Google lashes out at DoJ, Oracle as it asks US Supremes to sniff Java suit one last time Google has hit back at US president Donald Trump in the never-ending legal spat over its use of Oracle's Java code in the Android mobile operating system, urging the US Supreme Court to judicially review an appeal court's 2016 ruling against it. In a legal submission (PDF) made to the highest court in America, the Chocolate …
The real problem is that the rest of us are lumbered with a decision that turns on its head everything we've thought for all these years about coding against an API. That's the reason for all those amicus briefs. Maybe Google should have called James Gosling as a witness to tell which previous APIs he took various bits from.
"The real problem is that the rest of us are lumbered with a decision that turns on its head everything we've thought for all these years about coding against an API"
^ This is precisely why this case is so important and why Oracle (that bunch of litigation-hungry lawyers with an attached IT department) ought to ultimately lose this case.
If you are old enough to remember when Java was introduced. The mobile platform (ME) was never free. There was a reason for this. The trouble began when Moore's law made it possible for Java's standard JVMs could run on mobile devices.
From the lawsuit(s) between the two titans, it became apparent that Google muffed their clean room version, thus opening the door to this mess. So if anyone at Google wants to blame the need to continue to pay FRAND to Oracle... its on them.
As to the consumer... not much would change. Less than $1.00 per device at the time of purchase.
(Rough math YMMV but you get the idea. )
TANSTAAFL.
The way I understand it, the case if it stands as it is right now goes a long way towards eliminating "clean room" as it is currently done an option. Remember, Oracle does not contend that Google copied any executable source code, just the header files defining constants and public function names, properties, and methods.
If you are trying to write a program that is going to be cross-compiler compatible, the names of those files, and the definitions within them, must match.
There were a lot of accusations and depositions throughout the multiple trials.
This is not a clean issue.
Keep in mind that some of the depositions by Google damn Google.
My point was that Google didn't start from scratch and independently created a language that would check every box that Java does and that could be easy for developers to pick up.
They started by basing their efforts on Java.
It looks to be that they may be forced to enter into a FRAND agreement w Oracle.
8 billion - that's what, 10 minutes ad revenue?
For Google, that's about three weeks of revenue, three months of profits or 1% of the market cap.
For Oracle, that's about three months of revenue, ten months of profits and 5% of the market cap.
For Somalia, that's one year of GDP.
For The Register, that's 6.8 DUP or 76 Pogbas.
Copyright and patents on software makes no sense anyway, copyright makes sense on source code but not on running code. I hope that even with the closer Americanisation Brexit will bring, we remember that European philosophy. It's like copyrighting an accounting practice "nuh uh, you can't use P&L's, we invented that!". Aside from license restrictions, if you sit down and write some code you've just done the same thing as a brick layer. Can't copyright a brick wall either, come to think of it...... unless you're Banksy.......
Oh, what do I know. Beer time.
A fixed term would solve that problem, perhaps 20 years to match patents, or some shorter period for software to reflect the shorter useful life of most software.That wouldn't comply with WIPO rules, which keep the idea that Victor Hugo's grandchildren should be able to live off his work, but amending the WTO's rules to relax compliance requirements shouldn't be too hard since only 3 countries make a net profit from IP exports (and one of them is Ireland)
Copyright and patents for software in general make perfect sense as long as you have a patent-awarding body with more smarts than the USPTO.
If I write some software, why should anyone else be able to copy it buckshee if I don't want them to? If it were a novel it would be copyrightable, and software is only the same thing - text (or at least representable as text). Software patents make sense too if handled sensibly, I should know as I've got a half dozen or so of the things. If I make a hardware widget that performs a certain action, I can patent that, so why not a software widget too? Where it all goes wrong is the USPTO granting massively overbroad patents or patents for things that (patently) fail the test of non-obviousness.
Anyway, that's not the case in point here, what is being argued over is APIs, a very different thing from software and a lot more slippery to pin down. In a way it's more like the old spat between Intel and Zilog over the Intel 8080 instruction set, which is sort of like the API for a processor's internal workings. Zilog reimplemented the entire 8080 instruction set from scratch (and some) and then the Intel lawyers turned up. In the end, it was ruled that Intel couldn't stop other companies reimplementing the same instruction set, but they could copyright the instruction mnemonics and so Zilog had to give all the instructions different mnemonics in their assembler. A bit of a pain if you needed to work on both Intel and Zilog, but liveable.
Edit for crossposting: there is only one indefinite copyright that I'm aware of, and that's Peter Pan. Everything else is lifetime plus a certain number of years which varies from jurisdiction to jurisdiction.
When the USPTO made software patentable there was a mad gold rush to patent the 'bleedin' obvious' by all and sundry regardless of whether the idea was novel, published or even invented by the claimant. It was a true Gold Rush, one that corporations had to participate in because "if they didn't then someone else would". I've had some of my own 'original' work patented by others -- I use the term original in quotes because I'm pretty sure it was derived from earlier work, its what I'd call 'subjectively original'. I've seen people patenting things like framing patterns in data streams because they were not au fait enough with communication technology to understand that *all* data streams use framing patterns (and the USPTO naturally granted it). Its a mess and it has contributed little to the development of business and technology.
Copyright has its place but its also easily abused. When IBM made their desktop PC in the early 80s they did what was then routine for all of their computer products, they sold the PC with a three ring binder manual that included comprehensive schematics and the assembler source code for their BIOS. This was the origin of the PC clone industry, a mad scramble to build imitation PCs using their design and code. IBM eventually took note and sent legal notices to these companies asking them to stop using their BIOS code (but giving them a couple of months' grace to replace it - those were the days!). That's blatant theft. Meanwhile people patent and copyright stuff that isn't really theirs -- its obvious stuff (rectangle with rounded corners) but they expect to sit on this and use it to generate an income stream for ever. This does't contribute to society, its just theft of the commons.
As for APIs, if you don't want people to use your language or libraries then don't publish the API. Its simple. If you do publish the API then it will be in the form of both code and documentation.
In the UK the Authorised Version of the Bible also has indefinite copyright
I think you'll find that it no longer has copyright [1] - which is why all the Bible apps include a copy of the text. Newer versions (for example - The New International Version) are very much in copyright which is why you usually get charged to install a copy of their text.
[1] It is instead protected by a Royal Prerogative, restricting publication and distribution rights solely to the crown. Who has the attitude of "as long as you don't change it, you can distribute it all you like".
Software patents are unnecessary. Copyrights prevent software from being copied, unless the original author permits it. If you write software and don't want someone to copy it, that system already works. If someone figures out another way to implement the same thing with their own unique code, that's perfectly fine too, as it should be.
Functional devices can't be copyrighted, so we have patents instead - if you publicly document the workings (implementation) of a widget via a patent, in exchange you get a temporary legal monopoly for producing it. If someone comes up with a different method for doing the same thing they quite likely won't infringe the patent. If you don't patent it then anyone is free to copy or adapt the implementation.
Software patents on the other hand often prevent others from implementing something even if they don't have access to the inner workings (source code) and come up with their own unique implementation. This concept is absurd and is why patents don't translate well to software. Perhaps if software patents made working example source code a mandatory part of submission then they would better reflect their traditional counterparts, so that others would be able to invent their own non-infringing implementations.
Patent on software - no because the USPTO or local equivalent is not smart enough to recognize a truly novel software idea from the general application of a manual method or mathematical theory in code.
Copyright - no with the current copyright laws (too long a term). If the term is more reasonable say 15 years or so then ok.
Copyright and patents on software makes no sense anyway, copyright makes sense on source code but not on running code.
GPL2 relies entirely on copyright law to work. Major luminaries in the Linux world (eg Redhat) also rely on copyright protection of compiled binaries derived from GPL2.
It has been argued that dilution of copyright law much as Google have been seeking in this case would break GPL2. That’s not been tested, Google lost the “fair use” argument.
"GPL2 relies entirely on copyright law to work."
No, it's not. It relies on license as a contract. At least according to everyone from fsf...
Copyright covers individual code contribution - each coder retains his/hers copyright. If copyright is complete abolished tomorrow, gpl will be just fine.
I think you’ll find it otherwise. Wikipedia:
“Those who do not accept the GPL's terms and conditions do not have permission, under copyright law, to copy or distribute GPL licensed software or derivative works.”
So whilst it might be a license as contract, it’s strength relies on copyright.
"GPL2 relies entirely on copyright law to work."No, it's not. It relies on license as a contract. At least according to everyone from fsf...
If there was no copyright on the work, there'd be no way to force someone to engage in taking out a license, there'd be no basis to require a contract. If it's not copyrighted, patented, or a trade secret, then one wouldn't need a license at all.
With no copyright everything would be effectively under CC-SA (or CC-SA-BY in countries with "moral rights"). That's weaker than the GPL because it doesn't include the "preferred form for modification" and weaker than the GPL3 because it doesn't include the anti-TiVoisation clause, and the common F/OSS patent protection clauses wouldn't work (but presumably there would be no applicable patents either).
However, you can also fake copyright, by releasing something subject to an NDA that the recipient will undertake not to redistribute it except as listed, with their ability to access your secret literature as their consideration to make a valid contract. The major difference is that it would only allow you to sue the first person to release your item, not all the subsequent redistributors who never agreed to your contract.
.....not some kind of patent troll designed to snare successful users of the language to boost Oracle's bottom line.
Google didn't copy anything, it just used an API. APIs by their very nature use common code that's incorporated into the user's application -- we all know that and we all know that sometimes the definitions are just function prototypes, sometimes their code stubs, it just depends on the particular structure of the language.
All this has done is prove that Java is a useless language and that Oracle can't be trusted with any generic IP. Java itself is a derivative language, the whole concept of how it works is based on earlier languages (my guess Pascal and Forth**) so its not as if its some kind of 'must have' novel technology.
(**Pascal for the P-Machine, Forth for the variable stack / return stack concept.)
Google didn't copy anything
The courts decided that Google did copy Oracle code.
APIs by their very nature use common code that's incorporated into the user's application
Yes, but unless explicitly granted license you don't have a free hand in altering, distributing code that defines an API (or so it seems based on the long and turgid history of this case). And normally a dev wouldn't be interested in doing so anyway, because a dev is simply using the API definition to access the API implementation itself. What I think has happened (more scholarly and patient observers will no doubt correct me) in this case is that Google liked to the look of Oracle's API definition, and ripped it off as part of creating an incompatible implementation.
we all know that and we all know that sometimes the definitions are just function prototypes, sometimes their code stubs, it just depends on the particular structure of the language.
And if you copy them in the same way Google did, you are breaching someone else's copyright. Whilst it remains to be seen if this counts as "fair use", given Google's failures several times to establish that their use was "fair", it seems unlikely that they will succeed and nor will anyone else.
All this has done is prove that Java is a useless language
Millions of Java devs down the decades would disagree.
and that Oracle can't be trusted with any generic IP
What's that supposed to mean? What is generic IP, and what is specialised IP?
Java itself is a derivative language, the whole concept of how it works is based on earlier languages (my guess Pascal and Forth**)
Oh dearie me
so its not as if its some kind of 'must have' novel technology.
No, it's no longer novel, but given how much of it is out there it is totally essential.
I don't care. It's just that I think it ridiculous commentards make their own pointless assertions about what Google may or may not have done with Oracle's code, and the rightness or otherwise of it. The courts are the ones making the decisions here, and it's sensless to pretend otherwise. We pay the courts to do it, and they're a lot more thorough at digging into the nooks and crannies of a case than the commentariat who either think that freeloading is a God given right, or the complete opposite. I especially think this when they go on to suppose that Java is somehow descended from Pascal or Forth, of all things.
My own view is that copyright law should be upheld to the maximum possible extent. We've all got a lot to lose otherwise. If that means Google owe Oracles $8billion, so be it. If IBM could in principal have a similar case against Oracle but have decided not to in the interests of all, well that's very generous of them. Though I have to say that IBM suing Oracle over SQL would seem more like Bell labs suing Microsoft or Borland because they had the temerity to implement their own C compilers with an intent of meeting an agreed specification (i.e. the complete opposite of Google).
You're taking a leap of faith that the court decision is correct, though.
Coding is hard, and I posit it's unlikely the judge(s) hearing the case are experienced developers with a solid understand of how and why APIs exist and are used.
Instead, they rely on so-called expert testimony from both sides, and whichever side sways them the most is the one they base their decisions on. I find that wholly inadequate for determining legal positions because, as ever, it will come down to which side has the most compelling lawyers, rather than technical accuracy.
The very nature of APIs is that they are meaningless interfaces to the actual code underneath.
If you have a function called "get_value" in more than 1 language that does not make them the same.
Equally so, to common sense at least, the same declaration with 2 completely different implementations underneath is also not the same.
A, probably bad, analogy is the existence of 2 different houses with the same design and colour front door does not mean you can go in to the one you don't own and claim it is yours.
Obviously Google chose to use the same function names so that Java would just work in their JVM without Android devs having to rewrite anything, but they should have written the actual code behind the APIs blindly.
True, and on that hand I don't really care which screws which one.
But as already pointed out the Oracle win would set a very disturbing precedent to allow any company to block 3rd party products simply by claiming copyright violation on the API that is necessary to make something work.
> allow any company to block 3rd party products simply by claiming copyright violation on the API that is necessary to make something work.
There's no reason why these new entrants need to interact with existing market participants or established player's market. What resolves this in oracle's favour is that google needs a permission to enter oracle's java programming market. If they ruin the license negotiations by not agreeing to oracle's market rules/which seems kinda onerous), then they simply don't have permission to interact with any oracle products.
While this could be considered significant limitation that products cannot interact with each other, it's basically what copyright wants -- different vendors to invent their custom solutions with different pros and cons. If they all clone each other's interfaces, the new products will only be clones of products that are already in the market, and copyright's main purpose of promoting the progress and useful arts wont be properly implemented.
This is all a powerful signal to all businesses and individuals out there to avoid touching anything Oracle because Oracle is going to find a way to steal your kids. A
We have now banned MySQL, VirtualBox and most Java from this business because an employee downloading some seemingly innocent GUI tool might make us suddenly in breach of Oracle's world order.
"We have now banned MySQL, VirtualBox and most Java from this business because an employee downloading some seemingly innocent GUI tool might make us suddenly in breach of Oracle's world order"
...otherwise a rather large predatory bill would be finding its way from Oracle's accounting department to the company's chief financial officer.
I am sure lawyers at IBM are sitting on the side lines thinking boy are we in for a huge pay day if Oracle win. That is Relational Software, Inc. (now Oracle Corporation) didn't have a license for Structured English Query Language when they incorporated it into their flag ship database, and now we want a big a really big payday that will make the payday from Google look like loose change.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
