Oracle wants another go at Google over Android Java copyrights Oracle has asked a US Federal Circuit Appeals Court to overturn an earlier decision in the database giant's billion-dollar intellectual property lawsuit against Google, with an audacious brief that compares Google's use of Java in its Android smartphone OS to an author selling counterfeit Harry Potter books. Oracle's appeal …
fwiw, decades ago doing 'archives' for some product, we would have to print out the first 100 pages of code, then have that in the archive. The understanding was that this was what you needed in order to 'prove' the copyright..
now obviously this is a while ago..
but it was far far from 'just the headings'. So god knows what Oracle think they're playing at.
This post has been deleted by its author
There have been various cases that have shown that basic function prototypes and class declarations are not copyrightable. The ruling have been that headers are essentially functional - the specification is not distinguishable from the expression. How many ways are there to specify a function of a given name that returns an int and takes two ints as parameters?
USL v Berkeley and SCO v IBM come to mind. Both came to the same conclusion.
Not to defend Oracle (yuck) but I could see a grey area in some template meta programming where you use the prototypes and declarations to actually generate the implementation for you. Of course that doesn't really apply in this case but just the first thing that came to mind.
Should say at least in C++ the implementation obviously only gets generated by the compiler (compiler generates the code itself) once the API is actually called (and also obviously can generate multiple implementations depending if called with different types in different ways) just to be clear. Not an exercise to defend Oracle at all but more just a curious what if.
A better example would have been the lampoonings Harry Potter.
Barry Trotter and the Shameless Parody. It came out 2 years after the first harry potter book, and satirizes it. It doesn't infringe of copyright even though it uses most of the same elements with a few tweaks (not barry trotter, not harry etc)
In which case Oracles own example comes back and bites them in the arse.
That's not true. In the uk http://en.wikipedia.org/wiki/Fair_dealing_in_United_Kingdom_law
As the Harry Potter author is english I'd have thought she would have tried to sue under uk copy-write law if she could. However the analogy stands as none of the text is copied and tapping into a fan base of another author is legal if the elements are not trademarked/trademarkable.
If she wanted to contest under trademark law I'm sure she would have done so by now, so there might be a reason that this is not do-able too. (changing harry potter to barry trotter and keeping the character roughly similar is not enough of a defense).
"adding that "just about every smartphone carrier" at the time used Sun's Java Mobile Edition (ME) – a description of the market that conveniently ignores Apple and the iPhone."
Except the iPhone and the Android OS were developed almost alongside eachother, time-wise, so at the point Java was being stolen for Android, the iPhone either hadn't been launched, or was just being launched. Therefore, "just about every smartphone carrier at the time used Sun's Java Mobile Edition (ME)".
I believe that was an S80 or S90 device. S60 definitely had Java ME.
@Tom 35: Smartphone platforms of the day were BlackBerry, Symbian, and last and most definitely least Windows Mobile. True, the last two had their own operating systems written in other languages but both also had a Java VM and ran Java apps, which is the same set up as present-day Android based on Linux running Dalvik apps now.
This post has been deleted by its author
Starting by the fact that "tapping into the fanbase" of Harry Potter is in itself copyright infringement, since the character of Harry Potter is protected by copyright; while even Oracle had to admit that the Java language itself is not protected by copyright.
I believe you can pretty much do what you want in a court case without it infringing copyright, trademarks or patents.
If you couldn't then a situation could arise whereby you could not defend yourself because your defence materials infringed copyrights, trademarks or a patent.
It's nice to see that it's not just commentards on the interwebs who make tortured and piss-poor analogies in order to try to discuss copyright and intellectual property. If I read one more argument about how downloading music for free is or isn't like a car, I'm going to start hunting people down and using their goolies for golf balls...
But it's nice to see Oracle's highly-paid lawyers can do it in a court case too. Would 'Demonic Dalvik and the Virtual Machine' make a good book title?
I'd call it darn fine lawyering, getting the client to spring for briefing a wackily poor analogy to the next higher court.
I had some sympathy for the point of view that Google, in exploiting java's good will in the developer world, should have compensated Sun for its many years of building the community by giving away sdks, documentation, frameworks and tool.
Any way, Oracle decided to try and claim apis as copyrightable, but this would be such a large ip grab. I think, without being alarmist, that no one could implement a protocol except they license it from the originator, even if the implementation is entirely different, were Oracle's argument prevail.
Or maybe I miss the brilliant truth that boolean Object.equals(Object o) is merely a unlicensed character in a little drama I call My Program: Does It Work?
Isn't that how Exchange ActiveSync works? (You have to license it from MS, how you implement it is up to you.)
Software patents. See here: http://www.microsoft.com/en-us/news/press/2009/feb09/02-09statement.aspx
I suspect that Microsoft already tried to enforce copyrights on APIs back in the bad old days and got slapped down for their troubles (see also: Samba) but my memory is hazy on that matter.
In the original trail Google had the advantage that Alsup actually spent time learning to code and to understood code, he would never have bought this dumb down crap from Oracle and probably would have told them that in court. I suspect Oracle is hoping that the appeal judge they get is a computer illustrate fool that will buy their arguments. They could be right, there are not many judges in the US that already know how to code, and even few that will take the time to learn to code and understand code before issuing a judgement.
>I suspect Oracle is hoping that the appeal judge they get is a computer illustrate fool that will buy their arguments
Illiterate? I'm not judging- I've found myself, in the last couple of years, typing similar-sounding words in place of the ones I actually mean.
I must admit, the article left me feeling computer-illiterate. All I know is that in my head, Java = annoying for its insistence on being updated every fortnight and attempting to install some shoddy browser 'toolbar'. If there is more to it than that, Oracle have not encouraged me to find out more.
Perhaps google could counter with.
An API is much like a toolbox. Now, all these tools were created many many years ago, long before the Java brand tools came along, with a JSpanner. It's exactly the same as a spanner, except the name is different, and it's kept in a different place in the toolbox. As has been proven, their trademarks and patents did not cover the names of items in the toolbox, nor did it cover the box itself.
Google, seeing that this toolbox was good, took elements of it. The tools go into the same places, and they bear similar names. GSpanner for example. The big difference is the JSpanner was made out of hardened plastic, while the GSpanner was made out of cast iron.
Both are based on the Spanner of old, both have similar names, and both go into the same slot in the toolbox, but both are made in completely different ways even though they function the same.
Those 'you wouldn't steal a car' 'downloading is theft' adverts make me want to steal a car from them...
I wonder how they get those notifications past the ASA? I am very tempted after the next time I visit the cinema and watch one of those, to email the ASA complaining about the advert misleading the public as they are lying..
Copyright Violation is not always a crime, it is a civil offence most of the time if anything....
Yes, quite right, companies shouldn't be allowed to make a profit from their customers, to do with as they please... you know, like perhaps using it to defend their own software from what they believe is theft/copyright infringement/whatever. All profit they make should just be churned back into the software, to benefit you, the customer.
Maybe Oracle should dictate to you how to spend any money you didn't spend on their software...
Source Code != Symphony.
They both have the creative aspects, but the former is more of a "tool" to "do a thing" and that thing often boils down to basic math, or the choice between a couple of possible paths. It also can directly affect your privacy and security. It is manifestly not a symphony. Or even the sheet music for said symphony.
Source Code != Novel
For almost the same reasons as above. One is a work of art, the other is a tool (no matter how well crafted) to do a job. End of. About the only place we can enter into a grey area is where the source is laid out as a poem, or in a visually pleasing structure. But that's really just formatting - it doesn't usually affect what the code does.
I rather doubt Google had to "steal". It's not like they are technically inept, this is the company that dropped Windows and made their own GNU/Linux spin. Are Google as pure as the driven snow? No, I just don't see enough here to make me think they stole anything. They made the API (give or take) compatible with JavaME, everything else after that was almost inevitable.
What Oracle is pissed off about is that is dropped the ball. JavaME was the pre-eminent platform for mobile computing and was treated like the bastard child. Oracle ignored it, Google made a look-a-like and now Oracle is kicking itself.
To be honest the best way to resolve this would be to place the executives from both companies in a large sack and invite members of the public to beat said sack with a stick ($5 for 5 hits). No matter who you hit, you will always hit the right person.
This feels a lot like when Microsoft got into trouble copying other companies then passing it off as their own works - I think it was over file formats for Word, will have to go link-hunting...
The point is that the nail that sticks up is always hammered down eventually. Microsoft stuck up and is now slowly being hammered down. Combine this with missed opportunities and questionable decisions, and there's space for opportunists, and in this situation, the one who can get to the right solution fastest will win. "The right solution" is not necessarily predictable, but "fastest" in the corporate world virtually always involves cutting corners and treading on toes, hence creating targets - and so the wheel turns.
In fact, I'd be willing to bet that within the next 5-10 years, we will be getting anti-Google jihadists posting bile all over these forums as Google themselves try desperately to hold onto their position, just like currently happens for Microsoft.
Wasn't that some kind of XML thing? And wasn't it a software patent (spit) they infringed?
As for the anti-Googlers - we already exist. I do what I can to avoid the Demon of Chocolate Mountain. Same as I try to avoid the Beast of Redmond and the Curse of Cupertino.
(And I made a very, very bad choice of phone)
Oracle had the bad luck to come in front of a judge who seemed to semi-computer literate, and could even write simple code.
However the American legal system is such that if you don't like the judgement you can keep paying your money and throwing the dice in the hope you get the result you want. This will go on for years, but you can't help wondering whether they would be better off spending there money on developers rather than lawyers
"However the American legal system is such that if you don't like the judgement you can keep paying your money and throwing the dice in the hope you get the result you want. "
Depends on which court you started with, as each court you appeal to has to be a higher court. Since this case started in the Federal Circuit Court, it had to be appealed in Federal Appeals Court. If you want to appeal the Federal Appeals Court verdict, you have to petition the U.S. Supreme Court and hope they choose to hear your case (most of the time, they'll refuse).
And, if you do manage to get before the U.S. Supreme Court and you lose there, you have no one left to appeal to in the U.S. legal system. Although you could try to convince the legislative branch to change the law...
I think it's closer to a recipe book. There are only so many ways you can make an apple pie and although you wouldn't steal the exact wording from another book (unless you're an idiot) there's nothing wrong with reading a book, adding a bit of cinnamon, juggling the quantities, describing the pastry rolling out a bit better but in your own words, etc.
What Google copied was the necessary parts of the process. What Oracle are claiming is that only they can tell you how to cook an apple pie, and every apple pie recipe is somehow based on theirs, even when someone has entered into an "apple pie making industry". Collect recipe books and, pretty much, they will all have the same recipes in them, and the recipes will be very similar. The expression is in the details, in the actual description, and even in the collection and order of the recipes inside the book. It's NOT in the process of how to make an apple pie that is edible and recognisable to others as an apple pie.
Oracle really scrape the barrel, like SCO did. No sensible lawyer in the world would try making that argument after the judge ruled the code uncopyrightable - that's almost a direct attempt to say "we still think it's copyrightable" by even making the comparison, without actually expressing that directly. The judge isn't going to like that, and the more publicity the analogy gets, the more damage done to Google on a false analogy.
Makes you wonder who's behind it really, again. Just as the SCO thing gets settled (by driving themselves into bankruptcy deliberately to pursue a lawsuit totally without merit), suddenly there's a new threat to the copyrightability of code, on a Linux-based operating system, using open technologies, that compete in areas that a certain company is trying to push into.
How hard is it? You can't copyright a process. You can only copyright an expression of that process. If someone copies a Haynes manual, they would have grounds to sue them. But if someone else writes a book about the correct way to dismantle a Ford Mondeo, which uses none of their pictures and none of their text but might have similar (trivial) chapter headings ("The Exhaust System", etc.) then they don't.
The judge already decided this issue in terms of what's REQUIRED for interoperability (the "chapter headings"), what's uncopyrightable (the "chapter headings", and the trivial bits that are necessary to make them work properly which in code terms means quite a lot of code that has no other sensible way they can be written - e.g. returning certain structures means filling those structures, etc.), and whether Google did anything wrong. Annoying them by putting out a false analogy that suggests they were wrong, without formally challenging his individual points, is only going to end in tears for Oracle.
Damn, I'm glad they shoved OpenOffice out of their way because they didn't want to touch it. Probably the only saving grave to come out of Oracle in decades.
@Lee D: "I think it's closer to a recipe book. There are only so many ways you can make an apple pie and although you wouldn't steal the exact wording from another book (unless you're an idiot) there's nothing wrong with reading a book, adding a bit of cinnamon, juggling the quantities, describing the pastry rolling out a bit better but in your own words, etc.
What Google copied was the necessary parts of the process."
The thing is, I'm not entirely sure that's true. Whilst there are only a few ways of defining an individual function that is performing a very specific task, there are an enormous number of different ways an entire API suite could be designed. There really is no technical reason that Google had to present the exact same API calls as Java, other than to conveniently be able to "borrow" the toolchains being used for Java development without much hassle and to more easily persuade Java developers to jump ship.
I don't like the idea of Oracle winning this one (and I doubt they will), but at the same time it doesn't seem entirely right that all the effort that went into designing a good API (which is certainly a difficult thing to do) can be so easily swiped by a competitor without any compensation whatsoever. It makes the job of anyone who provides third-party software components rather uncomfortably fragile.
"There really is no technical reason that Google had to present the exact same API calls as Java"
Other than to make something that you can compile Java programs too and is compatible with Java, you mean? Sure they could have done it with Google Go but that wasn't the point. The point is that the LAW requires independent recreations that have been made without reference to the original to literally copy parts of the API where it is NECESSARY for interoperability. And the court said that that is all the Google did (apart from a minor copy/paste exercise that covered something like 9 nines of very obvious code - out of millions - by mistake and which Google didn't try to defend) - they just copied the bits that NEEDED to be called that in order for Java programs to compile and work as expected.
Just because you own Java, doesn't mean you can stop anyone else making a competing and compatible product. If it did, Samba and Wine and ReactOS would have died decades ago.
I believe publishers of maps and log tables sometimes introduce small deliberate mistakes in their work so that they can recognise when others copy them.
Just like how teachers can spot pupils copying each others homework.
So if only we could identify any bugs/flaws in Java then we could easily prove if Google copied any of Sun/Oracles bugs into Android!
According to Frank Castle (M. I. Mech. E.) 's log tables, 2 * 2 = 3.999. I'm not sure that error is deliberate, though; just a simple rounding error.
The existence of deliberate errors in maps has been proposed as a plausible explanation for some of the "SatNav SNAFU" stories that used to crop up every fortnight or so. It certainly makes sense. I wonder if a bridge has ever collapsed because the designer ran afoul of a deliberate error in a book of tables? How far can the publisher disclaim responsibility, knowing that they introduced deliberate errors?
The law is quite clear on the matter: Anyone else is allowed to make a product that can be used as a drop-in replacement for some product that you make. It's called "fair competition". You can't use copyright, patents or trademarks in such a way as to prevent this.
If the only way for them to make their thing do what it's supposed to do properly, then such bits of your work as they absolutely need to copy in order to make it work -- but no more -- are excluded from the scope of protection. For instance, if you use a special connector for the power supply that goes with your widget, other people are allowed to copy that connector closely enough for their after-market power supply to be able to make proper contact with your widget. If your games console hardware is checking for a signature block on a storage medium (such as a ROM cartridge or optical disc), then that signature block must be excluded from copyright in the interest of interoperability. Otherwise, you would be able to lock competitors out of the market, which is very Not Allowed.
Maybe Google should insist for Oracle to show them how they can make their own compatible Java replacement (which the law explicitly allows them to do: blocking interoperability is anti-competitive behaviour of the most blatant kind, and Oracle are now entering the realms of deliberately obstructive behaviour) without actually copying any Oracle code.
Sun never intended Java to be a single-vendor system. They recognised, correctly, that developers are reluctant to commit to a new environment that is someone's exclusive property. So from the word go there was a process to allow others to develop complete Java implementations. These had to be compatible at the bytecode level at runtime, so that Java could interoperate at the binary level, and pass an extensive compatibility test suite. Once complete, you'd get a licence from Sun to call your product Java(TM). 10 years ago, I was working on a JVM at a company that did exactly that.
MS also took a licence, and then released MS Java. MS Java included additional methods in the standard classes, methods that the documentation did not even identify as not being part of standard Java. The clear intention was that developers would be suckered into writing Java that was not portable beyond the Microsoft system. This was a naked breach of the terms of the Sun licence, and it was on that basis that Sun sued.
Android uses Java the language, and provides compatible versions of some of the standard Java classes. But it does not use the Java runtime bytecode; instead, it uses a completely different bytecode and VM, Dalvik. You can't plonk a standard Java binary onto an Android device and expect it to work. The important point here is that Google do not claim to be shipping Java(TM), believe that using Java-the-language is perfectly permissible, and have no Sun licence.
So the legal matters at issue are completely different.
IANAL. But if the original vendor of a language has the power to stop others doing compatible implementations, or demand royalties, as long as no trademarks or patents are violated, it's news to me. And as others have pointed out, if this were to turn out to be the case, m'learned friends at IBM, where SQL originated, will doubtless want a very, very expensive word with Oracle.
"IANAL. But if the original vendor of a language has the power to stop others doing compatible implementations, or demand royalties, as long as no trademarks or patents are violated, it's news to me."
The tricky part is saying what language you're using, and want developers to use, for your VM, as the name of the language probably is trademarked. Indeed, IIRC, it is this exact point that led to Sun developing Java instead of using Smalltalk as Xerox was asking too much for the license (seriously, is there any example of Xerox's computing work that wasn't screwed up by a complete lack of commercial sense?).
There's legal precedent that anyone can merely re-implement a programming language from scratch -- look up FoxPro.
Before Java was Open Sourced, Microsoft licenced it from Sun under quite reasonable terms. They then breached the terms of the licence quite spectacularly. Specifically, you are not allowed to call a software product "Java" (which is a registered trademark, and so requires a licence to use) if it does not pass a series of compatibility tests. (The actual compatibility tests were always Sun's proprietary secret, in order to prevent vendors from gaming the tests.) Microsoft's "Java" implementation failed compatibility (on purpose: Microsoft intended for their "Java" to do things that Sun's Java couldn't, as a way of persuading users to choose Microsoft Java over Sun's Java. This was exactly what Sun's licencing terms and compatibility-testing régime were intended to prevent), and so they lost the right to call it "Java". When they persisted in doing so anyway, Sun took them to court.
SQL is an ISO standard, so you can't be sued for implementing it. Java isn't, because Sun didn't want to play ball with others and insisted on their need to retain control.
Google could've avoided all this by going with the Microsoft CLR and C#, which are also ISO standards. You kind of have to wonder why they didn't (other than it might have damaged their "cool" factor amongst anti-MS geeks)
Leaving aside all proper legal process, I trust Oracle even less than Google and I think that consumer interest lies with prizing Java from the cold dead hands of the Oracle's corpse*. If Java was open source, it would have been forked just as MySQL spawned MariaDB. In this light, Google is playing hero.
In the unlikely event that I was appointed judge, jury and executioner**, I'd decide that Google must pay Oracle a percentage of the revenue they take from selling Android to hardware manufacturers. :)
* Fantasy
** Pure fantasy.
So according to the insane story of Ann Droid, Oracle is basically trying to say that they were going to release an open source mobile operating system?
Yeah right.
What planet is Larry on? Does he actually realise that Google dont sell Android for a profit? Ok they make money from Android by other means, but that's besides the point.
If I were a judge, I'd read that story and tell Larry to go back and fix his fundamentally broken Java client before he starts throwing stones, before all the big corps sue for the damage the recent security fuck up caused!
What I completely fail to understand is how Oracle manages to come to these conclusions even though Java is basically an open source project, you can see so for yourself at its Java.net download page.
When you grab it you'll immediately notice that you don't have to comply with any license or agreement before getting your hands on the code. And when you finally got the tarball and check its contents you'll notice the main license being the GPL. Now, I know GPL doesn't involve itself with copyrights, those will always remain with the original author, but just how far can you take that?
When I take a GPL project I can fork it; which means so much as using all the code one on one and adding my own code to it. How would copyright come into play there? Sure; the original author has a copyright on his code; but that doesn't mean I can't use it or even sell it.
Another thing... Oracle claims to know all that much about their licenses but it also seems they totally ignore whatever doesn't interest them. When you check Java SE's download page you'll notice a link to the Java Research License.
I quote:
"Sun is supporting the JRL for most Java technologies it releases through the Java Community Process as well as research projects surrounding this code. "
"...Sun requires a click-through license. "
"While not every Sun Java technology has been released under the JRL, many have. Please contact JRLFeedback@sun.com if you have any questions. "
"10. Am I required to purchase a support contract from Sun under this license? "
Not only does that license date from the Sun days, it would also appear as if Oracle never bothered to change or update it, which seems strange. And although this license does state that "COMMERCIAL USE AND DISTRIBUTION OF TECHNOLOGY AND MODIFICATIONS IS PERMITTED ONLY UNDER A SUN COMMERCIAL LICENSE." this is obviously superseded by the inclusion of the GPL in the actual sourcecode tarball.
So quite frankly I can't help think that its kinda obvious that this seems very much like a wild goose chase.
""just about every smartphone carrier" at the time used Sun's Java Mobile Edition (ME) – a description of the market that conveniently ignores Apple and the iPhone."
I'll bite.
What year was it "at that time"? If it was more than say 3-4 years ago, then the quote was probably correct and it ignored the iPhone because the iPhone presence was, in fact,irrelevant.
The analogy presented by Oracle is, as one might expect, completely meaningless. Comparing the copying of a (portion of a) copyrighted material is not the same as copying the definition of an API method.
The chapter titles in the analogy presented are in fact part of the material as a whole, and would themselves be copyrighted, I'm sure. Just because O'Reilly, for example, makes the contents of books visible prior to purchase does not automatically entitle you to duplicate them, nor give you any rights over the content.
Public APIs on the other hand do not work this way. They're designed for the purposes of inter-operability by 3rd-party content to the underlying application (there's the start of a clue in the name - Programming Interface. Not code library.) Presumably by trying to claim such APIs as copyrighted (and therefore as private APIs), Oracle is also making a statement about their objection to permitting inter-operability.
I can't help but feel that this is unwinnable for Oracle (as their prior court loss should have suggested to them): If they're intent on their claim, then I'm sure this is something for either the Competition Commission, European Commission or DoJ to investigate further as this would surely give them an unacceptable monopoly in this field?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
