About time! Google has responded to Oracle’s amended complaint in the big Oracle v Google patent and copyright hoedown, and it’s a contradictory grab bag of various defences, basically throwing everything and seeing what sticks – a normal and common course of events in cases like this. There are some juicy claims in there.
The biggest, and most interesting, claim relates to the Google-copied-our-code claim from Oracle, and is one that looks an awful lot like what happened with the 300 lines of code SCO claimed the Linux world had copied. In that case, IBM showed the judge that SCO had compared the supposedly copied code with Linux in code in such a way as to fabricate the idea of it being copied. Google is doing the same with the supposedly copied class.
“Google states further that Oracle has redacted or deleted from the materials shown in Exhibit J both expressive material and copyright headers that appear in the actual materials, which are significant elements and features of the files in question,” the filing reads.
Another interesting note is that Google specifically explains that while Android programs are written in Java, Dalvik bytecode distinctly different from Java bytecode, and the Dalvik VM is distinctly different from the Java VM. In other words, “the Dalvik VM is not a Java VM”.
Other than those, there’s a whole boatload of defences in here regarding the patent infringement claims. Google claims all patents in question are invalid, “because one or more claims are directed to abstract ideas or other non-statutory subject matter”. They also bring up prior art, obviousness, and even claims that the patent language is unclear. Then they go into things like that it took Oracle too long to assert its patents, that Sun’s and Oracle’s actions (and in some cases, lack of action) led to an implied license, that Sun/Oracle improperly expanded the scope of the patents (misuse), and more. Regarding the copyright infringement claim, Google claims, among many other things, fair use, that the bits used are too small to constitute copyright infringement.
This all seems contradictory, but that’s normal in cases like this. Google can throw whatever it wants at the judge and see what sticks, and that’s exactly what they’re doing. Some points raised are interesting, but of course, it’ll be up to the courts to decide.
It’s been fairly conclusively shown that Google’s PoligcyNodeImpl is a result of decompilation of Sun’s code [1] which is definitely illegal.
However, that is only one file and I can’t see why they couldn’t just say “Oh yeah we screwed up with that file, we’ll reimplement it legally.”. The damages surely couldn’t be too great…
[1] http://www.reddit.com/r/programming/comments/dxmvr/oracle_google_di…
That’s not what the link says.
The link says it was decompilation of Sun’s code. You understand how references work right? The little number comes *after* the thing you are talking about.
And when I said it is illegal, perhaps that was badly worded. I didn’t mean decompilation *itself* was illegal, but decompilation and then redistribution under a different licence is clearly illegal.
That is, unless the original licence says you can do that. Let’s see. It’s GPL plus classpath. Some quotes:
From the GPL:
It doesn’t explicitly mention decompilation, but I think it is pretty clear.
Besides that would be a really stupid loophole. You could just compile any GPL program with full debug information and decompile it to get an equivalent non-GPL program. Obviously you’re not allowed to do that.
In all seriousness, how is that any different than any other form of reverse engineering, considering you can have access to the machine code of any running/compiled program?
It’s not. You can’t just decompile MS Office, for example, and then put the decompiled source on your website for free.
That is true, but that is because the license for MS Office says that you cannot do that.
In this particular case, however, the license of the code in question requires you to do the equivalent of putting the “source on your website for free”.
For some inexplicable reason, people seem to be having a real hard time understanding this point.
Edited 2010-11-15 22:18 UTC
Yes, UNDER THE GPL! How is this so hard to understand?!
Ironic.
Whoosh!
I think you have utterly missed the point. Google made the code in question available via a git repository. Therefore, they made the source code available to the public.
Now, if Oracle undertake a long battle, and somehow convince the court against the evidence and the odds that Google have actually violated Oracle’s copyright, then Oracle may win the case and get the court to force Goggle to obey the terms of the license for that code in question.
The code in question is licensed under the GPL.
Goggle might be forced to … make the source code available. Perhpas they might have to put it in a git repository or something …
Oh wait.
Now do you see the irony?
If you do not, then my original comment stands utterly vindicated … people seem to be having a real hard time understanding this point. You amongst them.
Edited 2010-11-16 12:54 UTC
No, they’re just brighter than you. It’s actually not that complicated. Maybe I can simplify it for you.
1. Oracle/Sun writes PolicyNodeImpl.java. They release it under the GPL, which states that you have a licence to use the code if you release your changes AND DO NOT RELICENSE IT
2. Google copy the code (via decompilation, but that makes no difference) AND THEY RELICENSE IT.
3. Since they didn’t comply with the terms of the GPL, they no longer have a licence to use the code and have therefore committed copyright infringement.
I don’t know how to make this any clearer.
All of this is patently obvious, and indeed is exactly the outcome I have previously stated would probably be the case if Google were indeed found to have violated copyright. Google would have to re-license the code in question under the GPL.
I said this ages ago, in one of the early postings on this very thread.
http://www.osnews.com/permalink?449978
To quote myself:
How about them apples.
The point that has apparently gone miles over your head is that this will amount to a hill of beans. It will mean didly squat.
Google will have to change the headers of the source code text files, to indicate that they are GPL licensed. Since Google are already offering the source code to the public, that is the sum total of the remedy required.
Everything else will carry on as before.
This is the point that has gone waaaaaaaaaaaayyyyyyy over your head.
The point you have been trying to convince me of was already my position in the first place. It was my position from the very outset … IF google actually did breach copyright.
So, I guess the only thing that leaves to be said is for me to virtually-scowl at you and say: “Quit your jibber-jabber, fool!”
Oh yeah.
Edited 2010-11-17 10:16 UTC
From the comments against your link:
De-compiling GPL code is most decidedly not illegal.
From the authors of the GPL:
http://www.gnu.org/philosophy/free-sw.html
The first freedom allows you to run the code for the purpose of studying it and finding out how it works. The second freedom, “freedom to study how the program works” covers de-compilation. The last freedom covers re-distribution of modified versions.
I think you must have got confused between GPL code, and restrictions which are included in EULAs for proprietary code.
De-compilation of code is not even “illegal” under copyright law itself. Only distribution of illegal copies of a work is illegal under copyright law. De-compilation would at best be a contract violation of a software product EULA.
There is no EULA for GPL code, there is only the GPL itself, and the GPL grants permissions for activities such as de-compilation.
Edited 2010-11-15 02:47 UTC
But the decompiled GPL code wasn’t licensed under GPL license, wasn’t?
Your statement does not make sense. The “decompiled GPL code” is obviously under the GPL since you said that its, uhm, GPL. It can’t both be GPL and not.
Actually, it can; dual licensing is possible. Consider Qt, which is available under both the terms of the LGPL, and under a commercial development license.
As far as I can tell, the Classpath Exception is fairly similar to the LGPL. It allows you to link your own executables with the code covered by the Classpath exception, and release the whole under any license you want. In this case, the terms of the GPL would not cover the binary you release, even tho it includes code that normally would be.
If it was decompiled then placed into another source tree, then the resulting derived work would also need to be licensed under the GPL in order for Google to distribute it.
Google didn’t distribute it. The resulting derived work is not part of Android. AFAIK it is used as test code only, in the process of making Android. In other words, this derived work is for internal use only. GPL license is not therefore required.
Edited 2010-11-15 05:32 UTC
I beg to differ – decompiling a binary to source code and then making that source code publicly available is arguably a re-distribution of the original source.
Otherwise, any automated processing of the original sources would fall under the same rights, and people would be allowed to “compile and decompile” source code, labeling it as their own work, and distributing the resulting source under their own copyright. Which is what would have happened here (assuming the code was decompiled that is), even if no binaries were built and distributed from it.
In short, you can’t just take a binary compiled from someone’s copyrighted work and thusly remove their copyright and license via decompilation. The compiler isn’t granted the right to remove the copyright… this goes back to the whole concept of “colors” seen by laws – this code file was a certain color, the resulting binary created from this code file was the same color, and the decompiled code created from the binary also becomes the same color – you can’t just legally change the color.
http://ansuz.sooke.bc.ca/entry/23
Edited 2010-11-15 05:42 UTC
However, the GPL doesn’t grant you the right to redistribute the decompiled version under a different copyright owner and license… which in this case would seem to be part of the problem Google is facing.
Apparently it is test code, used in the process of making Android, and it is not part of Android itself.
Google therefore didn’t redistribute the code at all. The point about “the decompiled version under a different copyright owner and license” becomes a non-issue. The GPL permits anyone to take code, study it (including decompile it), modify it however they please, and run it (or modified versions of it) for any purpose.
The GPL license for derived works is only a requirement if one re-distributes the code to downstream recipients.
Edited 2010-11-15 05:38 UTC
If they never distributed the code, then why was it found?
Because it is in the source tree of their developmnet process. It is apparently used by Google for testing purposes when they make Android.
It is only Android itself that Google distribute. When you buy an Android phone or tablet, you don’t get this derived work at all.
Therefore, Google don’t need a license for the derived work … all that they need (for their own internal purposes) is that they received the original work as GPL code. They may then use this code (internally) however they please.
AFAIK, this is the thrust of one of the legal defences that Google has announced. There is no valid complaint against what Google did with the code … everything they did with it, it is permitted for them to do.
Edited 2010-11-15 05:59 UTC
WHICH IS PUBLICLY DISTRIBUTED!
Jeez, you have such a strange view of things sometimes.
Well, the aim of copyright law is to prevent financial gain from someone else’s work. Google putting that code in a git repository does not invoke an act that is punishable by copyright law. The act which is an actionable finding of a breach of coyright law happens if Google sells, in exchange for money, Oracle’s code. Didn’t happen.
So, if the git repository is found to be redistribution under the intent of copyright law (debatable) … then Google will have to meet the conditions as required by copyright to redistribute it, and pay Oracle damages.
Oracle’s damages: zero (because there is no part of the sale of the Android phones that icludes this code).
Action required by Google: comply with GPL license conditions = make the code in the git repository licensed as GPL.
It is a strange view of things that Oracle is owed some percentage take from the sale of Android phones for something which is not in an Android phone. It just doesn’t work that way, sonny jim.
Edited 2010-11-15 06:22 UTC
© Fail.
Google putting that code in a git repository does not invoke an act that is punishable by copyright law.
Placing the files in an publicly open repository IS distributing it. Gaining something financially or not has nothing to do with it. They are distributing the file as long as it is publicly reachable on their repositories.
Of course it’s distributing. How can anyone argue otherwise? Very odd.
If it is distributing, and that is indeed the complaint, then all that Google needs to do is remove the files from their git repository, and carry on. Then they aren’t distributing it any longer, and it makes no difference at all to Android itself. Problem solved as far as GPL compliance goes.
How does this help Oracle?
Given that it doesn’t help Oracle one whit, and that it is terrible PR for Oracle, and that it is no trouble at all to Google either way (to leave the files on git, or to remove them), then why did Oracle sue in the first place?
It makes no sense.
The only way that it makes sense is if Oracle believe they will get a cut of Android.
Oracle’s problem lies in the fact that the files in question are not part of Android.
That is just an expensive blunder for no gain on Oracle’s part, then. Now that makes sense.
Edited 2010-11-15 10:26 UTC
Putting code in a repository accessible by others is clearly distributing. I think your “If” is unnecessary here.
The complaint (I think) is that said distribution was done not in accordance with the offered license, and thus is a copyright infringement.
If the judge or jury agrees. Or Google may well be liable for substantial damages, if the commercial value of the unlicensed file is determined to have been damaged or Google is determined to have acted recklessly. In Jacobsen v. Katzer, for example, the damages were $100,000, although the case was quite a bit more complex than our “simple” discussion here (is the law ever really simple?). It also involved patent counter-claims, cyber-squatting claims, abusing cute kittens claims, and as much confusing language as legal minds could render. Kind of like Oracle v. Google. 😀
Look at it this way. Say I put all 7 Harry Potter novels on gitorious. Ms. Rowling sues my butt. Do you really believe that if I simply remove the files, I have no further liability?
Or do you mean to argue that distributing in violation of the GPL can be cured by simply not distributing any more? But that’s incorrect, too – if you distribute in violation of the GPL, then you are violating *copyright*, not the GPL (because you are distributing without a license – the GPL thus doesn’t apply, because the GPL is a license, not a contract).
And you can most certainly win damages for copyright violation. Happens with distressing frequency.
Now I have no idea whether Google violated Oracle’s copyright, of course, and I’m not a lawyer, but having studied copyright as it relates to open source software for many years now, I’m fairly certain the above isn’t terribly inaccurate (can you tell I’ve dealt with lawyers for far too long 😀 ).
Or an obscenely large lump sum of money. Or they are in league with another company that makes competing smartphones that is having their butt kicked by Android (and really, who isn’t?), and are hoping to damage Android’s reputation, market share, or profitability for smartphone manufacturers. Or they have motives we haven’t discerned, because we don’t know their trade secrets. Or…
Whether the files are distributed on a smartphone is irrelevant in a copyright lawsuit. Simple distributing (if indeed that happened) is sufficient to cause problems for Google and potential income for Oracle.
A matter of opinion, but OK …
Fine. There are levels of this. Although it is by no means a given, let’s assume for the moment for the sake of discussion that there was indeed copyright infringement.
Oracle/Sun released the file as GPL. The GPL requires that if you re-distribute the code, modified or not, then you must make the source code available.
This Google did. The only possible non-compliance was that Google did not offer downstream recipients the same license as the code had when Google got it. Hence the only possible damage is to downstream recipients … not to Oracle.
Except for the crucial bits that you left out. The original code in Jacobsen v. Katzer was not licensed GPL, but merely open source. The problem here was that the recipient of the code tried to close it, charge downstream recipients, applied for patents for elements within the code that the recipient did not invent, and then tried to sue the original author as having violated the patents!
Oracle vs Google involves nothing of the kind. After all, remember that the GPL license requires that Google make the source code available, which is what Google did. Precisely the opposite of Jacobsen v. Katzer.
Ms. Rowling, the copyright owner, requires that you do NOT copy & redistribute her work without her permission.
To get their full permission, Sun/Oracle, as authors of the code at question in this case, requires that you DO make the source code publicly available if you do redistribute it.
Just a teeny tiny bit of difference in the details there, don’t you think? Hmmmm?
No. I mean that: if you distribute the code, the GPL requires that you also make available the source code to downstream recipients of derived works. Google did make available to downstream recipients the source code of the derived work. Where is the damage?
You have to establish that there was a copyright violation, and that there was damage. In the circumstances, given what Google have shown about the code, both of those are going to be very difficult for Oracle to prove.
It isn’t terribly inaccurate as a description for some kinds of copyright cases … such as Ms. Rowlings type of cases, for example. Unfortunately it misses the most fundamental of important details about the particular case of Oracle vs Google. The detail that Google complied with everything required in the GPL except putting a GPL notice in the header is one such detail. Perhaps Google will need to add such a notice to the header. OK, so? How does that help Oracle to such a degree that a lawsuit is worthwhile?
Google didn’t make any money from the files in question. Therefore, under any reasonable evaluation, damages = zero.
Sorry Oracle, but there is no pot of Android gold here for you.
The damage to reputations is most likely to happen to Oracle’s reputation, suing other companies over nothing.
Oracle don’t make smartphones. The code in question is GPL.
Google didn’t commit industrial espionage and send “agents” into Oracle offices and somehow “steal” trade secret code … which Oracle had released as open source! Get real! At least try to make some sense! Sheesh.
Au contraire, it is entirely relevant to the question of damages. Precisely on point for that question, actually.
Sigh! Maybe you will get it one day. Oracle released the code in question as open source. OK? Got it? Think about it for a second. Mull on it. Let the concept seep into your head.
Not only just open source, but under the GPL, which requires Google to make the source code available if Google re-distribute it.
Ponder on that for a while, and get back to me.
Edited 2010-11-15 12:57 UTC
Watch out your statements dude, cause if a propietary company decompiles GPL code and relicenses it like google and using the same distribution trick google is using then I bet you would go ape shit.
As long as the proprietary company released the source code of the derivative work (which is what Google’s git repository does), then the primary condition of the GPL has been met. The intent, but not the letter, of the original GPL license has been preserved.
BTW, there is no need to decompile anything … under the GPL, the original source code is available as plain text. Just copy it direct … why do you need to go to all the trouble of decompiling?
So, the ONLY problem is that Google did not license the derived work as GPL. Since Google ARE providing source code of the derived work, then the remedy is clear and simple … all that Google need to do to be 100% compliant with every requirement of the GPL is to license the derived work as GPL. That’s it. All damage fixed. Every intent met, evry I dotted, every T crossed.
This is (one of) googles primary counterclaims … that there is no actual claim made by Oracle. Oracle have not managed yet to say how they were damaged. What did this act by Google actually cost Oracle?
So far, in all of oracles legal submissions, the answer to that question is apparently … zip. Google’s actions regarding the code in question have cost Oracle nothing. Not one penny. Zero. Nada. Zilch.
According to Oracle’s own accounting.
So, if the court does find a breach of copyright, as it stands that will the damages amount awarded. Not one penny. Zero. Nada. Zilch.
Google will have to make the code in question licensed as GPL. Nothing else will need to change.
Got it yet?
Edited 2010-11-15 22:21 UTC
Hold on, what’s the problem here? What Google did is perfectly legal: There is nothing in the GPL that says you can’t decompile GPL’ed binaries or redistribute said decompiled code; it’s still licensed under the GPL.
I would also mention that putting code in the same source tree doesn’t make all of the code part of the same work. For example, in one of my projects I have rhino.jar and closure-compiler.jar in my build directory. Rhino is MPL/GPL’d and Closure Compiler is Apache Licensed but my project is New BSD licensed; the same goes for jQuery. Am I infringing on Mozilla and Google’s copyrights to their respective works? No. These tools are in my source tree for testing and minifying, nothing else.
I’m afraid you’ve completely missed the point (and with a remarkable number of words, too).
You see, the GPL grants the right to copy only if you conform to its terms. If you don’t conform to its terms (which is what Oracle is alleging about Google, I believe), then the GPL’s remaining provisions (about which you write with such passion and at such length) don’t actually matter a whit. They don’t apply.
I have no idea if Google screwed up or if they complied with the terms of the license just fine. But IF Google didn’t abide by the GPL, they’ve violated Oracle’s copyright and are liable for whatever remedies the court decrees (which is, of course, the relevance of Jacobsen v. Katzer despite the other differences you list).
It really is just that simple.
Indeed. Google made the source code available in a git repository.
If because of that they are somehow found guilty of copyright infringement, then indeed they may be forced to comply with the terms of the license for the code.
Google might have to comply with the GPL. Google might have to make the source code available to the public. Google might, for example, be forced to make the source code available in a git repository …
Oh wait. Do I perchance spot a teeny tiny hole in your utterly circular logic there … Hmmmm?
I have to say it … it is you who have utterly missed the point.
The entire effort of Oracles lawsuit will amount to Google changing the license of code it already makes available from one open source license to another.
How does that help Oracle in any way? Why would this outcome be worth Oracle’s time and expense? For that matter, how does it hurt Google?
Sheesh!
Edited 2010-11-16 13:03 UTC
They distributed it:
http://android.git.kernel.org/?p=platform/libcore.git;a=blob;f=supp…
That’s why everyone knows that it looks decompiled.
If Oracle can mount an argument that having it in a git repository is redistribution, then Oracle do have a case that this code is derived from GPL code, and hence the derived work should itself be licesnsed by Google as GPL (since Oracle have convinced the court that it has been distributed to downstream recipients).
Fair enough (as long as Oracle can get the court to agree that Google re-distributed it, and not the Open Handset Alliance).
Financial damage to Oracle: zero.
Action required by Google: Amend the headers to credit Sun(Oracle) as co-authors of the derived work. Re-license that derived work as GPL. (Google are already providing the source code, so no corrective action is required there).
Court Costs: awarded to Oracle.
Next case.
Edited 2010-11-15 06:10 UTC
I agree – this would be the proper resolution (or simply, remove the file entirely, demonstrating that it is no longer distributed, and not used in any derivative work), and *should* solve the problem morally for all parties.
However, once copyright infringement has been proven to occur, Oracle could continue pursuing monetary damages… I’m not sure how they could ever quantify any in this case, but who knows what they have up their sleeves
Edited 2010-11-15 06:13 UTC
I think being on the wider Internet constitutes distribution, we accept this definition at any other time and now should be no different.
I can’t predict at all the trial outcome, but if they were to find for Oracle on this point, it would damage Google’s reputation.
I thought being on the wider Internet constituted “making available” not “distributing”. Am I wrong?
That would be a question for courts to decide, I suppose.
It would also be a very simple matter, if it is decided that being on the wider Internet constitutes distribution, for Google to remove the particular set of files from git, would it not?
So, if indeed the court determines that “being on the wider Internet constitutes distribution”, then the simple remedy is for Google to pull the files, and simply carry on with Android as they were.
How would that help Oracle? Why would Oracle sue Google just to achieve that end?
My take is that Oracle are after a slice of the take of Android. Since the files in question are not distributed with Android, my guess is that Oracle won’t have a show at achieving that aim.
So the whole thing boils down to a huge WOFTAM in the end, IMO. Oracle makes an enemy of Google, and most of the open source followers, for absolutely no gain.
Once again, I am reminded of my very favourite quote: “Against stupidity, the Gods themselves contend in vain”.
It’s the same thing.
Yeah…. 300 lines of Perl, C, or Python would be copyright infringement. 300 lines of Java and you barely have Hello World.
300 lines? You have got to be kidding!
http://en.wikipedia.org/wiki/Derivative_work
300 lines simply isn’t going to be a major element of anything these days. You would need more like 30,000 or even 300,000 copied lines to make a copyright case in this day and age.
300 lines simply isn’t going to be a major element of anything these days. You would need more like 30,000 or even 300,000 copied lines to make a copyright case in this day and age.
Lemur, you are so full of sh*t that it’s imbelievable. Copyrighted work is copyrighted work, regardless of the size of the work in question. Otherwise all companies would need to do is just copy all files with less than 30,000 lines in them from any open-source product they’d want and relicense them under a proprietary license and claim them as theirs. Hell, I could just copy all files with less than or equal to 30,000 lines in them from LibreOffice, claim they are all mine and relicense them under a closed license, and start distributing it all as mine.
Are you REALLY certain you want to go there?
But not everything is copyrightable. From what I’ve seen this is merely declarations in header files, not code (or am I mistaken?). Declarations can be considered as facts, and thus not copyrightable (also, was there ever a final ruling on public api in the SCO case?).
Things that are not copyrightable:
titles, names, short phrases, and slogans; familiar symbols or designs; mere variations of typographic ornamentation, lettering, or coloring; mere listings of ingredients or contents.
In my opinion either this copyright infringement is a red herring or there has to be more, I can’t see them actually going to court over this, it seems far too weak in my (admittedly IANAL) eyes.
Copyright law is copyright law. The size of the copying (relative to the whole of the work) is a crucial element of the law. In order to infringe copyright, one must have copied and then re-sold a “major element” of an earlier work.
Now, if the earlier work was 600 lines, then 300 lines was copied into a derived work of, say, 900 lines, then clearly this is a major element copied. The resulting 900 lines is clearly a derived work, under the definition of copyright law. No problem.
Hower, if the original work was 300,000 lines, and what was copied into another work was only 300 lines, then a perfectly valid legal defense arises that the second work is not a derivative of the earlier work, under the definitions of copyright law.
This is not just me saying it, this is the law of the land.
That is why I can tell you that as far as understanding of copyright goes, this is the best of times and this is the worst of times … and I would have no fear of being liable under copyright law even though I had used those particular words.
The extent of the copying (compared to the whole of the work) does matter. Very much so.
Edited 2010-11-15 22:22 UTC
Hah.
Hmm… I really can’t decide whether to mod you Funny or Troll.
At 300 lines, you’re barely past the import statements.
Import Java.Sun.NonOrthogonality
Import Java.Sun.OverEngineering
And so on.
How come that every poster so far has been oblivious to this bit of the news item:
I think this (if true of course) casts some serious doubts about the legitimacy of Oracle’s copyright violation claim. I for one call it “twisting reality to suit your agenda” (TM)(C). I guess people have some judgmental ability left when they hit the courthouse?
If this is true then the copyright infringement is dead in the water. IIRC this is what SCO did, removed/edited certain parts from the originals in order to make the alleged copies look more similar. Either way this copyright case seems very weak as it is, something tells me this is just for show and that the real meat will be the patent dispute.
Quite so. Even more to the point is that the requirements of the original license have been largely met. Even if the work is held to be a derived work under copyright laws (very debatable), then Google have published the source code of the derived work, as required by its GPL license.
So what did Google do wrong that caused any damage to Oracle? Oracle haven’t identified any damage yet, and that is a serious problem for Oracle’s case.
I don’t believe there is any meat at all.
The real intent and purpose of the Oracle lawsuit might best be found in this speculation:
http://www.entirelyopensource.com/Blog-and-Opinion/The-Apple-Oracle…
The only problem I suppose is that if Oracle are not able to state any actual valid claims against Google, then it won’t even go to court.
Edited 2010-11-16 01:30 UTC