, no one should including Google, but if they wish to persue the matter that APIs and interfaces can be copyrighted then maybe they should just pay up to show how strongly they believe in their own case.
Oracle were lucky they didn't win, it could've ruined them. There's no guarantee this will be appealed, will be accepted for appeal and almost no chance if it is that it will succeed.
understanding the situation isn't bias.
Easier in the sense of not becoming illegal?
What is the point of disclosing APIs?
Ya wanna go back to Microsoft's secret APIs that killed off all their software rivals letting MS Word inherit the earth?
This is about more than Oracle v Google
This is not a paraphrase. It's a direct copy and since it is apparently made in the UK, in all probability it's a violation of Fox's copyright under the Copyright, Designs and Patents Act 1988. The only real question is whether it meets the s107 test for a criminal offence, as that provides for a sentence of three months in the Scrubs.
Extreme? Well Oracle are taking a similar line in this case...
It's a good thing that IANAL
IIRC the case includes not only API declarations (and these shouldn't be copyrightable, of course), but implementations too (some were copied verbatim, and this certainly amounts to copyright infringement (i.e. theft), by every account).
http://nighthacks.com/roller/jag/entry/comments_around_oracle_v_google
This post has been deleted by its author
...and if you'd actually kept up with the case you'd know Oracle has agreed that Alsup will decide the damages on the 9 lines of rangeCheck that actually was in Android and the 8 files that *weren't*. That part of the case is still being scheduled.
And he made it very clear to Oracle several weeks ago that the amount awarded would be approximately $ZERO. Oracle (or more accurately the talking monkeys they hired from BS&F) gambled insanely on turning de minimis infringement into a massive payout. They lost. The maximum likely award was $150,000 before handing Alsup the decision, like I said more than a year ago: Oracle won't cover their legal costs. The judge in fact pointed out to them in court that it would cost them more than they can win in lawyer's time just to argue the point.
Here I was enjoying a good rant against those thieving bastages at the ulta-huge megacorp, and you HAD to find one thing I'll have to give the credit for: they had the stones to take their case to court even though they were likely to spend more trying the case than settling it.
Frankly, I'd like to see more companies do that when they think they are in the right. In the long run we'd wind up with fewer nuisance suits. But then again I think you Brits have a losers pay clause that sort of limits that, whereas we 'Mericans don't.
Have you actually looked at the 9 lines that were "stolen"? All is does is check that a value is between two other values. To quote the judge:
'All it does is make sure the numbers you're inputting are within a range, and gives them some sort of exceptional treatment. That witness, when he said a high school student could do it-- '
This post has been deleted by its author
Wrong, you cannot ever copyright ideas, that's what patents are for. Sadly our idiot patent offices seem eager to patent maths (AKA algorithms).
This distinction is explicitly referred to in the judgement, both as the reason APIs aren't copyrightable and to point out this would award patent like monopolies that last far longer than patents - 95 years vs 20 in the US.
As I understand it:
The Patent would be for "an encryption system that took in messages of such and such a sort, processed them with these steps, while asking the operators to manage the key in this way"
People feel that this ends up patenting the algorithm because, really, that's all the encryption system is. The protection would be a 20-year monopoly.
The text of the program witten to build the encryption system could be protected under copyright, gaining a 70-year protection in the UK (is this true? that can't be true!) But anyone else understanding the algorithm could write their own program and provided they didn't copy the original, they would not be violating the copyright. Apart from the length of the protection, I think software copyright is fine. It enables the GPL among other good things.
I share the general concern about patenting algorithms because I think it chills innovation, and patents are only there to ENCOURAGE innovation. But I have difficulty figuring out where to draw the line between a novel arrangement of parts and a novel arrangements of ideas.
What the judge has said in essence is:
because APIs expose the operation of the internals to other languages, they have a fair use exception from copyright protection. Since Oracle defined the header as part of the API, it also cannot be protected by copyright. Everything else you are reading is pure speculation about how this will be interpreted by other judges, assuming the current case stands after the inevitable appeal.
As Westinghouse discovered, patents are there to provide a monopoly to the inventor, in order to reap the benefits of the invention. Westinghouse is big, partially because of patented invention(s).
When a patent expires, everyone is welcome to the game.
Who knows what "software patents" are about or why, I don't.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
