US Justice Department Urges Supreme Court Not To Take Up Google v. Oracle 223
New submitter Areyoukiddingme writes: The Solicitor General of the Justice Department has filed a response to the US Supreme Court's solicitation of advice regarding the Google vs. Oracle ruling and subsequent overturning by the Federal Circuit. The response recommends that the Federal Circuit ruling stand, allowing Oracle to retain copyright to the Java API.
Clean room implementation? (Score:5, Insightful)
Re:Clean room implementation? (Score:5, Informative)
Yes. Exactly.
It's all about the term of copyright versus the term of patent. Patent lasts only twenty years at present, while copyright is effectively perpetual (whenever Pooh and Mickey might enter the public domain, the legislators fix it). If copyright governs interfaces, that part of the law will keep the government from stealing IP away from its rightful owners after twenty years.
Re:Clean room implementation? (Score:5, Insightful)
Can they make Compaq's reverse engineering of IBM's BIOS illegal retroactively, and take back much of the PC revolution?
Re:Clean room implementation? (Score:5, Interesting)
Or in a slightly less Dystopian view, computers today would look like iPhones, with one vendor having a stranglehold on the platform and completely anemic third party hardware support outside of cosmetics.
Re:Clean room implementation? (Score:5, Insightful)
Oracle must have contributed to the right Administration official.
Re: (Score:3, Insightful)
To be fair, if it was (R) it would have happened too. And when people say that the (D) and (R) are really different, you can point them to this case to show that ... no not really.
No. Solicitor General's Office is Good (Score:2)
Oracle must have contributed to the right Administration official.
No.
The United States Solicitor General's Office writes a *LOT* of these "op" briefs. Generally ten pages saying "This case isn't so magically important that SCOTUS should look at it among the thousands of cases people are asking to look at." Taking that position on a case is the default position and nothing should be read into it.
They are for the most part really top-tier professionals who are trying to make the decision based on what is best for the US Government as an institution. Not influenced very
Re:Clean room implementation? (Score:5, Insightful)
Why because the court asked for the government's opinion?
No. Because the government's opinion sounds like it was written by Oracle's general counsel
Re: (Score:2)
Why because the court asked for the government's opinion?
Yes, actually. Why should they bother, in this case?
Re:Clean room implementation? (Score:5, Insightful)
MOD PARENT UP! (Score:4, Interesting)
Or Google has been resisting the NSA a little too much.
This.
It's pretty obvious that this is a punishment for adding encryption to Android devices, and for going to SSL for all web transactions, making it much more difficult to spy, despite administrative objections.
The recommendation is clearly punitive because Google has pissed the executive off, and consistently opted on the side of data protection, and has disclosed many of the recently discovered OpenSSL and SSSL protocol flaws which made eavesdropping easier.
Re: (Score:2, Funny)
There are treatments for paranoid schizophrenia nowadays, you don't need to live with the disorder.
Re: (Score:2)
Re:Clean room implementation? (Score:4, Informative)
The Laws we have in place are the same as we had back then. The main difference today is that people holding public offices tend to flaunt their pay-for-play status, where back in the 80s/90s they were still attempting to hide it. The biggest harm to IT took a while to get precedents set, but really started almost immediately with "ideas" being patented and copyrighted (you can thank the first Bush for that lovely patent reform).
As an example, Athena (X) was developed mostly by DARPA funding and grant money. Yet we had to see 32 screens worth of copyrights just to start the Xserver (okay, 32 is an exaggeration but the point remains). Some of these were to Universities like MIT, Berkley, and Stanford. Many others though were to Novell, Sun Microsystems, Hewlett-Packard, IBM, etc.. etc... And no, these were not "credits", but copyrights. This is why Linux started with a pretty old version of X and basically had to reinvent the wheel. Linux had 1 crappy pay-for version of CDE because some schlep company ended up buying copyrights to extort money from people.
Re: (Score:3)
The Laws we have in place are the same as we had back then.
The DMCA most definitely wasn't in place back then.
Re: (Score:2)
Linux had 1 crappy pay-for version of CDE because some schlep company ended up buying copyrights to extort money from people.
I don't even remember there being a pay-for version of CDE for Linux. I'm not saying it didn't happen. I just remember you could buy a Motif tarball from Metrolink that would get you Motif and mwm, not like you would ever use mwm when you had fvwm. And then later you could buy Caldera Network Desktop, which came with Metrolink Motif. You could also buy AccelX, which got you a substantially faster X server back in those days, with meaningful support for your video card's 2d acceleration features... something
Re: (Score:3)
I am bothered by technical capabilities being copyrighted instead of patented
Interfaces, including APIs, should be neither copyrightable nor patentable. There is precedent for this in prior supreme court decisions [wikipedia.org] concerning the interface for printer cartridges. In principle, this is no different.
Re: (Score:2)
Re:Clean room implementation? (Score:5, Insightful)
If copyright governs interfaces, that part of the law will keep the government from stealing IP away from its rightful owners after twenty years.
You mean stop government from returning it to the rightful owners. The public (and public domain) are the rightful owners of all information and works - copyright/patents just give exclusive use for a time.
Re: (Score:2)
the public are not the rightful owners, either. After the patent or copyright expires, nobody is, and hence anyone is free to use it.
Copyrights don't expire. Eldred v. Ashcroft.
Re: (Score:2)
Eldred v. Ashcroft ruled that you couldn't declare a particular term extension unconstitutional.
No. It just established that the arguments put forth in that case weren't enough to overturn the 1998 extension. It also sets a precedent that will weaken any similar challenges. It does not mean that another case against term extensions couldn't succeed in the future.
A periodic formality, like adopting House rules (Score:3)
A pattern of Congress continually extending term lengths retroactively is not the same as a law declaring that copyrights do not expire, because the action that occurs if Congress does not act is that copyrights expire. Whereas in the latter scenario Congress has to act in order to make copyrights expire.
Each house of Congress also has to act every two years in order to set its rules. The requirement of a periodic formality to prevent copyrights from expiring does not change the practical outcome, just as the requirement of a periodic formality to readopt House and Senate rules every two years does not keep the House and Senate from having rules.
Nobody actually wants perpetual copyright terms, except maybe Disney.
And the Gershwin estate. And the leadership of the Motion Picture Assocation of America (to find sources, search the web for the phrase "forever less one day"). And
Re: (Score:2)
If we are splitting hairs nobody is the rightful owner before the patent or copyright expires either. Neither of those things confer ownership.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re:Clean room implementation? (Score:4, Interesting)
Google probably does copy/paste of the Java source code all the time, like lots of other people, because it's open source, so that is a meaningless accusation.
Google distributed interface definitions that look very similar to Oracle's. That's probably because in Java, there really aren't a lot of different ways of describing the same interface. Furthermore, interfaces should not usually be considered copyrightable.
Google also distributed some copyrighted Java source files. That was stupid, but those files appear to have been test cases, not code that ships on handsets, and it appears to have been unintentional. It's hard to argue that Oracle suffered any harm from that and Google came into compliance.
Re: (Score:2)
That's a pretty insane statement. Copyright doesn't grant ownership and there are no rightful owners. Copyright is government imposed limitations on the natural rights of everyone else. If you write a book, you own THAT book and that particular copy of it's contents nothing more or less. I have the right to copy any software, api, song, movie, etc limited only by
Re: (Score:3, Insightful)
Even so, I don't think an API should be copyrightable, but if it is, it should be considered fair use to actually use it. Otherwise, the software industry just wouldn't work.
Re: (Score:3)
Nitpick - technically, we're talking about Java, where there are no header files. The interface is defined in the same source files as the implementation.
You're correct that Java doesn't have header files; Java uses Interfaces which fulfill the same function. Best practice is to write your API using Interfaces, and not code to a particular implementation (classes).
You can define Interfaces and Classes in the same source file, but people typically create a separate file for each one.
Re: (Score:2)
I disagree. APIs are a valid expression of an idea. Given an area of computing (regex, networking), there are dozens if not hundreds of ways you can design the API -- so it's a creative form of expression which is what copyright laws are supposed to protect.
There are plenty of different ways to define class names (and their responsibilit
Re: (Score:3)
Sorry, but APIs are not fixed format like directories: "name: ph. number." As I stated above, there are plenty of subjective and creative design decisions that go into creating an API. This is very similar to the subjective and creative decisions that go into writing a novel. Hence both APIs and books should receive copyright protection.
For example, both OpenGL and DirectX are 3D APIs that ca
Re: (Score:3)
Doesn't the Linux kernel group hold a very similar stance in that you cannot use the kernels internal APIs without breaching copyright and thus falling under the GPL as a derivative work?
Re:Clean room implementation? (Score:5, Interesting)
I believe it's not that you can't use the API, but that you can't interface non-GPL code with the kernel due to the restrictions of the GPL. That you would use the API to do so is incidental to that restriction.
If instead you want to build your own kernel implementing the same API, I don't believe they have any objection.
Re: (Score:3)
The only sway the GPL has here is due to copyright law, so if the GPL is in force in my example then that means the API is considered to be copyrighted...
No copyright issues means no GPL restrictions.
Re:Clean room implementation? (Score:5, Informative)
No, it does not. Re-read my post after you've had your morning coffee. You're free to use the API however you want, it's presumed not copyrightable. The *one* exception is using it for interacting with the Linux kernel, because the kernel *is* protected from such access by the GPL, and only GPL-compatible code is allowed to interact with its internals. The API is irrelevant to that fact - it's simply the interface used by those who *are* allowed to interact.
No (Score:2)
Re: (Score:2)
Doesn't the Linux kernel group hold a very similar stance in that you cannot use the kernels internal APIs without breaching copyright and thus falling under the GPL as a derivative work?
Not really [tldp.org]. TL;DR: Linus doesn't say so, and he holds the trademark, so he gets to decide what makes "Linux(tm)".
Re: (Score:2)
Doesn't the Linux kernel group hold a very similar stance in that you cannot use the kernels internal APIs without breaching copyright and thus falling under the GPL as a derivative work?
Using the API doesn't invoke the GPL -- it's actually the linking process. When you link against the kernel, your binary becomes a derivative work partially based on the kernel. Since the kernel is covered by the GPL, creating any derivative work from it requires you abide by the terms of the GPL. I doubt any FSF lawyer would have anything negative to say about cloning a project's public API in order to create a completely new implementation under a different license.
Another example: If I created a GPL
Re: (Score:2)
Internal APIs are part of the kernel implementation; that's why they change so often. If you're using one, you're implementing a kernel piece, and therefore creating a derived work.
The external API that is for use by non-kernel software has no such restriction. You can write any code you want that makes calls to the Linux kernel. You may not be able to _ship_ the linux kernel along with your code, if your code isn't GPL, but that's a restriction on what you can do with linux code, not with your own.
Re: (Score:2)
flink's post below talking about linking makes more sense than mine.
Re: (Score:3)
You can use Linux APIs all you want, no problem.
You can't link certain code directly into the Linux kernel, because at that point you're creating a derivative work that has to be licensed under the GPLv2, and if the license of the code you added isn't compatible with GPLv2 you cannot satisfy both licenses simultaneously.
Re: (Score:3)
Right now they are talking about if the API is actually copyrighted. If it is, there is still a (good?) chance that fair use will allow you to reimplement it anyway; but that's going to be another court case, likely.
Nokia (Score:2)
If this is so, then Nokia can now assert copyright over fork().
Nokia now owns Bell Labs through a long chain of acquisitions. Bell Labs publicly asserted copyright over fork() in the Lions Commentary.
Nokia should now assert infringement over Solaris and the UEK. A sizable portion of Exadata revenues are fairly owed should this decision stand.
Re: (Score:3)
If I'm reading this right, Google incorporated Oracle's Standard Library wholesale, instead of re-implementing the Standard Library from scratch.
You are not reading it right. The standard library was re-implemented using the same API. That's why the government's stance would destroy the software industry as we know it because a complete re-implementation would still be covered by the copyright of the original implementation. This is exactly copyrighting an idea instead of a particular expression of an idea.
My guess is that the reason for this idiotic position is to intentionally kill off Linux and all independent software development in orde
Java is done (Score:4, Funny)
If this stands, Java's vaunted claim to being on "billions of devices" will soon become the punchline to a bad joke.
Re: (Score:3, Funny)
Re:Java is done (Score:5, Funny)
Yes, but that would take a couple billion years.
Re:Java is done (Score:4, Interesting)
Re: (Score:2)
Re: (Score:2)
Sun would be just fine today if Oracle hadn't bought them.
Sorry, I'm going to have to disagree. Sun was a hardware company that also sold an operating system. From the late 90's on Sun began losing marketshare to Linux on commodity hardware. Sun tried to respond by moving to x86 and publishing a companion CD/DVD of (old versions) of popular open-source software with limited success. It's a fact that Solaris was pretty much useless without also installing a bunch of GNU software as Sun shipped obsolete/buggy versions of quite a few programs. People who had ex
Re: (Score:2)
It's hard to believe that Sun would have been fine today, if Oracle hasn't bought them. We kept hearing the same story every year since 2000, and yet Sun's stock never recovered after the dotcom bubble burst. Indeed, the market has become very commodidized with Lintel and Wintel getting better and better at every iteration. Sun's corporate culture that aimed to protect Solaris and Sparc made the company take a wrong turn in the late 90s and early 2000s, by completely missing a chance to embrace Linux, x86,
Re: (Score:3)
That's retarded. Oracle had (and maintains) a sizeable investment in Java and the rack servers for which Oracle is optimized for.
The acquisition was about securing the investment. Not any devious scheme.
Disclaimer: I work for Oracle but am not in any way associated with the Java group nor am I part of the executive/decision-making chain.
Re: Java is done (Score:2)
So you're not privvy to what goes on upstairs. Go find somebody in the know, get them drunk, and ask them about massive scaling and Google's patents on map/reduce.
In the meantime this seems like a good idiological fit. The surveillance-funded corporations will be taken care of while the USG destroys the software industry, which is too wildly successful for a completely unregulated market. Nimble big-name companies have already fled or are in the process of fleeing the jurisdiction, leaving work-a-day pr
Re: (Score:3)
That's retarded. Oracle had (and maintains) a sizeable investment in Java and the rack servers for which Oracle is optimized for.
The acquisition was about securing the investment. Not any devious scheme.
Disclaimer: I work for Oracle but am not in any way associated with the Java group nor am I part of the executive/decision-making chain.
Your point of something being retarded is aimed in the wrong direction. "Securing" would mean that they originally owned it, but they didn't. They purchased Sun and immediately started legal actions which Sun was never going to pursue because they knew they had open sourced Java. In fact in the Google vs. Oracle case numerous messages from Sun came out expressing exactly that, which is why the first Judge ruled for Google. The Judge also understood the sheer idiocy of Oracle claiming patent and copyrigh
Re: (Score:2)
It's my opinion that Sun had a case but no the money to follow suit, until Oracle bought them.
But wasn't the whole thing about some private APIs that Google (or whatever was that company it hired) made use and actually copied verbatim?
IANAL.
And by "securing", I meant "securing it's investment in the technology", not ownership of Sun itself.
Oracle DBs and Apps make substantial use of Java. Had Sun been allowed to falter, or worse yet be bough by a (then) competitor like IBM, it would have been disastrous for
Re: (Score:2)
The only way to come to your conclusion is to ignore facts. You can go read the original decision and evidence which accompanied the decision. No, you don't need to be an attorney to figure this out.
But wasn't the whole thing about some private APIs that Google (or whatever was that company it hired) made use and actually copied verbatim?
No, again you can go read the decision and evidence (which includes the charges from Oracle against Google). It was one of the most open Civil cases I have ever seen.
Oracle DBs and Apps make substantial use of Java. Had Sun been allowed to falter, or worse yet be bough by a (then) competitor like IBM, it would have been disastrous for Oracle.
At best a straw man, at worst complete horse shit. IBM does not run around suing people over bullshit like this, and _IF_ they had bought Sun t
Re: (Score:2)
Someone claim (C) on something oracle depend on (Score:5, Insightful)
Like libc, or whatever, or change licenses to an "Oracle Exclusion License" so stupid things like "Copyrighting APIs" get dropped and common sense rules again.
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
The Open Group claims the copyright on the POSIX specifications. If APIs can be copyrighted and this copyright includes all implementations, then it would be problematic for all open source *NIX systems.
Only if The Open Group were acquired by a company with malevolent intentions. A company such as, oh say, Oracle.
Of course, they might decide to provide a license that's valid for everyone except Oracle (though writing such a license in a way that's GPL compatible would be very hard, so glibc might be in trouble).
How is glibc in trouble? Oracle doesn't have the copyright for it.
Wasn't Java open sourced? (Score:3)
Re: (Score:2)
By open sourcing something you don't necessarily give up your copyright claim. You're giving everyone a license to use and modify your code under certain conditions, and those conditions can be whatever you want based on your license (GPL/Apache/BSD/Whatever)
If you put the code in the public domain, you give up all claims to copyright.
Re: (Score:2)
Re:Wasn't Java open sourced? (Score:5, Informative)
Re: (Score:3)
Re: (Score:2)
Re:Wasn't Java open sourced? (Score:5, Informative)
Here is what the copyright case was about:
http://en.wikipedia.org/wiki/O... [wikipedia.org].
No matter what Google may have copied, according to Oracle's own court case and allegations, they did not create a single API by copy-and-paste, as you allege.
The only "actual Java source code" copied is these nine lines:
But, actually, these lines pretty much follow from normal Java programming conventions; whether or not Google actually copied them, they should not be covered by copyright law since they are not creative.
So, in different words, you're a liar.
Re: (Score:2)
Didn't Google take parts Apache Harmony as base?
Re: (Score:2)
Re: (Score:3)
Because open sourcing has nothing to do with giving up copyright. Sun didn't give up their copyright when GPLing it and Orcale now holds it.
Re:Wasn't Java open sourced? (Score:5, Insightful)
Re: (Score:2, Interesting)
The idea that Java was ever free was rooted in Sun's original promises about making Java an open standard. But Sun dragged that process out for years and eventually just reneged on it, all the while filing patents and making sure that they owned the platform completely. Sun probably did that because they had learned that open and free platforms like Linux and X11 were eating their lunch.
Because a lot of FOSS developers had counted on Sun's promises, by the time it was clear that Java was going to remain pro
Mini Sample (Score:5, Insightful)
Of what the TPP is going to do.
Re: (Score:2)
Re: (Score:2)
Trans Pacific Partnership
Re: (Score:2)
Of what the TPP is going to do.
Or perhaps the TPP people don't want the SCOTUS ruling on this, because it could provide a constitutional basis for challenging the TPP.
Nah, nevermind. That couldn't be the motivation of The Most Transparent Administration in History.
Justice Department? (Score:5, Insightful)
WTF do they have to do with this case? This isn't a criminal proceeding, it's a civil matter.
This isn't about "protecting" Oracle (though there may be some $$$ influence involved), but rather more about protecting the copyright racket, strengthening it beyond the accepted scope.
APIs should not ever be copyrighted. Once you start doing that, it's only a matter of time before Disney copyrights all cartoon renderings of a mouse, or Nickelback gets to copyright all formulaic/generic rock.
Unfortunately... the Justice Department, likely at the behest of the White House, is intervening to influence copyright law and give corporations even more power. Ugh. It's like our government is pushing to see how far it can go to enslave citizens (the real, human kind, not the corporate nonsense kind) before they decide they've had enough of this shit.
I'd be inclined to chalk this up as a "First World Problem" but clamping down on technology denies everybody equal access. This is a serious infringement of our freedoms that will have a chilling effect on the progress of technology to help people in their daily lives everywhere in the world. It's not just about Java - it's about any programming language interface.
Re: (Score:3)
WTF do they have to do with this case? This isn't a criminal proceeding, it's a civil matter.
The Supreme Court asked the government to comment, and so they did.
Re: (Score:2)
Where do you see anything about permission. They asked for *comment* - just because you ask someone for advice, doesn't mean you're bound to follow it.
Re: (Score:2)
You have already answered your own question:
The US government has more or less become the enforcement arm for the copyright lobby.
Which means they are now advancing copyright/corporate interests.
People who understand APIs understand it's a published contrac
Re: (Score:2)
WTF do they have to do with this case? This isn't a criminal proceeding, it's a civil matter.
I've absolutely given up on trying to figure out how our government works. Apparently SCOTUS specifically asked them for their opinion. And here I thought our founders wanted separation of powers, and in particular that the Supreme friggin' Court was intended to be insulated from the ebb and flow of political interests in the other two branches.
I could understand if the Justice Dept decided to file an amicus brief just like everyone else, but I can't understand why SCOTUS specifically sought their opinion.
Re: (Score:2)
Anybody who wants to know what these cases are about can just read the Wikipedia article and find more information on the web:
http://en.wikipedia.org/wiki/O... [wikipedia.org].
https://majadhondt.wordpre [wordpress.com]
Re: (Score:2)
The Groklaw archives have a good discussion of the original case heard by Judge Alsup.
Re: (Score:2)
You clearly haven't read the brief, or even have a basic understanding of the case (hint: it is about APIs being copyrighted).
The rest of your post is just complete bull. I wish I had mod points today...
Java API: Copyrighted, but hope for fair use! (Score:5, Informative)
TL;DR: US executive shares the appeals court opinion that APIs are copyrightable, but that does not mean the copyright is enforceable - there will be another court case that will be about if it's fair use to re-implement the (copyrighted) API.
Here is maybe the most important paragraph (italics mine):
The brief is quite well readable (modulo the awful scribus ui), try it!
Re: (Score:2)
Java API: Copyrighted, but hope for fair use!
HOPE for CHANGE. Where have I heard this before?
Re: (Score:2)
Probably the "bright line" copyright distinction between APIs and actual works of art should come from the legislature, but our Congress is just as technologically illiterate as the judicial and executive branches.
Maybe in another 20 years we can have laws that actually bring us in to the 21st century.
Re: (Score:2)
I think you give Congress too much credit. The executive branch, at least, has folks who have to actually work with technology.
Re: (Score:2)
The SCO dementors (Score:2)
Once they declare the API to be copyrightable, the SCO dementors will arise and attempt once more to cast darkness over Linux. Darl McBride must be snickering as he reads the Justice Department's response...
Re: (Score:2)
If you go back and read the extensive coverage of the original trial, you would discover that you are arguing for the opposite of what the original court decided.
The decision was that the API _interface_ itself is not copyrightable, but the implementation behind that interface was. That is what open source (and any sane closed source) software project requires. The judge took the time to learn how to program Java as part of his research, a
is there an undeclared interest here? (Score:2)
Just asking, because it seems to me as if someone wants this matter to be declared settled as is and for no good reason other than to guarantee a payout.
Supreme court to DOJ, Challenge Accepted (Score:5, Funny)
DOJ: We recommend you don't take this important copyright case.
SCOTUS: Oh really, why is that?
DOJ: Corporate interest mostly, we are looking to create a new form of monopoly power, and Larry Ellison has some really cool Sailboats.
SCOTUS: Thanks for your recommendation, we are looking forward to hearing this case and just added it to the docket.
Re: (Score:2)
One can hope.
legacy of this administration (Score:2, Insightful)
Looks like Obama's primary legacy may be to enshrine API copyrights in law. If this had been the law of the land in the 80's and 90's, Linux and FOSS would never have gotten off the ground.
Re: (Score:2)
If this had been the law of the land in the 90's, the BSD case would have gone very differently. So no Linux and no BSD.
Explaining API copyright to lawyers/judges (Score:2)
You're honor, you probably don't want to read the case. "Why not. It's a matter of public record". Yes, but the index is copyrighted. It's $5000/copy. Good luck finding the case without the index. "$5000 per copy? That's preposterous. Indexing is trivial compared to the arguments in the case". Why yes, yes it is...
Re: (Score:2)
Microsoft licensed Java from Sun, but there was problem that compliance to license meant passing compatibility tests so Sun later sued Microsoft. Microsoft paid their way out of that situation but later dropped J++ itself. It reused some of that tech in other products.
http://en.wikipedia.org/wiki/V... [wikipedia.org]