Java-aaaargh! Google faces $9bn copyright bill after Oracle scores 'fair use' court appeal win The US Court of Appeals for the Federal Circuit in Washington DC has revived Oracle's bid to bill Google for billions over its use of copyrighted Java APIs in its Android mobile operating system. On Tuesday, the appeals court reversed a 2016 jury finding of fair use that deemed Google's actions acceptable, and sent the case …
The CAFC is fscking insane. This decision is giving me a migraine just trying to figure out what the implications for software, system, and hardware engineering. What's transformative, or no, is another can of worms. What I do know is that my previous work is "technically" not infringing as I engineered my code from basic principles of the disciplines involved. No cut and paste there. On the other hand, proving it would be a problem.
Umm..., has anyone given thought on how this plays out on Stack Overload let alone GitHub and every other coding, especially sharing code? Thought not. Frag.
First off, IANAL. I did a walkabout at other sites to gain some clarity on this. If you have an API with a defined interface, it doesn't matter one whit that the underlying implementation differs from the copyrighted API's descriptions. Clean room implementation doesn't mean squat in legal terms. Since copyright is simply assumed on creation of the IP, we have a serious problem. Frankly, were IBM to be vindictive, everyone that was involved in the whole IBM PC clone process way back in the 1980's is a valid target.
What is not immediately clear is exactly how much correspondence there has to be in the header files between two different implemenations. What if I keep the function name, number of parameters the same between implementations but alter the variables names passed. That's a good Fair Use question right there. {Shrug?}
On further review, anyone using my API's from when I was serving in the military has a problem as I was instructed to assert copyright. [Something that was changed after I was out on disability.] I'm not the least bit interested in a lawsuit asking for damages. I wonder how many others feel this way?
IAANAL. However, it is my recollection from the late and much lamented Groklaw that changing the names, in a software copyright case is immaterial; that a sufficiently close "structure, sequence, and organization" match of the code samples determine infringement. It is not clear whether sorting the API entry points in the include file would get you any further from infringement.
This cannot be anything but very bad if it is allowed to stand.
The CAFC is fscking insane. This decision is giving me a migraine just trying to figure out what the implications for software, system, and hardware engineering.
I can tell you what is the decision outright - it is more lawyers.
We now have officially a precedent where a qualitative opinion by a lawyer is required in any case related to copyright infringement on software. This is as subjective as subjective gets, as idiotic as only the USA judicial system can produce it and to top it all it makes any software legal issues strictly a matter of "the best judicial system money can buy".
Frankly, if you are writing software and working to an API - I suggest you move to the other side of the pond and NOT to the UK as it will probably be made an IPR vassal state of the USA as soon as 2020.
"Your parents may wish to sue. Did they grant you a licence?"
Well my mother must have granted me a licence to replicate my DNA while in vitro, she seemed rather keen to help with that entire process. Not like I could do anything if she chose to not grant me that licence at the time. From what my parents told me, they where both very keen. In the 57 years since, neither parent has tried to revoke that licence, and one of them is still alive. She knows where to find me if she changes her mind.
If Micky Mouse keeps extending the length of copyright, all three of us will likely be long gone before the copyright expires. Once mum dies I'm not sure if the copyright passes to me, or either of my siblings, unless specifically written into her will. If by law it's a three way split, I'm sure we'll negotiate a settlement where we each have full rights to our own DNA. We are reasonable like that, unlike Oracle.
"I think I hold a copyright on my DNA, or maybe my parents..."
Don't make any assumptions ....... If you don't have it in writing then it does not exist !!!
Behind your back maybe Google or Facebook own it !!!
One can never be sure the way the world is going !!! :) :)
I would start filling out the form now ...... remember your coffers are not likely to be as deep as 'The Googleplex" and their 'Friends' :) :)
Best of Luck,
(SoylentG-74868774863-837637-7428487BEEF-UK) aka S. Roth.
"No, copyright actually doesn't require a written statement or registration."
It does not strictly require it as copyright is applicable on creation of the 'new' work BUT there are registration services such as UKCS that you can register your copyright with for 'further' protection. See https://www.copyrightservice.co.uk/services
As proving who has copyright can often be 'difficult' such registration is useful to prove via an independant organisation that you have copyright and applied first for registration of such copyright.
Such deconstruction does somewhat ruin the 'joke' but with the like of 'The Googleplex' and others default copyright is minimal protection when US of A courts seem to be driven by how much money you have NOT whether the result is just !!! :)
I don't think you do own the copyright to your DNA. You haven't published it: these days you can register your 'publication' of the original work.
Key consideration is to ensure that blood & other samples you give for healthcare reasons are accompanied by a form that you make *them* sign that they cannot use your samples for any form of Intellectual Property assertion or claim.
actually, given the behaviour of the Vote Leave alleged in Parliament yesterday, the original advisory referendum ought now be ruled null and void *
* = unless such Putinesque behavior is now the Legal standard for the conduct of UK ballots.
>OK for oracle to keep appealing until eventually they get lucky and find a Google hating judge.
>not OK for second referendum, the people have spoken..
The idea of appealing up a court decision is that you reach a more competent court.
Are you suggesting there is a group more morally or technically qualified than "the electorate" to make the Brexit decision? Perhaps an elite who think just like you? Are you are thinking of forming "The Central Committee"? We can't really have the electorate making such important decisions can we?
"The idea of appealing up a court decision is that you reach a more competent court"
In this instance, I hope that Google does take this case right up to the Supreme Court if necessary given the adverse ramifications for developers if this most recent ruling is allowed to stand.
These days, Oracle is pretty much the template for Evilcorp given that it's trying to use litigation to extract money a la patent trolls instead of innovating and because it's so anti open source.
In summary:
OK for oracle to keep appealing until eventually they get lucky and find a Google hating judge.
not OK for second referendum, the people have spoken..
Pray tell us what these things should have in common.... As I for One am failing to connect the Dots.
And as to the People "Speaking", I thought an albeitly very small majority already thankfully have.
>Refer to Monsanto.
There's a huge difference between patent and copyright, one is 20 years the other is death plus 70, I wonder why corporations are pushing for the expansion of copyright into the patent field ?
It was our 18 year old lab assistant that copyrighted it and we are going to make sure he lives until 95 even if they snuff it before then, wink.
"hence (c) Sony/Disney etc."
... though from something a read years ago i think this may not be a valid copyright claim as the copyright symbol is a c in a circle and not "(c)" - and I seem to recall someone managed to argue that using "(c)" without the word copyright meant it was an invalid claim of copyright and the fact that an attempt had been made to expliclitly claim copyright meant the implicit Berne convention copyright didn't hold either.
Personally I'm still trying to get over the fact that an API can apparently be copyrighted.
So the means of interoperating with something, which was specifically and exclusively designed for the purpose of interoperation, is now apparently restricted by law to prohibit interoperation?
More "IP" lunacy.
This post has been deleted by its author
>Personally I'm still trying to get over the fact that an API can apparently be copyrighted.
The EU has some sanity by explicitly saying it cannot be copyrighted:
https://arstechnica.com/tech-policy/2012/05/eus-top-court-apis-cant-be-copyrighted-would-monopolise-ideas/
Sorry Reg, couldn't find a Reg link story.
The EU has some sanity by explicitly saying it cannot be copyrighted:https://arstechnica.com/tech-policy/2012/05/eus-top-court-apis-cant-be-copyrighted-would-monopolise-ideas/
Ooooh - another area of EU/USA courts disagreeing, time to order more popcorn!
I hope that the EU view prevails.
It doesn't mean that at all, it just means that you can't go copying someone's API. as Google did in this case.
You can still write code to make use of someone else's API, but whether or not you can go and re-implement their API depends on the API copyright holder and their terms of use, as it should be.
Google knew all along that this was the likely outcome, they were warned internally by staff that they should just come to a licensing agreement in order to use Java. But Google didn't get where they are by paying anyone when they could just rip them off instead.
AT&T (C) can sue Stroustup (C++) because C++ has printf. Who can sue Oracle (Java) because Java has printf. Who can sue Walter Bright (D) as D has printf too.
Now that van Rossum (Python) thought he could rename it to str.format() and get away with it, as did Mozilla (Rust) with println, but the thing is they both of them were foolish enough to admit it's based on C's printf in their documentation, so they can get a sueball for their trouble too.
AT&T can't sue itself because C copied B, but I bet Martin Richards (BCPL) is laughing all the way to the bank, as he can sue them all.
Now if you'll excuse me I'm going to look at how CPL did it, maybe there's money to be made there too.
Repeat ad infinitum.
You can still write code to make use of someone else's API, but whether or not you can go and re-implement their API depends on the API copyright holder and their terms of use, as it should be.
Sorry, did you just post something as ridiculous as I did above?
Why should AT&T should sue one of its employee working for the company benefit? AFAIK C++ was developed at AT&T Bell Labs.
Anyway, *can* sue doesn't mean you have to - you may not be interested in protecting your copyright - and copying a single function name may not be enough to invoke copyright protection (as you can't use a single word or even sentence in a novel or song), but copying each and every function and parameters in each library is a different thing.
Why should AT&T should sue one of its employee working for the company benefit? AFAIK C++ was developed at AT&T Bell Labs.
Because he's left and he's got their proprietary super special secret-sauce API thus denying AT&T of billions.
Or maybe it wasn't just the API and had something to do with the fact that Sun/Oracle couldn't really make a useful mobile platform from Java ME. All those man hours to make a language which barely managed to faithfully reproduce Tetris.
Actually no. At the time everyone knew that Oracle were desperately trying to stretch and bend the law so they have some excuse for a case against Google's money. When they came up with "you cannot write your own code to be compatible with a published API" the only reason programmers were not shocked is because they expect this kind of rubbish from Oracle's lawyers.
Oracle did not create Java. Is was made by Sun. During Sun's death throws they tried to jump on the open source bandwagon. They released stuff under their own license (CDDL) that allowed anyone to contribute for free but no-one but Sun to profit. For some reason there were not swarms of developers rushing to work for Sun for free. Later Sun released Java under the GPL, so everyone and his penguin (including Google) had an explicit license from the copyright holder.
Oracle then bought Sun and hired David Boies so they could sue Google. (Yes that David Boies, the lawyer SCO hired to get $600 per Linux seat license fees for code that SCO were distributing under the GPL and nobody used.)
... if Google had chosen *not* to base Android on Java, then Java would be completely dead in the mobile space by now, relegated to dusty back-end server applications.
As it is, Google have handed Oracle a whole new generation of Java programmers on a plate. And what thanks do they get? Oracle sues the crap out of them.
Wrong. Sun releases only a part of Java under GPLv2, what is now in the OpenJDK - but there are still parts of Java which aren't GPL.
And I would like to see what FSF would do is someone makes a phone OS that copies everything from Linux but it's not Linux....
"And I would like to see what FSF would do is someone makes a phone OS that copies everything from Linux but it's not Linux...."
Nothing, as the FSF have nothing whatsover to do with Linux.
Not that your example makes any sense anyway, since Linux is Unix inspired.
@Flocke Kroes
No, the internal warnings came when they were deciding to base Android on Java, long before Oracle bought Sun, and long before the lawsuit.
Java SE was open sourced, but with a field of use restriction, it was not to be used in mobile platforms. That's where it all falls down, they copied the API outside of the allowed field of use.
"You can still write code to make use of someone else's API, but whether or not you can go and re-implement their API depends on the API copyright holder and their terms of use, as it should be."
Ah, but which side is the implementation? If it is a device driver interface, the OS is the caller and the third parties are the callees. Google could argue that they simply produced a new "client" of all that existing "application software". In fact, in the context of producing an alternative implementation, this is a perfectly reasonable argument and has precedents with several other groups that have produced similar "clients" over the years. Likewise, internet RFCs are (at least traditionally) not accepted unless they have multiple implementations.
End-users (including the average lawyer or judge) may think it is obvious that APIs are one sided and it is obvious which side is which. That's because they only ever see it from one side. However, to a software developer, APIs are two-sided. If you only have one implementation then you don't have an API, you just have a stamp collection.
Big APIs are hard! Take at Apache open source projects, PHP, or any of the hipster languages du-jour. You get inconsistent naming, inconsistent parameters, data structures that don't pass from component to component without translation, and language syntax hacks that make future changes impossible. Java isn't perfect but the effort shows.
It seems reasonable to protect a good API with IP laws but the amount of money involved here seems offensive to the Sun engineers and armies of outside developers and researchers that contributed too.
It seems reasonable to protect a good API
Which is why the majority of software specifications are open and royalty free? If you want interoperabiity then you don't build barriers.
I suspect one consequence of this ruling, no matter how much damages are awarded, will be to encourage Google to get rid of Dalvik. It's been working its way toward this for the last couple of years.
> Which is why the majority of software specifications are open
> and royalty free? If you want interoperabiity then you don't build barriers.
But it should still be your choice. (Think licensing)
> I suspect one consequence of this ruling, no matter how much
> damages are awarded, will be to encourage Google to get rid
> of Dalvik. It's been working its way toward this for the last couple of years.
Dalvik went years ago. They now use ART. They are DEX compilers, and are not relevant to this discussion.
A.
> Personally I'm still trying to get over the fact that an API can apparently be copyrighted.
Why? If you design a clever system, why can't you own the copyright on how it works?
> So the means of interoperating with something, which was specifically and
> exclusively designed for the purpose of interoperation, is now apparently restricted
> by law to prohibit interoperation?
No. You may design your API so paying customers can more easily integrate your product. That doesn't mean you have to give your work away for free.
A.
'Have you seen how many yachts Larry Ellison owns? Well, he "needs" some more.'
When the North and South Poles melt, and the sea level rises enough to wipe out most coastal cities, Larry wants to have enough yachts to lash them together and build his own floating country.
mysql, OpenOffice, OpenSolaris, now Java ?
Oracle has managed to kill all that in no time, good job, guyz! This ruling basically means that, if you make money distributing a Java VM with your solution, you are f'd ... I already see Larry leisuring in Walldorf ... You might get more money, but everyone and sundry will stay clear of Java, now, and Oracle! I suppose shareholders will rejoice, until the 2018 sales results come in ...
NB: Oracle bought Sun for ~$6bn, now stands to milk Google for ~$9bn and SAP ? I say good purchase!
But unlike those other projects, Java is the only one that has not been forked. Why? Simple. Nobody likes Java.* No sane person will produce code in Java unless you have a gun on your head. That gun could be an adequate compensation or just some PHB asking for "enterprise" stuff, whatever that means.
(*): People who know nothing outside of Java possibly love it. Many "computer science" programs are more like "Java science" programs.
Why? Simple. Nobody likes Java.
No, because it cannot be forked courtesy of the various non-free licensing in it. That is what this lawsuit intended to prove and it proved it - Java is NOT Free software. End of story.
Now, let's see if Google will finally grow a set of balls and throw out java completely as Android runtime in let's say Android-NG. There is a multitude of choices out there which will provide similar functionality when backed by their existing bytecode interpreter and low level runtimes. In fact ANY interpreted language will do - even Python or Perl. Though my bets will be on Go.
Considering this madness, Apple's decision to force-feed developers some Objective C no longer looks insane. Just the opposite.
I wonder if Windows is GPL compliant with their Linux subsystem ... if they ain't and I cannot see how they could be (see linking binaries, as they basically re-implemented Linux system calls) ... hm ... this is gonna be bloody ...
My gut feeling says the FSF is waiting for the prime time to sue ...
Shows the braindead oxygen hogs in Redmond, they could have implemented BSD system calls and had all the goodies, but no ...
The FSF don’t own the copyright to Linux. You obtain the GNU operating system from Ubuntu or SuSE via the Windows app store, and they own the copyright to some of that.
FreeBSD does provide Linux compatibility, but it isn’t installed by default, and I’ve never had any reason to try it out.
So does Solaris, and QNX. Linux system calling interfaces are two-a-penny these days.
The copyright to Linux is indeed not owned by the FSF. It is owned by the individual contributors of the various source code files (some of whom have now passed away). So those who wrote the Linux system call interface code going back over the past 27 years are the ones who own that API.
I've not heard that they're complaining about Solaris, QNX or WSL...
"They're trying to implement the Win32 API"
Which is what Google did with Java, they implemented the Java API with their own source code (excluding a few thousand lines iirc) - that is what Dalvik and ART are, Google implementations of the JVM/Java API and this decision is saying APIs are copyrightable, that includes the Win32 API that WINE are implementing.
They're trying to implement the Win32 API, using the reference manuals as the specification.
The text of the reference manuals is copyrighted, so if they use the same text verbatim for function names and signatures, they are - under the CAFC ruling - in potential violation of copyright. Having access to the source code is irrelevant.
... got off to a strange start because to begin with Unix was distributed as source code with no copyright notice. People fixed and improved it and some of those changes got back into what AT&T distributed. If there ever was such a thing as Unix copyrights, when AT&T was split up for abuse of monopoly the copyrights went to one of the fragments: Unix Systems Laboratories. USL promptly sued the Regents of the University of California because Berkeley Software Distribution distributed their own implementation of the Unix APIs. (USL wanted to abuse a monopoly in operating systems and lacked the brains to sue the correct organisation.)
USL proposed a settlement that had USL paying court costs, legal fees for both sides, giving up and going home with only a non-disclosure agreement to cover their embarrassment. They sold the Unix copyrights to Novell (who did not transfer them to SCO) and AFAIK the copyrights now belong to Attachmate. That did not stop the court appointed trustee for the bankrupt shell of SCO selling the Unix copyrights to some twit who probably believes he owns the Golden Gate bridge too. The intermediate court rulings from USL vs RotUC were later released as part of another case. The judge had ruled that header files were not copyrightable because they embody the API, not the implementation.
Oracle had the foresight to include some patents in their original complaint which gave the US Court of Appeals for the Federal Circuit jurisdiction - even though all the patents were thrown out of the case early on for being irrelevant, invalid or both. Although copyright law is not really their thing, the USCAFC are good at ramming patent law down the throats of anyone doing anything innovative and as we have just seen, did not let established copyright law or findings of fact from two jury verdicts stand in the way.
Unless, of course, whoever owns Novell nowadays wants to do it.
Micro Focus owns Novell. I think it very unlikely we'd start an API-copyright war. Legitimate software licensing is one thing; pursuing quixotic IP claims that fly in the face of decades of software tradition is quite another.
Plus a big part of our business is emulating mainframe environments such as CICS, JES, and IMS. Including APIs.
Like most here, I'm hoping SCOTUS hears a Google appeal and overrides CAFC. I can see CAFC's point in this ruling, but it's disastrous.
> hopefully they'll get a full one dollar as compensation
It's Oracle we're talking about here. They'll just sue and appeal and sue again and re-sue and re-appeal, forever.
Lawsuits is one of the two things Oracle knows. The other one being Mergers and Acquisitions that end up in failure.
"But where, as here, the copying is verbatim, for an identical function and purpose, and there are no changes to the expressive content or message, a mere change in format (e.g., from desktop and laptop computers to smartphones and tablets) is insufficient as a matter of law to qualify as a transformative use," the appeals court's ruling stated today.
Does that mean that tacking 'on a mobile device' isn't transformative, innovative and new for patent?
IBM on the line for Oracle. They want a word about SQL. And AT&T on the other line want to talk about C.
Might be the only thing that could revive IBM. Not that we want the Beast of Armonk becoming a patent troll (not that they have anything else going for them).
Jobs wanted to go "thermonuclear" on Android but a decade later at best Apple will walk away with maybe half a billion from Samsung, meanwhile it looks like there's a chance Oracle will collect more than 10x as much (possibly with more in the future depending on the exact ruling) thanks to purchasing Sun who didn't even care that Google was borrowing Java for their own purposes.
"it looks like there's a chance Oracle will collect more than 10x as much (possibly with more in the future depending on the exact ruling) thanks to purchasing Sun who didn't even care that Google was borrowing Java for their own purposes."
That right there is the reason:
- Oracle is still in business, and Sun is not.
- Larry is a multi-Billionaire, and Scott/Jonathan are not.
Oracle has always seemed to focus on profits. Sun, not so much.
All creative work is derivative
Possibly. This is an open problem in aesthetics. A perfectly random "work" is not in itself derivative,1 but may not be creative - that's a matter of definition. Similarly for a purely found work may not be derivative in itself,2 even if it is interpretable; consider Knapp and Michaels's "wave poem" thought experiment.
On the other hand, if you adhere to any of the schools that locate meaning-production primarily in the audience, then the aesthetic weight is in reception, and arguably so is at least a significant part of creativity. In this case all creativity would by definition be derivative, since interpretation can't happen without a ground.
1Except in the degenerate and meta-artistic sense of being generated, labeled as a work, etc.
2For found works, it's often argued that the creative act is entirely in the process of selection and publication, which is a stronger version of the qualification described in note 1, and clearly derivative both because found works have been published before and because publication follows established protocols. Thus the "in itself" qualification is particularly important in this case, and if it's disallowed we can discount found works as possible exceptions to the rule.
The problem is not the language - it's everything around. Google had to copy everything, not just the language syntax, because it didn't want to spend money into a new toolchain for Android, it wanted to reuse as much existing tools and libraries as possible. Google always takes much more than it gives back...
Surely we need a system where they can both lose?
It's pretty clear from Google's internal correspondence that they knew they were pulling a fast one, that it was all a bit dodgy, but they'd just go ahead anyway. It was cheaper, quicker and got them what they wanted. They were taking someone else's system and using it for things that Sun didn't want.
On the other hand, there are reasons why this should be possible.
So in a perfect world, both companies will spend a fortune on lawyers, in a case that causes them huge management headaches and goes on for years and years. Making both of them look bad. Then ends in an inconclusive judgement where they've both lost a fortune.
So here's to endless appeals, and huge retainers and refreshers for m'learned friends! Cheers!
So in a perfect world, both companies will spend a fortune on lawyers, in a case that causes them huge management headaches and goes on for years and years. Making both of them look bad. Then ends in an inconclusive judgement where they've both lost a fortune.
So here's to endless appeals, and huge retainers and refreshers for m'learned friends! Cheers!
Problem is, it would cause years upon years of uncertainty for vendors, developers and end-users. Sure, the Java part would be simple: just don't use Java for *anything*. But the possibility of a blanket precedent of "APIs are copyrightable" could halt technological advancement. Of course, at the fast rate the internet itself is turning to shit, we'll soon be back to letter-writing and old-school "dumbphones". Maybe even back to typewriters when lawsuits and patent-trolling make even our computers untenable.
"Who'd use Java for any kind of project now?"
To be honest, I see nothing in this judgement that changes the situation for programmers using Java on top of Oracle's JVM and very little to concern programmers on Android, unless they work for Google. Java has been a fairly unattractive language choice for a while because Oracle have been fairly Neanderthal stewards for quite a while, but if you already have 90% of the code you need and it's all Java components, the argument for writing the remaining 10% in Java is probably pretty good.
I see nothing in this judgement that changes the situation for programmers using Java on top of Oracle's JVM
I don't see anything here that affects any other JVM either. It might affect other Java SDKs, but OpenJDK's license seems pretty secure to me (IANAL) and IBM and other powerful players have big investments in it. Taking on Google was already a large risk; I don't think Oracle want to tangle with IBM on top of that.
And many organizations have huge legacy Java code bases. This decision won't even give them pause.
On the one hand this idiotic judgement will, if it stands, have horrible implications throughout the industry.
On the other hand Java is a fucking abomination. I'd like to see all vestiges of it attached to the leading edge of a descending techtonic plate.
The best outcome would be for both Oracle and Google to loose. The judge would say "No, you can't copyright an API. Also, you are assholes. Oracle is hereby ordered to pay $80bn to some guy named Bob who lives out in the desert somewhere. Google has to pay Bob $50bn, because you are also a bunch of pricks."
Nolveys wrote:
The best outcome would be for both Oracle and Google to loose. The judge would say "No, you can't copyright an API. Also, you are assholes. Oracle is hereby ordered to pay $80bn to some guy named Bob who lives out in the desert somewhere. Google has to pay Bob $50bn, because you are also a bunch of pricks."
Hmmm. I am making a wild guess here that the commenter's first name is Robert. And that he lives in a remote part of Nevada.
US courts' capacity for eyewateringly stupid decisions are renowned, so, who knows ... good luck.
Hmmm. I am making a wild guess here that the commenter's first name is Robert. And that he lives in a remote part of Nevada.
US courts' capacity for eyewateringly stupid decisions are renowned, so, who knows ... good luck.
Sounds perfectly good to me, even though I'm *NOT* Robert in Nevada.
The API is the document that tells you things like the cos function takes one double precision floating point angle in radians and returns the cosine of that angle as a double precision floating point number. The API for a language is vital for programming in the language. Sun made it easily available, but I do not know the license they chose. Such a license would give or withhold permission to copy and distribute the API documentation.
The API is also an excellent place to start coding an alternative implementation of the language (life gets difficult if the original implementation does not match the API). Before Oracle, everyone knew that creating your own code to match API documentation was legal. It turns out some judges do not understand what an enormous bomb they have just detonated under the software industry. It is almost as if learning a little about software is not a requirement for making multi-billion dollar judgements about it.
It is almost as if learning a little about software is not a requirement for making multi-billion dollar judgements about it.
Well, it isn't. The circuit court exists to interpret the law. It would be nice if it took into account the damaging effects of its decisions, but some courts (and CAFC is one) tend to prefer the "kill them all, let the legislature sort them out" approach.
Well, it isn't. The circuit court exists to interpret the law. It would be nice if it took into account the damaging effects of its decisions, but some courts (and CAFC is one) tend to prefer the "kill them all, let the legislature sort them out" approach.
And when their legal aides are reduced to using eye shades and quill pens in a dank candle-lit room just to transcribes the coutt's judgements, they may regret that.
Apple went with the NeXT codebase and the BSD-userland.
That should have told them something.
But they were probably too much Java and Linux fanbois.
Steve and Larry were good friends, neighbours at some point....
Maybe, when the cheque comes through the mail, Larry will put a flower on Steve's grave and silently mumble "I did it, buddy, I did it!"
I remember reading Andrew Orlowski’s piece https://www.theregister.co.uk/2016/06/02/google_oracle_comment/ two years ago making the case for the Java copyright to be enforced – and at the same time successfully predicting today’s legal outcome.
The article gives a small example of the code that Google allegedly copied from Java (https://regmedia.co.uk/2010/10/28/oracle_google_code_comparison_large.png) and it’s clearly not an interface but an actual implementation.
This makes Oracle’s win much more sensible to my hardly biased (I really don’t care for either party), law-illiterate mind. But then I'm wondering what the outrage is all about: am I missing something obvious here?
Orlowski does not exactly have a reputation for fair and balanced reporting when it comes to Google (or even getting important facts right). If you want to understand what is going on, spend a day or two with Groklaw's thorough coverage - up until when Lavabit shut down to avoid having to provide a back door for the TLA's
So.
In order to avoid any nasty fallouts due to languages, it is deemed best to code directly in Assembler.
Good God, and certain business groups complain that there's a skill shortage now...
Think what the IT sector would be like if that became the rule now.
Mmmh, UK Gov. projects would not see much change in usefulness or delivery delay.
"In order to avoid any nasty fallouts due to languages, it is deemed best to code directly in Assembler."
Assembler may be covered by both copyrights AND patents. Not sure on that one.
You would also need to avoid using libraries, those come with APIs that might be copyright.
Best bet is to use butterflies, coz at least you have plausible deniability. https://xkcd.com/378/
"a mere change in format (e.g., from desktop and laptop computers to smartphones and tablets) is insufficient as a matter of law to qualify as a transformative use"
So if doing something that has been done before but "on a computer" or "On a smartphone or tablet" is not legally transformative, doesnt that scupper a ton of software patents ? Including many of Oracles ?
How many dumb obvious patents have been issues that were just "Oh, this old process people have been doing for years but now on a computer or a server"
...that's the problem here. Oracle had the money to keep fighting, little people don't so if Google had been doing this to a little person they'd have just had to except it, probably declare bankruptcy from not winning the original case. But with big business, they can afford to just keep it going forever. Google will probably appeal now if allowed.
Amusing point here.
Google and Oracle can continue to appeal this for the next 10-20 years, throwing it back and forth between courts until they reach the top. The cost of litigation is a fraction of the cost of licensing or paying the damages demanded, so there is no incentive on Google to settle here.
Oracle's best option is to negotiate a lower pricing for historic and ongoing licenses for a guaranteed period of time, allowing Google to develop Android in such a way as to no longer utilise the original Oracle owned code. Google feigning moral outrage as not being allow to use code under "Fair Use" would work if they were a small software development company with a turnover of under $100k a year. Its bollocks to even try and argue that with the size and scale of the product.
I hate seeing lawyers get rich because they advise clients to fight when any sane person would tell them to negotiate and settle.
And what about the carbon-copy of Java called C# ??
Will Microsoft have to pay 100x the bill, as it was a major copyright violation to carbon copy Java back in 1999 and call it C# 1.0. First they tried it already with calling it Visual J++. Then Sun Microsystems sued Microsoft, then they tried to call it MSJava with slightly incompatible API instead. That annoyed Sun too. And then they just renamed their carbon copy and refactored it slightly and named it C# 1.0. I doubt Google's use of Java in Android is a copyright violatition, as it uses parts of OpenJDK anyway and even their own compiler - right? But doing a carbon copy of Java and before that using the Java name and trying actively to the EEE strategy, is evil and violates the copyright. Oracle, better check this one instead of the other thingy.
Oracle should read these Wikipedia articles, in case they forgot about it, what troubles Microsoft caused the former Sun Microsystems (bought by Oracle):
Visual J++: https://en.wikipedia.org/wiki/Visual_J%2B%2B
MSJava: https://en.wikipedia.org/wiki/Microsoft_Java_Virtual_Machine
I was at Sun (2004-2010) when SavaJe was purchased and J2ME was in the throes of standardization hell. I lobbied for investment with a handset manufacturer (Samsung was my choice) to partner with for an "on device portal" using Java as the base dev platform and a more functional user experience based on work done by SavaJe. The UI was wireframed out, looking somewhat similar to IOS/Android and with similar functionality. I was told there was no money or will to do it right so it wasn't going to be done at all.
In swoops Larry circa 2009, buying Sun and firing everyone even remotely connected to the project. Here he is, in hindsight, claiming someone "stole" his IP when he effectively binned it himself. Really, Larry? At long last, have you no decency?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
