Slashdot Mirror
Google Takes the Fight With Oracle To the Supreme Court
whoever57 writes Google has asked the Supreme Court to review the issue of whether APIs can be copyrighted. Google beat Oracle in the trial court, where a judge with a software background ruled that APIs could not be copyrighted. but the Appeals court sided with Oracle, ruling that APIs can be copyrighted. Now Google is asking the Supreme Court to overturn that decision.
(Also of interest.)
Fuck you. You are everything wrong with the software industry.
Sonia Sotomayor: isnt...didnt steve jobs invent API's when he did the iPod?
Stephen G. Breyer: im pretty sure its bread...ive had an API before and it gave me horrible gas.
Samuel A. Alito: no guys seriously its Oracle this is about religion. the API is like a god, but also a man i think.
Elena Kagan: I believe API is the knob on the old chairs that used to control that horrible frontward tilt feature we all hated. Oracle owns that?
Clarence Thomas:: zzzZZZZzzzzzZZZZzzzz
Antonin Scalia: It should be outlawed. these damn hooligans play their API at four in the morning and I hate rap bands.
Chief Justice John G. Roberts: I dont know about you guys but I rather liked the front tilty chairs...just sayin'
Anthony Kennedy: does everyone get an API or just some people. does this disenfranchise minorities or is there an API for everyone?
Ruth Bader Ginsburg: Youre all wrong. I just had the API serviced in my mercedes benz last month. im pretty sure its the part that works with the seat warmer or cooler or something.
Good people go to bed earlier.
If Oracle has their way, then IBM should be able to sue them out of existence since their entire business was based on copying IBM's database source code and APIs to make the Oracle Database.
Oracle are clearly trying to pull Java back from public domain and back to within Oracles control to undo the GPL license Java is under. It's a clever lawyer trick, but Java itself uses copyrighted APIs.
Java was not the first to use Vector, or String classes or Views or any other API and classes it built on. It would be difficult to even identify which *names* of classes are actually new APIs and which are copied from others.
In Oracle's mind, it thinks if it can get away with this is can seize Android, which contains a Java compatible API named set, even though it doesn't use Java and the code is not theirs. But that API set is itself copied from many previous products.
Why do we have trademarks if you can copyright the name of something? (Which is what an API is, its the names of the methods).
1. The don't use the Java brand and don't call it Java.
2. Its not based on Sun/Oracle sources and relies on a clean room implementation
3. Google didn't sign a licensing contract then violate it like MS did.
IIRC Sun Microsystems got into similar legal battles with Microsoft over the latter's implementation of a Java Virtual Machine. Sun claimed it licensed use of the *Sun* JDK and JVM to Microsoft, not the rights to re-implement the language with its own engine.
Of course, Microsoft eventually responded by creating a competing set of virtual machine-based languages that were sufficiently different from Java as to not violate copyright.
Now, that might have been a licensing issue, but the copyright issue is interesting too. There were the famous battles of spreadsheet "Look and Feel" where Borland Quattro Pro emulated the Lotus 1-2-3 menu hierarchy; Lotus successfully sued for damages.
Can someone explain how it's not a slam-dunk argument that APIs fall under the scope limitation of 17 USC 102(b)? Isn't that a key underpinning of decades of case law on very nearly this exact subject (Computer Associates v. Altai, Lotus v. Borland, Sega v. Accolade, Sony v. Connectix)?
"The Congress shall have power ... TO PROMOTE THE PROGRESS of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries..." [Caps mine]
This fails the promoting progress requirement.
Prove anything by multiplying Huge Number times Tiny Number
Copyrighting APIs makes no techinical sense, while copyrighting maps and manuals does.
The only thing remotely plausible is patenting APIs. And even that I'm not sure technically feasible.
I've witnessed probably man/centuries of wasted efforts due to crappy APIs - and I've seen single-person projects springing to life and seamingly leveling mountains when the available APIs were good.
Some APIs fit the puzzle of any applications - some force projects as a whole back to the drawing board.
All hope abandon ye who enter here.
> I've witnessed probably man/centuries of wasted efforts due to crappy APIs -
> and I've seen single-person projects springing to life and seamingly leveling mountains when the available APIs were good.
> Some APIs fit the puzzle of any applications - some force projects as a whole back to the drawing board.
So you're saying there are good APIs and bad APIs, that there is an art to writing a concise, balanced, beautiful API?
I want Google to win this one, but copyright _is_ supposed to cover stuff you write where there's an art to writing it well.
The concept of APIs can also be extended to any "interfacing" whether software, electrical, mechanical or perhaps even biochemical (proteins etc?) or "human communication protocols".
Where do we go from here?
4wdloop
The full source code of the UNIX v6 kernel, as published in the Lions commentary, bore prominent copyright notices from AT&T Bell Labs.
If the system call and C library API interface is thus still owned by Bell Labs, then that covers Oracle Linux, the POSIX standard, commercial UNIX, as well as all the phones (including QNX), routers, UNIX/Linux/BSD servers/workstations, and likely much more.
Oracle had better pray that they lose.
There is definitely a value (art) in good API. However...then there should be a similar balance as in patent intentions (no, not US implementation of patents) ie keep balance of reward/incentive to innovate with general society benefit.
If API was found to be copyright-able everybody would stay away of everybody's else APIs unless given to public domain. This at least would make clear statement of what to do - do not use copyrighted APIs. However this defeats the purpose of APIs: cooperation.
I at least hope they would not apply it retroactively - this would be mayhem.
4wdloop
Nice subtle moving of the goalposts there...
An API is not the same thing as API code. API code is an implementation of an API, and not the API itself, even if you describe the API by using a programming language notation as if you were writing code.
Code is obviously copyrightable, regardless if it forms part of an API or not... But the API itself is only the interface, and does not actually contain any code, and such an interface is ultimately only an idea... there's nothing concrete about it like there is with code.
Of course, a particular expression of any idea may be entirely copyrightable, but even that doesn't give one a license to exclude other people from expressing the same idea on their own... especially when the idea is narrow enough in scope that there aren't many unique ways to express it in the first place.
File under 'M' for 'Manic ranting'
They will have to use sock puppets and crayons to argue this case with these judges.
There were the famous battles of spreadsheet "Look and Feel" where Borland Quattro Pro emulated the Lotus 1-2-3 menu hierarchy; Lotus successfully sued for damages.
How successful was Lotus actually? The Court of Appeals for the First Circuit ruled in favor of Borland, and the Supremes agreed.
It's hard to decide how the SCUSA would rule in this case because it is two big corporations fighting. If it were a corporation versus a human being then it would be an easy decision for the SCUSA, but I don't think five of those nine Justices have any compass for how to rule in cases between two corporations.
It'll be really confusing for them. Imagine the conversation:
Roberts: "Hey, Thomas, how are you going to vote?" ... ... ... ...
Thomas: "I don't know, Johnny. I read the briefs and shat my pants when I saw that both litigants were corporations."
Scalia: "Me too. I even asked both sides if maybe they were not a corporation, to simplify things, but it seems like they really are both corporations."
Roberts: "Yeah I'm flummoxed. There's just no way to decide."
Alito: "Maybe one of them is a small corporation? Then we could just round them down to 'human' and rule against them."
Roberts: "No such luck. These are both huge corporations."
Kennedy: "Hey, guys, I'm thinking maybe we could decide the case based on legal principles."
Roberts:
Scalia:
Alito: "I don't get it. Legal principles? You mean like, we should check the documents of incorporation to see whether they are real corporations?"
Kennedy: "No, I mean like, we should decide the case based on what the law says, and based on previous legal decisions in American courts."
Roberts:
Scalia:
Alito: "I still don't get it. What does the law have to do with it?"
Thomas: "Yeah, seriously Kennedy, we let you screw up the DOMA decision and we're not going to let you do it again. Is this 'API' thing religious? Maybe we could rule in favor of religion, if the other corporation is atheist."
Alito: "No, apparently it's some kind of computery thing. I don't really know, I was doing a Sudoku during arguments."
Roberts: "Look, maybe we should just defer to the appellate court."
Scalia: "But the appellate court ruled against a corporation!"
Roberts: "Yeah but they also ruled for a corporation."
Roberts, Thomas, Scalia, Alito: heads physically explode.
Kennedy: "Ah, sheesh, they got blood on my robe."
Ginsberg: "Anthony, come over here, maybe you can help Sonya and I write a decision."
Kennedy: "I guess so, but your lap isn't as comfortable as Scalia's. Will you scratch me behind my ear?"
Ginsberg: "I'll scratch you behind your ear but you have to promise to stop mentioning legal principles in front of the conservatives."
I seem to remember the opinion of the Supreme Court in Eldred v. Ashcroft (the Copyright Term Extension Act case) stating in effect that "to promote the progress" means "to attempt to promote the progress". It defers to Congress on what constitutes promoting.
Again, I'm expressing arguments against the outcome I prefer.
If, as you argue, the purpose for creating your API is to facilitate cooperation, would you not license it to allow cooperation?
Patent/IP Attorney chiming in.
Sure. You're very specific expression of both. You don't get to copyright the "facts" in either case. Everyone is free to copy the factual aspects of both maps and manuals. The distinction is an important one. And while not addressing maps and manuals directly in their petition, Google takes on the concept: you can write a book about art, but you don't get to stop other people from doing that art.
You can draw (and potentially get a copyright for) an outline of the world, but you can't stop other people from doing the same.
Google's argument about the API isn't all that much different: you can copyright an implementation of the API, but you don't get a copyright for the "facts" of the API: function names, arrangement, etc. The argument is that copyright, by statute, expressly does not extend to "extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work." (17 USC 102(b)). This is not a trivial argument. It's also pretty important because if Oracle/Federal Circuit are correct, then you can have de facto patent protection for a century without any of the procedural protections such as examination.
You can't copyright a single telephone number, but you can copyright a telephone book.
Actually, you cannot copyright a telephone book.
I have no problem with the idea of an API being copyrighted. But please note, "we needed to do it for interoperability" has been found to be a fair use justification for reverse engineering undocumented APIs; if you actually publish it, you're doing it to encourage interoperability and writing to the API is fair use.
Trollmod for a perfectly cromulent explanation.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
> If the cost of making something is so low that a >100 year monopoly isn't needed for its production, then either the copyright term must be diminished or the product should not be covered by copyright.
The term of copyright is ridiculously long. I would think that anyone who thinks copyright is a good idea should support reducing the length considerably. Having it so long encourages and provides good arguments for those who wish to greatly restrict copyright. A term of around seven years would accomplish the stated purpose, without unnecessary drawbacks.
Didn't Oracle prevail on appeal by claiming that the structure, sequence, and organization of the Java API was copyrighted, not the fact that there's a function called this, a function called that, etc.?
In 3010, the potatoes triumphed
How will this impact Microsofts Android Tax? I mean doesn't Microsoft have patented intellectual property rights to Android?
M-Cam casts doubts on Microsoft's Android patent portfolio
if an API is a list of facts.... Maybe Oracle seeks copyright protection, as a phone book would, of their compilation of facts.
This is precisely what copyright does *NOT* cover. There must be millions of equally good ways of doing the same thing, and you may protect the non-functional aspects of the one way that you creatively chose. Here, the name is purely functional -- play(soundname) is the only way to get the API to play a sound -- it is the one way to achieve a particular functional result, so it cannot be protected by copyright.
You can copyright a sentence you solely authored if it has sufficient creative expression. But you cannot enforce that copyright against someone who uses that sentence because it is the only practical way to achieve a particular functional result. If the sentence uses an API, then complying with the API is the only way to achieve the functional result desired. The purpose of APIs is to standardize the way to achieve particular results.