Court: Oracle Entitled To Copyright Protection Over Some Parts of Java

An anonymous reader writes "Remember the court battle between Google and Oracle? It's the one where Oracle claimed Android violated Oracle's patents and copyright related to Java. Oracle thought they deserved $6 billion in compensation, but ended up getting nothing. Well, it's still going, and the tide is turning somewhat in Oracle's favor. An appeals court decided that Oracle can claim copyright over some parts of Java. It's a complicated ruling (PDF) — parts of it went Google's way and parts of it went Oracle's way — but here's the most important line: '[T]he declaring code and the structure, sequence, and organization of the 37 Java API packages at issue are entitled to copyright protection.' A jury's earlier finding of infringement has been reinstated, and now it's up to Google to justify its actions under fair use."

Comments

lesson to be learnt

[Anonymous Coward]

There is a lesson to be learnt here: Never depend on programming language, which is not under appropriate free license.

Apache Foundation, do you hear me?

Re:lesson to be learnt

[MightyMartian]

I can't wait for phone books to be copyrighted.

Re:lesson to be learnt

[phantomfive]

This has nothing to do with using Java. It has to do with implementing your own incompatible version of the language. If all you want to do is use Java, or implement a compatible version, the license is good and you will have no problem.

Re:lesson to be learnt

[Rich0]

This has nothing to do with using Java. It has to do with implementing your own incompatible version of the language. If all you want to do is use Java, or implement a compatible version, the license is good and you will have no problem.

The problem with this is that you now are subject to somebody else's sole determination that you are/aren't "compatible" or else you get stuck in endless litigation.

If you build entirely on free-licensed components, then nobody has any control over what you do.

Re:lesson to be learnt

[king neckbeard]

That's incorrect. With Java, you have a few options: 1) Fork the GPL version of Java. In such a case, you can have absolutely no compatibility and still be in the clear. 2) Write your own Java implementation, but have it meet the standard. I'm pretty sure what is and isn't compatible is laid out pretty clearly. 3) Pay Oracle enough for a license to do what you want.

Re:lesson to be learnt

[phantomfive]

The problem with this is that you now are subject to somebody else's sole determination that you are/aren't "compatible" or else you get stuck in endless litigation.

The test suite that tests for compatibility is now open source (although the licensing is messy, it is ultimately usable), so that isn't a problem anymore. It was a problem before, you are right.

Re:lesson to be learnt

[symbolset]

The test suite is under a different and nonfree license.

Re:lesson to be learnt

[ADRA]

Tell that to Apache Harmony. Oracle refuses to make the certification 100% free / license un-encumbered, so here we are. You can't release a 'Java' runtime without certification, and you can't be certified unless you sign contracts with Oracle to bend over the barrel.

Re:lesson to be learnt

[ADRA]

1. Isn't actually true. You need to stay well within the lines or draw the ire and lawsuits from Oracle
2. Isn't true because you need to to be licensed by Oracle in order to be verified as 'compatible' and if they say no then guess what?
3. Yup, that's pretty much the only route you have

Re:lesson to be learnt

[HiThere]

They needed it because they were trying to speed up execution on smart phones. As I understand it, Dalvik(?) compiles class files produced by the javac compiler to optimize register allocations, and what they are arguing about is the documentation. They aren't trying to use the same name, so it's not related to the MS ploy of defining a non-compatible Java. They're trying to define a (very large) subset of Java+libraries that can be handled by their compiler.

OTOH, it's been months since I paid close attention to this, so I may well have the details considerably garbled. IIUC, however, the question was whether the order in which class methods were listed in the documentation was copyrightable. Originally the answer was no, but it sound like that answer has been changed to yes.

Re:lesson to be learnt

[DrJimbo]

If all you want to do is use Java, or implement a compatible version, the license is good and you will have no problem.

This is completely false. Oracle changed the rules around for what it means to be "compatible" so that only projects that Oracle likes will be deemed compatible. Apache is being forced into a Java Fork:

The problem's core is that first Sun, and now Oracle, won't give Apache a chance to certify Apache's Project Harmony as being Java Platform, Standard Edition (Java SE) compliant.

Apache: I know my rights. I want my compatibility certification!

Oracle: How can you get a certification if you can't take the test?

Oh PJ, where art thou?

[pegr]

If ever there was a time we needed you... :(

Re:Oh PJ, where art thou?

[phantomfive]

Check out the ruling yourself, it's surprisingly readable and will make you smarter.

Re:Oh PJ, where art thou?

[maroberts]

It may be surprisingly readable, but Justice Alsups original ruling was extremely clear cut, no nonsense and demonstrated a clear understanding of technical issues and accepting Oracles arguments. Accepting it also would have also limited or prevented a lot of litigation in the future.

Re:Oh PJ, where art thou?

[phantomfive]

It may be surprisingly readable, but Justice Alsups original ruling was extremely clear cut, no nonsense and demonstrated a clear understanding of technical issues and accepting Oracles arguments.

It may have been clear cut, but he was wrong. The present ruling explains why, read it.

Essentially: 1) He muddled his logic between 'copyrightablity' and 'fair use,' two different concepts.
2) He didn't respect previous rulings and standard court procedures (such as the Abstraction, Filteration, Comparison test).
3) A lot of his logic rested on the Lotus case, which isn't commonly used as a precedent, and isn't relevant to this case anyway.

Realistically there's no reason to believe that APIs aren't copyrightable. Do you deny that building an API requires a lot of creativity? Making a good API is hard work, and deserves protection as much as music does. This isn't even a particularly important question (whether anything at all deserves protection is another issue; here we are talking about what is currently legal).

The more important and relevant question is whether the Google usage falls under fair use. If it does, then they can use the API anyway.

Re:Oh PJ, where art thou?

[kqs]

Oh dear. This could be a problem for anyone who uses a C compiler. Who owns the original C APIs anyways? Dennis Ritchie maybe?

I wonder who owns the SQL APIs? Oracle may have a problem here...

Re:Oh PJ, where art thou?

[phantomfive]

You obviously haven't been paying much attention to copyright law.

Re:Oh PJ, where art thou?

[WaywardGeek]

Thanks for the link. I read: The jury found that Google infringed Oracle’s copyrights in the 37 Java packages and a specific computer routine called “rangeCheck,”

Fuck rangeCheck. I don't care if Oracle gets $1B for that stupid 10-line function that any moron could write in 5 minutes. Oracle succeeding in copyrighting an API, which last a freaking 100 years, is death to our industry.

Coder Boycott

[RichMan]

Ok this ruling would seem to invalidate any ability to reproduce any interface.

This needs a coder boycott of anything Oracle until Oracle stands up and pubclically disavows this ruling and claims the court was wrong.

Re:Coder Boycott

[RichMan]

The ruling means that any library in any language can be shut down. It means that public interface declarations can be copyright. It means it could be impossible for anyone to reproduce a public interface.

AkA it makes all public interfaces private. It is not just a Java specific ruling, it has implications across all coding environments.

Re:Coder Boycott

[phantomfive]

The ruling means that any library in any language can be shut down.

No.

It means that public interface declarations can be copyright.

Yes.

It means it could be impossible for anyone to reproduce a public interface.

No, because if you are reproducing a public interface for compatibility purposes, it is fair use. There are lots of ways something could be fair use. Even if Google fails to show that this particular case was fair use, that won't prevent the fair use argument in other cases.

In reality, purpose matters for fair use. If your goal is to reproduce the public interface for compatibility purposes, that is fair use, because that is the only way compatibility can be reached. However, the goal of Google here was different, it was to make the life of programmers easier on Android by presenting them with a familiar environment. It will be interesting to see if they can defend that as fair use.

Please don't make the mistake of thinking this will kill Wine or all programming languages or something. There's already enough irrational hysteria on the internet.

Re:Coder Boycott

[phantomfive]

This needs a coder boycott of anything Oracle until Oracle stands up and pubclically disavows this ruling and claims the court was wrong.

Hey, I've been boycotting Oracle for the last decade on grounds that their software is garbage! Count me in!

Re:Coder Boycott

[Kaenneth]

It WILL have a huge 'Chilling Effect' though, even if you are in the right, and it is fair use, can you afford to defend against a huge corporation suing you?

Re:Coder Boycott

[reg]

Don't be naive. This will be used to shut down APIs. Increasingly the software world is a set of web based and hosted APIs, with big money but little business behind them. Imagine, for example, someone like Snapchat copying Twitter's API to enable their service to grow faster. This ruling, it is stands, will be used by incumbents to shut down start-ups or open-source/non-spyware clones.

Probably Google's biggest mistake at the get go was to not do a /Java/Davlik/g. Since all code needs to be recompiled, this can be done easily by the build system while maintaining a single source file...

Regards,
-Jeremy

Re:Coder Boycott

[Kaenneth]

No, it makes the difference between a frivolous suit which will get their lawyers sanctioned, and paying your court costs VS 'We thought it was reasonable given this case law.'

Re:Coder Boycott

[blackiner]

The mere fact that this case even exists is already proof it will be abused. And please, if the copyright regime has taught us anything, it is that fair use will continue to get shat upon.

Re:Bye-Bye Java

[BronsCon]

Sure, if your entire userbase is on Windows.

Re:Bye-Bye Java

[i kan reed]

Mono is a thing you know. A thing not being sued by Microsoft.

That's some crazy shenanigians right there.

[HeckRuler]

API calls.... their "structure, sequence, and organization" are copyrightable? API calls... really?

So I have a door, it has a door bell. It follows a nice standard that if someone wants to get my attention at my door, they ring the doorbell. If they want to leave a message, they put a note in my mailbox. That's a rough equivalent for my house's API. I can copyright that (baring prior art and fair use)?
The contents of my house are my own of course, but the procedure I ask everyone to follow when coming into my house? I also own that?

Really guys?

Re:That's some crazy shenanigians right there.

[dfsmith]

As I read it*, the argument is over the 37 verbatim copied headers that define the API. That's like Oracle making a beautiful (ahem), elaborate sign explaining how to ring their doorbell. Google made their own doorbell but copied the sign, embellishments included. While the content of the sign is "fact", the decorations are arguably product of a creative process.

While I'll have to wait for better analyses of the ruling, I think we can take away that if you're reimplementing a library, you might want to reimplement the headers too.

* IANAL, and I'm not speaking for my employer. I only scanned the ruling.

Re:That's some crazy shenanigians right there.

[k8to]

There have already been rulings that decided that headers that define a public api are not under copyright if they represent the only way that that public api can be declared.

In other words, this judge did not follow precedent, or they're in different jurisdictions (I don't actually know).

Copyrighted buy who cares?

[tomhath]

Federal Circuit Judge Kathleen O'Malley wrote. "On this record, however, we find that the district court failed to distinguish between the threshold question of what is copyrightable — which presents a low bar — and the scope of conduct that constitutes infringing activity."

Does this mean that even though Oracle can copyright something (not sure what), Google might still be able to use it without infringing? That's what it sounds like to me. And it took a whole lot of wasted money for Oracle to barely make it over the "low bar".

Re:Copyrighted buy who cares?

[phantomfive]

Yes, it must now be decided if Google's actions fall under fair use.

Results

[phantomfive]

In the original trial, the jury found that Google had infringed on the Java API (37 API packages including the declaring code and the structure, sequence, and organization). Shortly thereafter, the judge ruled that those things were not copyrightable, thus Google didn't need to pay.

Now, the appeals court has reversed that, and said that those things are copyrightable.

Because the original jury was deadlocked on the question of whether Google's copying was fair use, it needs to go back to trial. But only the fair use will be considered in that trial, not copyrightability.

Re:Results

[zarr]

...which nicely demonstrates how hard it is to come up with good analogies. If I wrote a book, and you copied all my chapter titles for your book, be certain that I (or rather my publisher) would come after you. I put a lot of work into those titles after all.

APIs are different though. They are meant to be copied. You can't use them without copying them.

Re:Bye-Bye Java

[BronsCon]

And the implementation is 100%? Nope. With Java, though, I get everything Java has to offer, anywhere Java is available. Maybe platform consistency isn't important to you, but it matters to some people.

From a purely logical standpoint, Java wins if you don't want to have to double-check whether each of the APIs you're about to use is actually implemented on all of your target platforms. From an idealistic standpoint, yes, I can see why someone would avoid Java (and, in fact, I have managed to do so for the entirety of my career, thus far), which is why the Mono projects exists, and why it is important. However, it's just not there yet, from a logical perspective.

Re: Bye-Bye Java

[VTBlue]

Umm if you actually check mono's compatibility notes, it has ridiculous good compatibility distinguished between the various versions of .NET. While there will always be a lag, if you develop with Mono, you know what works and what doesn't. .Net 3.0 and 3.5 are pretty mainstream and 4.0 is pretty much good to go for a broad set of use cases.

Re:They will keep trying until they get a judge th

[tomhath]

This is about copyright, not patents

Re: Bye-Bye Java

[BronsCon]

There's a gap between "ridiculous good" and "perfect" that's simply too big, just by the way of its existence. Add to that, most consumer users will have Java installed already; how many will have Mono? Not such an issue if you're talking about Windows users, as they'll likely have the .Net libs installed already, and the smallest subset of desktop users, desktop Linux users, will be able to fetch Mono from their distro's repository (and hopefully it's a recent version), so the #1 and #3 groups are covered. The #2 group, however... You don't expect the average Mac user to track down the Mac port of Mono, do you? And I'm asking this as a Mac user.

Until it's there by default, It's simply unacceptable for consumer software.

Not Getting the Strategy Here

[Stormy Dragon]

I don't get why Oracle bothered to buy Sun since they seem to be systematically destroying the value of everything they got from the purchase.

Re:Bye-Bye Java

[OhPlz]

"This new Microsoft has not only removed the problematic restrictions on its licenses, but also worked with Xamarin to solicit design feedback, and published documentation under a Creative Commons license so that it can be redistributed."
http://arstechnica.com/informa...

It's bipartisan

[tepples]

Do you have any questions which political party

Both major U.S. political parties have shown themselves to favor expansion of the exclusive rights of copyright owners. See the No Electronic Theft Act, the Copyright Term Extension Act, and the Digital Millennium Copyright Act.

Wrong, wrong, wrong

[reg]

This is a very bad decision and is only going to harm the software industry. This is Google's fault for using the wrong arguments. APIs are digital forms. You fill one in and give it to a worker, it does what you asked (possibly with side effects) and returns results. This is not an analogy, it is a fact. Forms are not copyrightable, for good reason. Imagine if every bank had to make up a new name for a 'deposit slip', and someone could copyright "First Name, Last Name" on a form! Google copied Java's API, the same as businesses have been copying each others forms since the dawn of time, and for the same reason: its easier to present a known interface to customers.

Regards,
-Jeremy

Re:Wrong, wrong, wrong

[reg]

I've followed trial very closely, and I read every line of the court transcripts of the original trial, although not every exhibit or submittal. They did not make this argument. They also did not make it on appeal, as far as I know. ("The parties have not disputed the district court's analogy: Oracle's collection of API packages is like a library, each package is like a bookshelf in the library, each class is like a book on the shelf, and each method is like a how-to chapter in a book." pg.7) Instead they used poor analogies for what the API is and does, and allowed it to be defined badly.

Having finished reading the ruling... pg 28 is about the doctrine of merger (expression being dictated by idea). That's not what I am talking about. But it does discuss the issue I suggest on pg. 19. I'm talking about 37 C.F.R. Â 202.1(c). The only reference I can find to that statute in this case is in http://www.groklaw.net/pdf3/Or..., where the idea of blank forms is only tangentially mentioned. There was some discussion of Baker v. Selden, but mostly in the context of the SSO of the API. Oracle actually concede in that that the individual method specifications are like a blank form, but not explicitly. Google never picked that up.

*plonk*

Regards,
-Jeremy

Analogies are Killing Us

[Mateo_LeFou]

You wouldn't steal a car, snatch a purse, etc. So why would you reproduce the sequence and structure of an API !? !

Re:Bye-Bye Java

[BronsCon]

.Net isn't consistent from one version to another, either, so I might be missing your point. You target a version of the Java API just like you target a version of the .Net API; the difference being that you can trust the implementation of a given version of the Java API to remain consistent across platforms, whereas on non-Windows platforms, how consistently a given version of the .Net API will be implemented depends on which version of Mono the user has installed.

I'm sorry, but I prefer to be able to debug my application against a known system, rather than an array of unknowns.

Legally correct decision with awful results

[Anon E. Muss]

(I actually read the court ruling before posting this)

tl;dr version: The results will likely be awful, but the decision appears legally correct.

Google won at trial because the judge decided that the Java API was not copyrightable. I absolutely believe that API's should not be copyrightable, but that isn't what the law says. Copyrightability has a very low threshold. The trial judge screwed up by applying legal standards related to fair use to the question of copyrightability. The appeals court was correct to reverse.

The case now goes back to the district court. There will be a new trail with a new jury, but the only issue will be whether Googe's copying of the Java API is fair use. The original jury deadlocked on this question. Fair use decisions are very subjective, so it's hard to predict how this will turn out. All I can say is that I hope Google wins.

P.S. None of this decision was related to patents. Oracle lost on their patent claims at trial, and that stands.

Re:Legally correct decision with awful results

[Garfong]

It kind of looks like the appeal court decision is setting this up for a potential appeal to the Supreme Court. The decision repeatedly talks about differences and conflicting rulings between different Circuits, which is one of the things the Supreme Court looks for when deciding whether to approve an apeal.

IBM Should Be Warming Up

[StormReaver]

If this ruling doesn't get struck on appeal, IBM's lawyers should be drooling oceans as they warm up to sue Oracle for copyright infringement on SQL. Oracle owes IBM many billions of dollars in infringement, by Oracle's own logic.

This is yet another judge that is completely incompetent for the job.

Re: Bye-Bye Java

[lgw]

If you're targeting the "non-windows, non-mobile Linux home PC user consumer", then, yeah, Mono sucks for that user base (both guys!).

But for the interesting consumer Linux market, which is to say Android, Xamarin has it sorted. I was skeptical of that dev environment for a long time, as the legal situation with Mono seemed unclear to me (even though they're in the right, MS could still sue to be a nuisance). But all that recently changed with an official MS-Xamarin partnership.

C# is a joy to work in compared to Java (and I've spent years writing in each professionally), and now the legal issues flow the other way - MS is partnering while Oracle is suing.

Re:You can't copyright facts

[pegr]

I can be creative in solving a math problem, but the expression itself I create is purely functional. You cannot copyright a process, only creative expressions.

Re:Bye-Bye Java

[benjymouse]

Name a platform that is end-to-end not proprietary in any way shape or form?

Even if such a platform exists, how does that preclude Microsoft from suing? Remember that the thesis here is that Microsoft would disregard the licenses already granted for C#, .NET Framework, compilers etc and just sue to exhaust your funds. Why couldn't they claim that you infringed an algorithm (or whatever) even if you were using Java or Python? After all, they have no legal standing but are considered *so* malicious that they will sue even when they have no legal standing.

The whole "Microsoft will sue!" is nothing but FUD.

In reality - because of the promissory estoppel of the community promise - users of .NET and any other technology under the community promise is much better protected than when using alternatives. This is because the promissory estoppel can be used to dismiss a lawsuit outright.

The judges don't make the rules

[Chirs]

they only interpret them.

If the law was written such that APIs can be copyrighted, there's nothing the judge can do about that.