Oracle v Google round-up: The show so far After years of waiting for the contenders to open fire, the Oracle-Google shooting match is now on, and the bullets are pretty expensive. The opening salvos have landed, so let's take stock. It’s worth keeping at the back of your mind how strange it is to be here at all. Judge Alsup has spent three years trying to persuade …
No it doesn't make him corrupt but it puts a huge slant on anything he says - something that Andrew fails to point out in his article
"As Florian Mueller points out in his analysis, Schwartz’s comments cut little ice with the Judge" has a very different meaning if he's being paid by Oracle - "As an Oracle paid consultant points out in his analysis, Schwartz’s comments cut little ice with the Judge"
It's a big difference.
Doesn't 'work for' is a consultant of.
There's a hell of a difference in law and actual fact.
Here's what's gonna happen...
1. Google found guilty with a one off payment and ongoing licensing fees to every handset maker. So much so that it becomes unfeasible financially to produce an Android phone because of the additional Microsoft licensing fees and zero profit for all manufacturers but Samsung
2. Google abandons Android and releases 'Chrome for Phone' - they're already working on it.
3. Samsung forks Android - it will happen, but not for these reasons
4. Profit???
Sorry should have said "takes Oracles money in return for services rendered". That of course doesnt make him an employee but he does "work" for Oracle just like he does work for Microsoft. Now if I am a "consultant", that has just signed up another large multi-national customer, my first public act is not going to be report negativily on their current biggest trial. So I stand by my statement he is not un-biased.
Quite what he would do if MS and Oracle ended up in a big patent trial against one another would be really interesting. Keep his fingers off a keyboard and not return calls from lazy jounalists would be my guess.
His track record on actually being correct with his opinions is not exactly stellar,except possibly in his own eyes.
"Only need to show one"
Go lookup what Googles 'de minimis' affirmative defence means... *amount* of copying *does* matter and 1 case of 9 lines in 15million seems a pretty good candidate for de minimis.
Particularly when the copied function is also a good example of unprotectable code - on grounds that you can't implement it in any functionally different way and that's its function and use is required to fulfil the Sort method contract.
I expect de minimis will cover the test suite files HP smuggled into the tree as well, on the basis they had no actual use or value to Android.
US software copyright law is not as simple as most of you seem to believe. It's messy, looks unfair (multiple ways to get away with copying for instance) largely because it evolved to prevent a whole family of abuses by copyright holders. There's a reason Oracle are trying to create new law here, the existing precedents don't support their claims.
Sun wove a contractual web designed to lockin control of Java but failed to cover all loopholes.
While the entire affair is not particularly pretty all code shown by Oracle in this slide deck puts Google firmly in the clear.
There is a well known precedent in IPR law known as the Lego precedent. Lego sued some of its competitors for allegedly copying its "bricks" and tried that based on both Copyright grounds and Design Patent grounds. It failed.
The reason is that in order to get a functional "brick" you need to have the recesses on the bottom, the pointy bits on top and in order to connect bricks in different combinations you have to have those bits rounded. The reason for the Lego brick shape is not design, not copyright - it is because it is functional and any other brick based system will have to have a similar shape.
Coming back to Google vs Oracle, it has no case for copying an API and there is very little case to apply copyright to an API. Anything trying to implement the API will be functionally similar to the existing implementation - same as Megablock bricks are similar to Lego bricks. From there on, if an API is public and if there is no _OTHER_ means of prohibiting the implementation of said API, an implementor can copy the definitions of the functions and there is very little that can be done against him. It is a form of the "Lego precedent" - there is no other way to do that.
Oracle is showing only definitions and function declarations in their slide decks. These fail "The Lego Test". So will any piece of code for which there is "only one sane way of doing it". They have to show a piece of code which requires a non-trivial implementation which has been copied.
Did Google copy chunks of code is not relevant here, it is "did it copy something that is subject to IPR protection".
It will be good if this sees its day in court and it creates a precedent, because regardless of who wins the scope of frivolous application of "Copyright" tags to include files, library definitions, schemas, etc - things essential for interop is likely to see some drastic reduction and this is good for everyone.
By the way, SnOracle is not the biggest offender here - most opensource projects are way worse including. They frivolously slap GPL2 on include files, API definitions and other items from the 2+2=4 variety.
This is like copying commentary. Yes, in this world where you can trademark A Goat On a Roof, this counts as "Intellectual Property". Remember we are going up against copy-friendly people like China with that kind of stupid.
The only really nasty thing one can say about this that in both cases the error report in the exception will certainly confuse the hapless developer. Too half-assed to actually say "Out of range [X,Y] with index Z", instead just saying "Z"? Yes, good work, one side. Or both sides.
Strange. Read the dissection at Groklaw and you get a totally different perspective.
The general view there is that this is largely about whether or not Google need a licence to use the JAVA library API, so this largely rests on whether or not an API can be covered by "copyright" as a creative work. Just because an ex-SUN engineer thinks that Google DO need to acquire a licence is entirely irrelevant: he is not a lawyer and probably doesn't know what he is talking about. The context of that statement is likely to be risk aversion and playing "fair" to his prior employer.
Most of the examples of supposed "copying" are debatable, although not necessarily without merit.
Im not interested in bias.
However, I am interested in resources that have a good track record for making the right call.
Groklaw consistently calls it correctly, predicting trouncings for SCO, Psystar and others. Mueller consistently calls it incorrectly.
At Groklaw you can ignore the editorial and go to the actual documents. I don't recall any such luxury afforded by Meuller and his ilk.
Groklaw's efforts are voluntary. Meuller makes $ from relationships with the companies he shills for.
Disappointingly, Groklaw has its own bias as even *facts* can be interpreted any way you want - a good example of that is their stance towards Google, Google can simply do no harm in their eyes (yes, I know that will give down votes, but I happen to be in a position where I can accurately judge this).
That aside, that ex-SUN engineer has been pretty close to the creation of the platform and can thus be taken as having a reasonably good idea of what works or not. In other words, if he says "other options suck" than you come back to the issue of "how did Google end up using Java commercially without a license".
OK, I'm possibly biased too: Google has a rather long history of trying to get away with anything it damn well likes so from a simple behavioural angle I cannot see them changing their MO, which makes the accusation more likely to be true.
> I happen to be in a position where I can accurately judge this
Your vague, unsubstantiated claim of authority from a position of anonymity certainly strengthens your argument. Well, I'm convinced!
Of course Groklaw is "biased". Anyone with any critical thinking skills knows every source is biased - that's what it means to have critical thinking skills. "Unbiased" is at best some sort of asymptotic ideal.
> In a nutshell, Google admits copying Sun Java code into Android.
Actually no. Google admits reimplementing the Java SE APIs in Android. The problem is that due to the nature of APIs, the Java originals and Google's (well Apache Harmony's) reimplementation are very very similar. The irony is that if the APIs were *identical* there would be no risk of Java fragmentation.
Also Andrew is still retailing the old FUD about the Lindholm email - which was written *after* Oracle began the process of suing Google and the latter were wondering whether they should just ditch Java.
The sad thing about this is that it all started because Sun chose to fragment Java into a multiplicity of editions, and flat-out refused to license Java SE for any handheld device.
The sad thing here is there are no real villains in the story...
Hmmm, you clearly not a Java developer if you think it would be a good idea to run the full J2SE stack on a smartphone (going back 10 years or so). It might be cool, but no, you have to work with subsets of the main J2SE platform. Also I am not sure how that would be fragmentation, It's just good platform design.
> The sad thing about this is that it all started because Sun chose to fragment Java into a multiplicity of editions, and flat-out refused to license Java SE for any handheld device.
I don't think that producing the different variants for different applications was too much of an issue. At the time, Java ME was fine for low-end devices and a 'full' SE implementation was never going to make much sense anywhere except the desktop. To me, the sad thing was the way in which Sun carried on insisting that Java ME was still good enough for the new generation of mobile devices when it clearly wasn't up to scratch. Faced with that, you can understand why Google decided to fork off and build Android instead.
Extending copyright to APIs: well, you know what I think about that.
http://www.theregister.co.uk/2012/04/17/google_oracle_copyright_on_languages/
I'm really surprised more people haven't picked up on it, because it has huge disruptive effects on the industry. And not in any good way.
The Lindholm email is just what it is. Page asked a very knowledgeable Java veteran what to do. The guy said: we need a license to do what we're doing. Google chose to poker bluff its way into court instead.
Have you researched the circumstances of the Lindholm email? Why have you assumed that "Page asked a very knowledgeable Java veteran what to do"? My understanding is that Lindholm was NOT asked directly by either Page or Brin but by the Google committee tasked with dealing with Oracle's legal action; this committee included Google's top lawyers (Kent Walker and Ben Lee) and Andy Rubin, head of Android operations at Google. Presumably that's why the e-mail was addressed to only Lee and Rubin with a salutation to "Andy".
Lindholm is technical - would you rely on his judgement on the necessity for (further) licencing in a billion dollar lawsuit when you're paying for a bunch of in-house lawyers?
In court, Tim Lindholm said that he didn't mean that Google needed to get a license specifically from Sun.
Wall Street Journal: "Google Employee Testifies His Java Email Was Misinterpreted":
http://blogs.wsj.com/digits/2012/04/19/google-employee-testifies-his-java-email-was-misinterpreted/
Actually, no! Lindholm was tasked to see if he can find other technically superior means of achieving the same goal. Nobody asked him if Google should take a license on anything, that would be a job for the lawyers to decide if Google is doing something illegal or not.
> If Oracle shows evidence that suggests that the clean room code isn't clean
Oracle isn't presenting any evidence that it isn't clean.
What Oracle claims is that any attempt to make a work-alike of Java is necessarily a copyright infringement. This is clearly contrary to the way the software industry has always worked, and appears to be contrary to the way US law works.
Judge Alsup has not ruled on this yet - but he has asked some *very* pointed questions about it. It would not surprise me one bit for him to so rule before the end of the trial.
BSF pulled the same sort of stunts when they represented SCO. It didn't work. SCO would have been demolished in court had they not filed for bankruptcy. I don't think Oracle will try the same tack.
Vic.
What I can see in the article is just Hilarious. There are not many ways a stupid range-checking function can be implemented in a sane way. I can see exactly three sane options:
A) check upper and lower bounds at the same time and throw exception if wrong
B) check upper first and then lower
C) check lower, then upper
Hopefully Google will communicate that little fact to the judge/jury. This is very much like SCO BULLCRAP !
> Actually no.
Strictly speaking, "yes".
> Google admits reimplementing the Java SE APIs in Android.
They also admitted to carrying a few copied files - believed to be put in the repository by a third party, and removed once Google found out, prior to the trial beginning.
Judge Alsup was apoplectic. Oracle did themselves no favours whatsoever by pushing that bit...
Vic.
As I said when this case was filed couple years ago, Google should just have licensed Java and be done with it. In the end, Android would look exactly the same, aside for some additional J2ME stuff added in. Also, with time, the J2ME and Andriod APIs would have likely just been merged.
Aside from the legal issues, a more interesting aspect of the case to me is the fact that Google seems to be one of the few mobile platforms developers who actually did their research to see what SDK would work in that space. If you look at RIM, HP (WebOS), and Nokia, all the other mobile Linux distros they all pursued some other brand new, and unproven (in the sense of mass developer uptake, not technical merits) "html5/css APIs", or worst some native C/C++ library like QT (nthoign wrong with QT, but number of mobile developers who would know QT is very low) etc ...
Kudos to Google, and Apple for that matter, for realizing that success in the mobile space has less to do with developing a sexy, new set of APIs, but more about leveraging what's already being used in a similar capacity. Or a better way to put this "first and foremost, appeal to lazy developers who don't want to lean yet another language/APIs, just for some potential incremental benefit".
You forgot it was Sun that absolutely vetoed licensing anything better than J2ME for mobile devices. Not that it stopped Google attempting to negotiate but ultimately they couldn't unilaterally decide to licence the version of Java they needed. Sun sank that.
What's more worrying is, if Oracle can force Google to licence they can force them to accept that web of legal traps and buying the licence will shut down Android instantly. The only upside of that is it's so outrageous a result Google will get 5 years of appeals or so to cleanse Android before it's enforced.
Sadly I cant even agree Google should have bought Sun. If Oracle paid $7.4bil and would likely go much higher to protect their main business's reliance on Java, in a bidding war Sun would cost much more than any imaginable fine Google could expect. Though Oracle still shout $1bil the court is strongly suggesting $30-40mil as a starting point.
It's interesting that Larry Ellison admitted Oracle considered building a phone based on JavaFX, an option only available to the platform owner and close to an admission you just can't build a smartphone on J2ME.
Oracle aren't going to cover their costs when this is over.
This is exactly it. Google wanted to license Java, in just the way that Sun wanted, but needed some freedom of action. And Sun wouldn't give that freedom of action at any price. So there was no deal, and Google had to find a legal way around that roadblock.
No doubt on good advice of counsel they found a way around. Now they have to convince the jury though, and that can be tricky.
I think it was because Sun simply didn't want to sell what Google needed. This is possibly where Google went wrong: they choose to do something that looked like going ahead without license.
Personally, I wouldn't put it past Google that they did that and tried to get away with it (it seems to be their MO to ignore laws), but it has to be proven that that is what they did, and there appears to be enough confusion to warrant doubt. So, it's up to the judge to decide if they did or didn't.
If they lose this it may blow a ginormous smoking hole in the whole Android platform and lead to all sorts of followup lawsuits from OEMs (read: its of lawyer money earned and possibly a seriously large dent in the Google share price), if they win Oracle will surely follow this up - I cannot see Oracle spending so much effort on this unless they think they have a chance, although it could also be the classic play garden for *WAY* too rich people.
There are two people involved here who both have wide and high doors to their offices so as not to impede their ego entering with them..
> they choose to do something that looked like going ahead without license.
If Google had obtained a license from Sun it would have been for Sun's _implemenation_ of Java and the APIs, but would only have been for ME.
Google determined that they did not want the limited ME version and so used the Apache Harmony implementation for which a license was available - the Apache licence. Google did not need a license from Sun as long as they did not call it Java.
Oracle is trying to sue over alleged 'copyright' over the APIs organisation and sequence. This is like one publisher suing another over the use of 'Chapter 1', 'Chapter 2', etc because the 'selection, organisation and sequence' is the same.
the Apache Harmony project ? If I'm not mistaking it is Java without the right to be named Java.
Most of Java has been open sourced under GPL license and (correct me if I'm wrong here) you may copy it freely so it would be nice if you could enlighten us regarding the parts Google copied, except for the bit where Oracle claims the mere creative arrangement of APIs involved a lot of precious IP Google has stolen
Also, instead of Florian Mueller, why didn't you put here some quotes from Google side, we know they also put some slides illustrating their position.
Of course you have the right to cheer Oracle's performance in court but we also have the right to fair and unbiased reporting here at El Reg.
This is Oracles week. They control pretty much everything while they present their case and it really should be looking devastating to Google 5 days in. But it's not.
Really doesn't help their case that they keep banging on about issues like Java fragmentation that aren't in the case and would only affect licensor-licensee relationships, something Google careful avoided. Remember, copyright claims, patent claims (later in the case) but NO contractual claims at all. Sun did indeed carefully spin a web, Oracle are clutching at straws because it didn't catch Google.
"That’s the meat of the prosecution’s copyright case."
That pretty much sums up this article. Had you even tried to include the defense's side, the article would be a very different thing indeed.
"So... it doesn’t look great for Google."
Not with the one-sided presentation you've given, no. But it didn't look good for OJ Simpson, either -- until the defense had their say.
"That’s the meat of the prosecution’s copyright case."
Hah, sums the article up for me too: seeing as how there's no prosecutor in a civil proceeding, methinks this article is somewhat under-researched, or at least there's a lack of understanding of the prceedings on show.
Don't even get me on to the quoting of dear Florian as some sort of impartial pundit...
I think you're being grossly unfair - you're effectively asking "who's winning?" in the first innings of a cricket match. That's why newspapers and telly news people tend to end their reports of on-going court cases with the words "the trial continues".
You have a point about prosecution v litigant. This is a civil matter, but that doesn't distract from my above point.
C.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
