back to article Partial victory for Oracle in Java case

After deliberating over the weekend, the jury in the Oracle v. Google Java-copyright trial has returned a partial victory for Larry Ellison's crew. The jury found for Oracle in agreeing on the initial question: that Google did infringe on Oracle's copyright with the use of 37 APIs in Andorid, including nine lines of software …

COMMENTS

This topic is closed for new posts.

Page:

barely meaningful

well, there's two huge caveats here:

1) the jury doesn't decide the question whether the APIs are copyrightable at all. The jury was asked to decide 'in the case of the APIs being copyrightable, did Google infringe that copyright'?

2) the jury did not resolve the question of fair use yet: in the U.S. system, the precise way 'fair use' is implemented is that content is defined to be 'infringing copyright but under fair use'.

Google has of course run its defence on every possible grounds available, but the most solid ones from the start have been a) the APIs aren't copyrightable in the first place and b) if they are, it's fair use. So this jury decision barely means anything. Frankly it'd be a bit weird for a jury to say that Google *hadn't* infringed Oracle's copyright, if it's assumed that the APIs are copyrightable and fair use is left out of the question.

8
0

Re: barely meaningful

yep.. its a hollow victory, if you call it that.. oracle simply won on an assumption that SSO is copyrightable. I suppose we'll just have to wait for the judge to decide if it so (he promised he will if the jury thinks it is).

I think, like the EU courts ruled recently, he will say it's not copyrightable.

5
0

Yawn...

From trial of the century that was to decide the fate of Open Source, to the anti-climax of the century? So the verdict boils down to: "We think we know you stole, but we think its not what we thought you stole" ?

Oh well, there's probably time for a quick trip to the neighborhood cafe, before they make up their minds on this one.

0
0

This post has been deleted by a moderator

This post has been deleted by its author

This post has been deleted by its author

Thumb Up

Re: Bloody hell Iain!

Withdrawing that one, as the article has been updated now and contains the full facts :)

0
0
Happy

Re: Bloody hell Iain! even Ginny LaRoe did better - the shame ;-)

Google Is Probably the Big Winner in First Phase of Oracle Trial

By Ginny LaRoe

http://www.law.com/jsp/ca/PubArticleCA.jsp?id=1202552698215&Google_Is_Probably_the_Big_Winner_in_First_Phase_of_Oracle_Trial&slreturn=1

3
0
Silver badge
Pint

Nine lines of code?

Nine lines of code, out of how many thousands?, could be coincidence. Especially if the basic function is the same.

5
1
Happy

Re: Nine lines of code? Hypocrisy alert.

Quote: 1. Those are the "9 lines of code" I mentioned earlier.

Not all of the Java libraries were written by Sun or Oracle. There are a lot of 3rd party contributions. timsort was a 3rd party contribution. One of timsort's developers copied an internal Java function as an interim solution until he could link to that function directly. This function is a simple array bound check function a 1st year CS student could write call rangeCheck.

That 3rd party contributor was Google. When it came to implementing timsort for the Android, Google took their contribution and accidentally failed to replace rangeCheck. A minor mistake (Copyright violation) of no consequence (outside of this trial).

http://www.perlmonks.org/?displaytype=print;node_id=968578;replies=1

10
1
Anonymous Coward

@SilverWave

You're forgetting that Google copied 37 Java APIs too, not just 9 lines of code, according to this verdict.

0
14

Re: Nine lines of code? Hypocrisy alert.

Agree, I don't know the legalese of the situation, but I don't know how all of those class libraries donated by the community process could then be Oracle IP. Are they not open source under the GPL? Wasn't that the whole idea?

2
0

Re: @SilverWave

Assuming APIs are copyrightable (they shouldnt be really)

2
0

Re: Nine lines of code? Hypocrisy alert.

It's possible they might have had a copyright attribution system going on - when a dev submits code, he is alerted to the fact that the copyright to his code is now with Sun/Oracle. They had a similar system with OpenOffice too, so to is possible

0
0
Silver badge

Re: Nine lines of code? Hypocrisy alert.

Canonical have a contributor agreement they make you sign before they will use any of your code in one of their projects. In the current version of the agreement you licence your copyright to them but prior to July 2011 they asked you to sign over all copyright.

So yes, GPL projects can include copyrighted code.

0
0
Silver badge

@bean520: copyright attribution system

Probably... except Oracle forgot to check it at the start of the case. That's how we went from 51 infringing packages down to 37 *after* Google showed someone else owned the other 14 ;)

Right there we have proof that being in Java doesn't give Oracle ownership and an admission of that on record.

0
0
Anonymous Coward

Re: @SilverWave

"You're forgetting that Google copied 37 Java APIs too"

API's aren't copyright. I think this has already been covered in multiple cases on the subject of reverse engineering and interoperability.

Of course, case law being the screwy joke it is, all it needs is one stupid judge or jury to ignore the law and make it copyrightable. Which is why case law is evil, stupid, and bad.

1
0
Anonymous Coward

Google victory really.

As much as this has been spun antiGoogle, the only Oracle "victory" is the 9 lines of code out of the millions that were inspected, and found to have copyright problems.

But Oracle can only seek statutory damages, ranging from $200 to $150,000.

What's not been mentioned, are the two counts that Google were cleared of today.

All in all, Google are the ones that must surely be breathing a sigh of relief.

9
1
TJ1
Mushroom

Good day for Google... so far

The jury were not "deliberating over the weekend" - they were sent home Friday and told not to discuss. They continued deliberating Monday and reached a partial verdict.

The judge had earlier reserved to himself the decision on whether the "'structure sequence and organization" (SSO - in other words, the APIs of the disputed 37 packages) are copyrightable. For the purposes of question #1 he instructed the jury to assume SSO is copyrightable therefore the jury could hardly do anything else but find for Oracle.

Q1. As to the compilable code for the 37 Java API packages in question taken as a group:

Q1 (a) Has Oracle proven that Google infringed the overall structure, sequence and organization of copyrighted works. YES.

Q1(b). The jury could not decide if Google's use of the SSOs constituted fair-use or not.

Oracle made a point of shifting their accusations late-on to the Java API documentation - as most of us know this is generated by javadoc.

Q2. As to the documentation for the 37 Java API packages in question taken as a group:

Q2(a). The jury found Google did not infringe.

Q2(b) moot (no need to decide).

Q3 Google had already conceded it copied the following, the only issue to decide was if the use was de minimis (and therefore non-infringing).

Q3(a). Google DID infringed for the rangeCheck() method in TimSort.java and ComparableTimSort.java.

Q3(b). Google DID NOTinfringe for source code in 7 "Impl.java" files and onr "ACL" file.

Q3(c). Google DID NOT infringe for the English language comments in CodeSourceTest.java and CollectionCertStoreParametersTest.java.

Q4. An advisory for the judge. If Q1(a) is found for Oracle then...

Q4(a). Has Google proven that Sun and/or Oracle engaged in conduct Sun and/or Oracle knew or should have known would reasonably lead Google to believe that it would not need a license to use the structure, sequence and organization of the copyrighted compilable code. YES.

Q4(b) If so, has Google proven that it in fact reasonably relied on such conduct by Sun and/or Oracle in deciding to use the structure, sequence and organization of the copyrighted compilable code without obtaining a license. NO

Q4(b) is irrelevant if the judge finds that SSOs (APIs) are not copyrightable. If he finds they are copyrightable, well Google's loss is the least of your worries since it means that if upheld on appeal that the American software industry will implode as originators of APIs begin suing others using those APIs.

Google is putting forward a motion for mistrial based on Q1(b) not being answered. The basis of the appeal will likely be that it is established case law that all parts of a question should be answered.

If you thought the USA's software patent situation absurd, copyrightable APIs will have you retiring to a quiet cave with plenty of provisions whilst USA goes into meltdown.

12
0

Re: Good day for Google... so far

"Has Google proven that Sun and/or Oracle engaged in conduct Sun and/or Oracle knew or should have known would reasonably lead Google to believe that it would not need a license to use the structure, sequence and organization of the copyrighted compilable code. YES."

Not really. There are internal emails that prove Google employees knew Suns API's were copyrighted, and that demonstrated Google knew it needed a license. (i.e. Andy Rubin - the founder of Android Inc. and SVP of Android development at Google).

And while Schwarz supported Googles position, no less a person than James Gosling was pissed with Google for what they had done - along with the previous Sun CEO.

It's pretty obvious Google knew exactly what it was doing. Knew it would eventually end up in court, and was hoping for precisely this kind of outcome. i.e. one where they could delay and delay until it didn't matter because they dominated the market.

However you feel about Google, it's hard to imagine anyone but the most die hard Google fan wants a device market where only Android devices (and perhaps the odd iPhone) are available.

0
10
Anonymous Coward

Re: Good day for Google... so far

"it's hard to imagine anyone but the most die hard Google fan wants a device market where only Android devices (and perhaps the odd iPhone) are available."

Errm, what's that got to do with the trial? Or should they be convicted for making too good a mobile OS?

The key point you clearly don't grasp, is that Android is a OS that's available on pretty much any hardware manufacturer you want. For a consumers perspective, that's great. Nobody wants locked down OS.

I currently have an Xperia, but next year it might be a HTC or a Samsung, or a LG, or a Motorola.... And I can take my content with me on that journey.... That ensures Android players are competing and upping their game.

5
0
Silver badge

Re: Good day for Google... so far

And that is why the jury answered no for question 4b.

Sun did engage in conduct that would lead you to belive you didn't need a licence but Google did know they should really have a licence.

0
1

Re: Good day for Google... so far

"Errm, what's that got to do with the trial?"

Android devices are essentially being dumped onto markets at price points far lower than anyone else can manage (Nokia, RIM etc) in part because they don't pay the same license fees as others (Nokia, RIM etc).

Mobile operators push Androids (often to people who ask for other devices) because they make a stack of money.

For most people other devices are perfectly suitable but they don't get to buy them.

The idea that a market the size of the smartphone market is developing "by chance" when you have organisations the size of Google and Apple at play is naieve at best.

I get your point about the OS and content, but that doesn't make Googles actions any less evil. (And you can do the same with other devices - some app vendors will even let you switch devices).

"Or should they be convicted for making too good a mobile OS?"

Certainly not. Linux is a half decent OS, though not the best techincally by a country mile. They should be convicted because of anti-competitive action, only possible because of their sheer size. They've been very clever of course, so even I have to admit it's unlikely anyone can stop their domination.

I take it you're a die hard Google fan. Some of us would rather have a choice.

0
4
Happy

... there's nothing except one line of statutory damages. 150K USD tops.

Judge: Unless court can give verdict on 1B in favor of Oracle,there's nothing except one line of statutory damages

http://www.groklaw.net/article.php?story=20120507122749740

6
1
Anonymous Coward

Re: ... there's nothing except one line of statutory damages. 150K USD tops.

Well Groklaw would say that, wouldn't they.... Google can do no harm in their eyes.

It's as unbiased as Fosspatents. Well maybe less since in Groklaw's case we don't know exactly who's paying the bills..

1
16
Anonymous Coward

@AC 22:40GMT - Re: ... there's nothing except one line of statutory damages. 150K USD tops.

Instead of a cold down-vote, I'll try to help you understand why others did it.

1. Groklaw talk is always based on accurate court documents and first hand testimony of eye witnesses present during the trial. You seem to be reading their articles so you should already know this by now.

2. They are as unbiased as Fosspatents with the exception that they are right almost every time, and this is because of what I said in #1 above.

3. In case you feel Groklaw has a special affection for Google, please go back to their site and read about the Apple vs. Psystar. You will discover to your horror that they were clearly with Apple on that matter not because they love Apple much but because that was in their view the correct way to apply copyright laws. It was about right vs. wrong.

4. As for paying their bills, there are some people who don't mind donating to a website like Groklaw while others prefer to finance shills and astroturfers. Guess who are wasting their money here.

12
1
Anonymous Coward

No, Oracle lost

The jury found for Oracle only on question 1A, and that only because the judge instructed them to *assume* that API's can be copyrighted. They deadlocked on question 1B (fair use).

Whether API's can be copyrighted or not - as a matter of law - will be ruled on by the judge later on this month. If the Court rules that API's cannot be copyrighted, then the entire Question 1 becomes moot.

Google has already admitted to copying 9 lines of code - they say inadvertently - and they removed these lines from Android. Potential liability for copying 9 lines of code: USD $0.00, as per Oracle's own expert's testimony.

The only question where Oracle might have won was the documentation. Oracle lost on that one.

Why does the title of this article state that the verdict is a "partial victory for Oracle" is very difficult to comprehend.

The way things are going, when all is said and done, Oracle won't even recoup the cost of litigation.

10
1
Happy

the jury seems unswayed. Ah... NO see 4a

4b is a trick question.

0
0
Happy

Re: the jury seems unswayed. Ah... NO see 4a

question 4 was this:

A. Has Google proven that Sun and/or Oracle engaged in conduct Sun and/or Oracle knew or should have known would reasonably lead Google to believe that it would not need a license to use the structure, sequence, and organization of the copyrighted compilable code?

And the jury said Yes.

6
0
Silver badge

On 1A: a better portrayal would be that the judge ticked the YES box on 1A, handed the form to the jury and assumed jury nullification would not be used to cross it out. He didn't give them a choice on how to vote.

It's unfortunate the jury couldn't agree on fair use, it almost guarantees a retrial unless he now rules there's no case to answer at all (AKA APIs aren't copyrightable). It should scare the shit out of Oracle that they couldn't win on the major use Google made no attempt to deny, that it forces a precedent setting ruling the court so obviously wanted to avoid.

5
1
Bronze badge
Facepalm

Partial Victory?

Not sure what Oracle won - this part of the trial was really about whether Google had fair use of the APIs. It didn't deny copying the (cough) SSO of the API - this was just deemed to not be subject to copyright previously and the question of whether it is is still to be decided.

The judge asked the jury to consider it copyrighted and therefore, basically, told them to tick the first box.

Do you mean the partial win was because they copied rangeCheck? Really?

This article seems to be a cobble together of the BBC and ZDNet articles and seeing as Rachel King seemed to be lost in a world of her own, I'm not sure the ZDNet articles should be used as source material.

6
0

Re: Partial Victory?

no, with the assumption to copyright, they are talking about 37 whole APIs

0
0
Happy

winners and oracles

So, I can still write software if I want to. I win too.

1
0

Unless Oracle gets the APIs, they are going to need to take the bus...

back to Redwood Shores. Those nine lines of code that Google was found to have violated have no monetary impact on Oracle. They probably were developed through the community process and are not even Oracle's IP. They will not get a settlement large enough to cover their legal fees from those nine lines of code.

To win a meaningful victory, they will need to get a ruling that: 1) Yes, APIs are copyrightable. That would fly in the face of decades of precedence and every software developer and company on the planet would be sending in letters to prevent it from happening. 2) If they are copyrightable, that they can enforce that copyright (not fair use). The odds of Oracle prevailing on both of those counts is highly unlikely. The judge told the jury to assume Oracle prevails on 1) (presumably so he can later rule that APIs are not copyrightable) and the jury still was deadlocked on 2).

2
0
Silver badge

Schwartz definetly influenced the jury. Oracle are trying to block him right now

"Although the former Sun CEO Jonathan Schwartz testified in Google's favor, saying the company never had a problem with the Chocolate Factory's implementation of Java, the jury seems unswayed."

That's not what happened.

They answered YES to 4a: the jury did believe Sun 'never had a problem' or at least told the world there was no problem. Oracle are fighting to keep Schwartz from testifying in the next phase because the jury WERE swayed by his testimony. With the evidence that's since emerged about how official the blog really was they're right to be scared.

4B really is a trick question. Not a ruling, just an opinion for the judge to consider and compared to 4A pretty minor, if only because the burden of proof was on Google to prove it, not Oracle to disprove it. Would have been nice to have but 4A did the damage.

Ultimately, if Oracle don't beg for a settlement before phase 3, the 4A result also applies to the patent claims and can only reduce any payout the jury decides. There are only 2 ways to kill a patent suit, prove you aren't infringing OR get a low valuation on the infringement: looks like Google have a low valuation coming.

Said it at the start, I expect Oracle to spend more than they win. Or put another way, Google saved nearly $7.4bil when they chose not to bid against Larry Ellison for Sun. Sun should have taken the chance to be relevant when they had it, they wouldn't have needed to sell to Oracle if they had.

2
0
Anonymous Coward

I guess after this court battle

and the subsequent appeals, Larry Ellison will have to come in front of Oracle shareholders and explain how he managed to ruin whatever value they might have got from Sun acquisition. Oh, boy, how they will grill him!

Oracle seems to have made the same mistake as SCO. They figured out who to sue for billions and went ahead without taking a good advice from their lawyers and engineers. This is what happens with lawsuits against common sense.

1
0
Linux

speculation in lay terms

I believe that when the dust settles on this, Google will at worst be brow beaten by the courts slightly and told to be more careful and they would have spent a lot of money to lawyers to settle the issue once and for all forever over a non question raised by bad boy Oracle.

Oracle on the other hand will have embarrassed itself and fubared the already fubared reputation they already have with the "needed Open Source community". Not only that but having lost the case, being so roundly trounced and made to look more foolish than they did in even attempting this nonsense... API copyrights. they will likely be forked out of any loop with future developments coming down the road. Would you trust a backstabbing raving madman claiming to be an Oracle of omniscience.

They spent a fortune only to settle once and for all an issue best left alone... They were forewarned it might not be in their best interests and did it anyway... Greed pure greed and now everybody knows what their game plan is. In a way :-) Larry got it right some years ago, Linux is unbreakable.

1
0
Silver badge
Devil

Hmmm

Copying 9 lines of code in a product which some claim to not be Java but a separate clean room release.

Doesn't seem like much, although here's the sticky part. 9 lines of code found to be infringing kills any 'clean room' defense of Google. You may not grok the significance, doesn't mean its insignificant.

Then there's the copyright issue on the APIs. That's going to be an interesting situation.

The issue of Fair Use is also an interesting one.

Google wants a mistrial and that's an important thing to understand. If this was insignificant, they wouldn't care about it, now would they?

0
7
Anonymous Coward

Re: Hmmm

Whether or not Android is, or is not, a clean-room reverse engineering of a Java implementation, is not being litigated in this trial.

Insofar as the 9 lines of copied code are concerned, here's a verbatim quote from judge Alsup on this matter: "Zero finding of liability so far".

You can read the transcripts yourself.

3
0
Silver badge
Big Brother

@AC. Re: Hmmm

'Whether or not Android is, or is not, a clean-room reverse engineering of a Java implementation, is not being litigated in this trial.'

Do you play checkers or chess?

I doubt you would get that reference. The point is that you can't think in terms of a linear battle.

First, this is the first of several lawsuits. You should know that by now.

Second, there is this thing called discovery... ;-) yes, that's right. It's not a fishing expedition because as you can see, the jury sided with Oracle on the major points. (do you honestly think ou can get a jury together to decide a very technical argument on what define's fair use?)

There is more to this lawsuit than you and most commentards realize.

In the background, you have the DoJ and the EU watching.

While the jury was out, the judge read in to the record that Google lost millions on Android.

I seriously doubt you would grasp the importance of that.

But I digress. Getting back to the issue of Dalvik not being a clean room version of Java... It shows that it is a derivative work. Kind of important for the next trial, no?

Like I said, big brother is watching...

1
4
Silver badge

Re: @Ian Trolls-a-lot Gumby

One minor problem with your clean room argument. Google contributed TimSort TO Java, they didn't take it FROM Java or reimplement an existing Java API. They created both versions from the freely usable Python original. There are no clean room issues here, you don't need one to contribute NEW API and source.

You also need to remember why RangeCheck was copied: the implementation was being contributed to Java with the intention of refactoring Java to use the existing internal RangeCheck support. Not removing it in their own copy is simply careless, not the making of a $bazillion fine.

3
0
Silver badge
Mushroom

Shirley you jest...Re: @Ian Trolls-a-lot Gumby

Sigh.

I guess because you don't grok the issue, I'm a troll?

The jury just found that Dalvik is a derivative of Java and not a clean room. Like I've said in earlier posts, you show one instance of the 'clean room' being breached, its no longer a clean room release.

Look at Harmony. Its a clean room implementation.

Even Apache distanced themselves from Google when Google claimed it was a clean room based on Harmony.

You find the infraction. Small, or not, the infraction occurred.

While everyone is looking at a minimal amount of damages which would be a pinky slap against the back of the writs. (Not even a full hand, but just a tip of the pinky.)

But it sets up the larger issue.

It also sets up the fact that this was a willful infringement.

There's more to it. But you wouldn't understand it. ;-)

The key is that this was round 1 and its in the following lawsuits that will start to spell out the issues.

0
1

meanwhile

Meanwhile North Korea continues to sap millions from the world through terrorism, and possibly is continuing trying to build a nuke, kids all over are starving, being abused, waiting for adoption, being sold into slavery, getting poor brainwashing education that will forever confuse the morality of some and the majority of us are still stuck on Earth, which is rapidly accumulating mercury and fluoride and insecticide poisons on the ground, while run off drains the fertilizer into the sea, causing massive dead zones and deadly algae blooms and bees continue to die off in large numbers, and the rights of people all over the world continue to be taken away, and police and evil leaders and oppressive banks using outrageous fees continue to get away with crimes.

Spend your money on things that matter people, that goes for the poor, and the rich.

0
6
Silver badge

Re: meanwhile

I think you're lost, the hippy drum circle is over that way >>>>

3
0

Java API's are copyrighted...

...or at least that's what Andy Rubin thought. And he should know.

It's pretty obvious someone at Sun wanted to generate revenue from Java on the mobile side (hence J2ME is not open), but figured they'd do ok selling support for server side Java - so open sourced J2SE. And they added some protection for the open source by registering patents. Perhaps they suspected someone might try using J2SE on mobile.

Open source J2SE was licensed using GPL - a license that allows unlimited use providing you also provide GPL licensed software extensions.

Google does not. It licenses Android using ASL. Using ASL allows Google and Android manufacturers to protect their software extensions (they're not open - but should be).

This case is enormously important for one of the worlds fastest growing industries. Having to license Java from Google would increase the base cost of Android phones and reduce the price differentials to other brands leading to more diversity in the market place.

0
6
Anonymous Coward

Re: Java API's are copyrighted...

> Java API's are copyrighted or at least that's what Andy Rubin thought. And he should know.

What Andy Rubin thought he knew, or what you think Andy Rubin thought he knew, is completely irrelevant to whether or not Java API's are subject to copyright or not. The legal status of the copyright of Java API's is a matter of law, not a matter of opinion.

4
0
Anonymous Coward

Re: Java API's are copyrighted...

THe emails you refer to, he was asking if they were, not stating they were...

BIG difference.. And even if Google at one time may have questioned or thought they might have been, bears no relevance on if they were or not...

3
0
Silver badge
Boffin

Re: Java API's are copyrighted...

Just a small nit...

GPL isn't as open as you think.

APL is more open, and if you want to get down to it... Stripping out the headers and comments would be a violation of the GPL...

0
1
Silver badge
Devil

Re: Java API's are copyrighted...

Another small nit...

GPL and ASL are not compatible. You can't take GPL code and reissue it under APL.

That's a no no.

So while the first trial won't garner Oracle billions, it sets up the next trial.

Had the jury not ruled against Google, would be able to defend itself by saying it was a clean room release like Harmony. But it's not...

0
1
Silver badge
Devil

Re: Java API's are copyrighted...

Third nit.

Using android has hown that there are hidden fees.

Phone manufacturers are actually paying more in legal fees because of Android.

0
1

Page:

This topic is closed for new posts.

Forums