“Oracle’s case against Google has evolved primarily into a copyright infringement suit over the past several months, and with the full trial scheduled to begin this coming Monday, the court is making an effort to get down to the nuts and bolts of copyright law. The judge issued an order last week requiring that both Google and Oracle provide their respective positions on a fundamental issue in the case: ‘Each side shall take a firm yes or no position on whether computer programming languages are copyrightable‘.” Seems like an easy enough answer to me, especially since Oracle’s example doesn’t hold up at all – Oracle points to Klingon’s custom glyphs to illustrate that a language can fall under copyright, but unlike Klingon, a programming language uses standard glyphs we all use every day. Arguing you can copyright that is borderline psychotic, and opens up a whole can of worms.
Wouldn’t that make the inventors of C able to enforce copyright infringement of the C programming language, effectively bankrupting the entire industry?
If only Dennis Ritchie was still about to test this theory
But who could out-asshole Larry Ellison who wasn’t killed by their own pancreas?
Dennis Ritchie’s children got the copyrights and Brian Kernighan is still alive.
And since most opensource programming languages include only licenses fro the code, I’m pretty sure that Google could assert copyright on Python against Oracle via their own Benevolent Dictator for Life Guido van Rossum or something similar.
We’ll just all go back to writing machine code and they can all put their copyrighted languages where the sun don’t shine. IIRC object codes are not copyrightable.
Thing is, if programming languages are copyrightable, then every coder performs (c) infringement every time writing down a word of that p. language, it doesn’t even need to do anything meaningful. Oh, imagine you reading out loud a line of code for a friend sitting across the table and a few other people hearing it. You should all pay damages of 9 gazillion silver dollars for each word of Java you ever used digitally, in writing, or in speach.
Machine code can also be called a language too, just harder to be read by humans. =(
I agree, that it can be called a language, same as a higher level one, but if I remember correctly (would make some searches to be certain) machine code is not copyrightable, since (see copyright.gov) object code registrations “are made under the Office’s rule of doubt since the Office cannot determine with certainty the presence of copyrightable authorship”.
Ya, that’s right. But the language that you will use to craft your code (if Oracle is right) can be subject to a copyright.
One thing is the code done in machine language (that you can’t copyright), another thing is the language itself.
For real? Klingon?? Their example is a fictional work of art with a disputed copyright? Wouldn’t it have been better to point to a constructed language with an upheld copyright?
Yeah, doesn’t sound like a great plan to me.
Thats the problem, I don’t think there is a constructed language that is protected by copyright. Languages really seem out of the bounds of copyright. Can you copyright grammar, syntax or vocabulary?
Hmmm…you are right. It seems they can’t. I remembered it differently but there you go.
You can’t copyright language.
Languages falls in the same realm of mathematical formulas, and indeed, is a mathematical construct itself. Indeed, in a programming language, this is even more obvious, since its nothing more than a high level mathematical notation to describe logical procedures to a turing-like machine, who is nothing more than a form of calculator.
That won’t be known for sure until someone actually manages to get a definitive court ruling. Paramount considers Klingon to be their property as a work for hire, and there are even a few Native American groups who are claiming copyright over their languages.
Agree wholeheartedly. If you’re using a fictional language as a basis of your legal defence, you know Oracle doesn’t have a leg to stand on. If they show us an actual, oh I don’t know… PROGRAMMING LANGUAGE as a precedent example, then maybe they wouldn’t get laughed at.
Should the argument of Oracle be considered valid they would very much be on hot waters as everyone else. There is a long string of “prior art” on computer language development way back to FORmula TRANslator and back again to standardization of commerce/production practices.
This is so ridiculous. That is what we get when lawyers take on subjects they don’t understand on its fully. Back off!
This situation is potentially a lawyer’s wet dream. Think of how much money wold get pumped into the legal profession if a judge could be fooled into oracle’s argument? Just the confusion alone i worth a lot of money. The more people fight the more money lawyers make. Nice system isn’t it?
Hush, or some economist would find your ideas to be the ideal way to get the economy back on tracks. Let’s all start a litigation bubble!
Sadly, the problem is that the lawyers understand both the law and the technology too well. They just don’t care if their actions destroy an industry because it isn’t theirs.
Many lawyers actually believe they are doing the right thing. If you take out the political/philosophical debate over what the law should say, and just look at what it does say, then a lot of what the lawyers argue makes sense.
It can be argued then, that if what they say follows the law, and following the law would kill an industry, then perhaps that industry should not exist anyway. After all, “the law” has existed for centuries and people have done well following it. On the other hand, very few industries today are that old, and industries tend to shaft a lot of people as they get older anyway. (Probably why the oldest industry still in existence spends just about all of its time “shafting” everybody involved and doesn’t seem to care what they law says anyway.)
In any case, don’t blame the lawyers. Blame the greedy business and technology school grads, or the corporate investors, who want the lawyers to do these things. And before you say the lawyers never refuse, I have personally seen a grown man cry because he had to refuse a client with a practically unlimited litigation budget because the client wanted something that was completely frivolous.
Actually the law is a lot younger than most professions/industries. Take hunting, farming and similar activities that are actually catered by the law, the law rarely defines these industries/professions.
You can relax, the lawyers are scum because they are trained to stow away their personal beliefs for their clients’ needs.
And in this particular case, I beet it’s the lawyers that came up with the idea of language copyright. The business person isn’t that educated in legal matters.
Cool, so when do they start paying royalties to IBM for SQL?
I have not read either brief, so I’m not sure exactly what is being argued. My facts could be wrong as they are taken from the article. Heh, I’m not even certain if it is Oracle or the article that makes the Klingon argument.
By the way, does Oracle have a language programmed in Klingon?
Joking aside, programming languages are not native languages. They do not spring up from a collective effort. It takes somebody, or a team of somebodies, a long time to author the grammar that defines a language. That grammar is a set, defined written thing, just like a book or article. This sounds like it could be copyrighted and entitled to copyright protection.
Further, even if the base grammar isn’t copyrightable, what about all those standard libraries that come with the language by default. A lot of work went into those, they are as written as any piece of software or paper I have ever written. If you cannot copyright your libraries, then how can you copyright your third-party libraries, or even your final application.
Oracle argues that Google copies quite a few of the base libraries too. It seems like Google is arguing that the standard default libraries are not copyrightable either. Honestly, Google seems to have the weaker argument.
Rather than see this as an evil that must be fought, just don’t use products from the company that treats its customers badly. Though I’m not certain who that is in this case.
Java was never as open as a programming language should be. Perhaps we should just use this as a Stallman-esque lesson in the virtue of only using actual open sources.
A) They do make the Klinglon argument.
B) Are you aware of the history of Java syntax? B(1969), C(1973), C++(1983), Perl(1988), JavaScript(1995), Java(1995), C#(2001) are all related in syntax. They very much are a result of “a collective effort”.
C) The argument isn’t that the actual implementation code of the base libraries isn’t copyrighted, it’s that the API(Application Programming Interface) is copyrightable in general. That is a case that all sensible software developers would like to be ruled in favour of Google.
Oh, yeh, because Java structure, grammar and syntax is so different from everything that predates it and it encompass new symbols and words not seem before. Not!
.
Like the IO? Or the Math?
Keep in mind that many parts of Java were released under an open source license or were published under a standard body to everyone see. This greatly lower their instances. The interface was made public, not the implementation. To me it is like publishing what is in your recipe to cook something, without telling how to do so. Google did not copied the implementation, what could bring copyright issues, just used the same name, arguments and, possibly, issued the expected results on code return, what is the sane thing to do if you want to keep interoperability.
I disagree with your stance about Google having weaker arguments, but, as we all know, our law system is not about what is reasonable but, instead, about what is set on letters. It is a common mistake to think that the Justice System is about to be fair, one that is repeated by judges and lawyers to further the authority of the decisions achieved, it is a myth.
I really hope Oracle get what they deserve: a huge bill on legal costs and nothing more.
The lawyers, though, will get paid anyway, unluckily.
This is incredibly surprising. I thought ALL languages were absolutely NOT copyrightable. It seems like the concepts ‘Freedom of Speech’ (for all forms of Speech) and ‘Copyrightable Language’ (in all forms of language) cannot exist in the same legal system because they are polar opposites.
I mean, it makes no sense for a language itself to be copyrightable. It does make sense for the works created using language to be copyrightable. Just like you can’t (or shouldn’t be able to) copyright English, Spanish, Dutch, or any other language but you CAN copyright things made using that language.
Also, as many others have pointed out, there’s so much evidence of prior art in this case that it doesn’t make sense at all for Oracle to even go this route, and it comes off (to me at least) as looking desperate.
Last but not least, this wouldn’t even be an issue if Copyright terms weren’t so ridiculous. There’s no good reason for Copyright to last longer than 5 years. What started out as an artificial monopoly granted to reward the creator for their work and contribution to Science and the Arts has been perverted by the legal system into something that actually stifles Science and the Arts. It’s despicable.
Please correct me if I’m wrong by providing links to relevant material.
IBM claims copyright over a character table, so it doesn’t surprise me if someone claims copyright over a language.
Actually, IBM uses Java a lot. It would be funny to see an Oracle vs. IBM lawsuit over that.
If Oracle can claim copyright on Java then IBM can claim it on SQL. That battle would not end well for Oracle.
I understand your outrage, I really do. I used to feel the same way. Then I went to law school. That is where I learned that lawyers, for the most part, do not “stow away their personal beliefs for their clients’ needs.” Rather, they believe very strongly that they should, like scientists, question everything, even their own personal beliefs.
Often, when you look at the bigger picture, you realize that things you believe with intense ferocity, are not what you think they are. I believe that this issue is one of those cases. As a programmer I am shocked that somebody would try to claim copyright over a language that I have used for years and love. At first, I was angry at Oracle for even making the argument for copyright of a language. Then I thought about it some more. While I still dislike the terms for IP protection, like pretty much everybody else here, I have to begrudgingly admit that protection should apply in this case.
I will try to address all the issues my original comment seemed to spark. If I miss something, please call me on it. I’m not trying to ignore inconvenient arguments.
Prior Art
First, and probably foremost on everybody’s mind, prior art. A lot of the comments I read on this issue are basically arguing that you cannot copyright Java because it is built on so much prior art. Prior art is a non-factor in this case.
A really good IP attorney I met told me, and I’m going to paraphrase the thought a bit, that prior art is the first defense of the legally clueless. In copyright it only matters if the person who holds the “prior art,” which isn’t even the right term, is the person suing. In this case, Oracle has the prior art. Google copied Java. Google has the derivative work, if it can even be called derivative because it copies so exactly.
If this was a patent case, prior art might matter. But it isn’t a patent case. And if it were, a good lawyer would argue prior art last.
Now, if the copyright owners of B, C, C++, etc wanted to sue Oracle for making an unauthorized derivative work (copyright infringement), they might have a case. It would depend on the terms under which those languages were distributed. But they could do it.
Take, for example, “Romeo and Juliet.” There are probably thousands of stories that retell this really old story that was itself probably a retelling of yet another story. I didn’t study literature, so I don’t know. Now, could it reasonably be argued that the writer of “West Side Story” does not deserve copyright because the story was based heavily on the prior art of Romeo and Juliet. You might not think so, but the courts and the law certainly do.
Java, like “West Side Story” adds to what came before. It was obviously different and probably better because so many people adopted it. Which leads to the next argument.
Sun open sourced Java and put the API through a standards body
This argument is very fact specific. It will greatly depend on the open source license they used and the promises they made to the standards body. Standards bodies have poor track records in regards to making their standards copyright/patent free/unencumbered.
This may be Google’s strongest argument, but I do not know enough specific facts to address it. So I will continue with the issue that seems to be irking people the most.
You cannot copyright/patent natural language or mathematical ideas
Copyright protects expression of ideas, not the ideas themselves. Patents protect original ideas. There was no mention of patents being involved, so they do not merit discussion. However, Java is a very specific, written, expression of an idea. It has to be very specific, or it would not work.
Having taken courses in programming language design, I have written my own custom language. It did not spring out of thin air. I took ideas from other languages and put them together for my own grammar. It sucked, but it was mine. Nobody would want to use it, but it was mine. Many other people took the same ideas and created (probably) the same grammar without ever looking at mine. That was theirs, even if it was identical.
However, if another student had copied my floppy disk without my permission (yeah, we still used those back then) then they would be using my language, not theirs. Even new adjunct professors understand this simple concept and would, hopefully, fail the thief and not me.
It was a simple concept, the grammar was the assignment, everybody had to create their own. If it makes sense in the classroom why do people want to look the other way when corporations do it. The Java grammar belongs to Oracle. If they don’t want people to copy it, and they didn’t let people copy it, then they have that right.
So what about the APIs. One comment here mentioned IO and Math. They agreed that the implementations were copyrightable, but not the API. Well, what about the designer who sat down and wrote the API. Software engineers spend more of their time writing the APIs than they do implementing them. There are even whole languages written just to write APIs.
Not how many times “written” was just used. You write a language, even one to write an API. That sounds like an implementation of an idea, doesn’t it?
So, if the grammar and the API specs are protected as written works, what about their implementation. Google does not publish the grammar or the API specs. Assume that they implemented their own version of the language and did not directly copy any code.
Under copyright, it does not matter. This is called a derivative work. Google gets copyright in their implementation of the language (assuming it is different enough to be a derivative work). However, if the derivative work is unauthorized, then it is still infringing. That is not to say that there fact specific arguments on fair use, authorized/licensed use, public domain, etc. Those arguments are where this case needs to be argued.
What about Freedom of Speech?
I’m sorry, but I just don’t see what Freedom of Speech has to do in this case. No person speaks Java, only machines. Extending Constitutional protections to machines is a debate for science fiction and not really applicable at the current time.
Conclusion
A programming languages is a very specific, written expression of an idea. The implementation of that expression is as protected as the implementation of your application. One could argue that losing protection for programming languages could eventually cost protection for all applications. After all, what is Microsoft Word or Adobe Photoshop but a way of expressing yourself
Would any real software developer, who makes their money writing code, really want to lose copyright completely. Maybe they would. That is definitely worth debate. But the courts are not the proper place for that debate. Societies need to make that determination for themselves and make it part of their law. Otherwise the law means nothing.
Bottom line, from a programmers perspective, Java is not a native language. It isn’t even the only option to do what it does. As has been pointed out, there are many other options. So, instead of whining about evil corporations, get off your butt and learn another language that is more free and open. Thank you Mr. Stallman. People didn’t understand before, but they are starting to get it now.
It is very disputed whether java is distinctive enough to qualify as a new language or a derivative work. One that unluckily will be decided on court. To be true, the only distinctive quality of java, when it was “released” to the world, was the promise to write once, run everywhere without the need to recompilation, what it never fulfill. Your examples about what you consider your creation is also a source of dispute, one that we see every day in day out. If you doubt, just “google” for plagiarism. Modifying little blocks or add some does not constitute a new work. The problem is: who will be the arbiter?
Anyway, keep in mind that Sun released Java under GPL on May, 2007, under pressure in compliance with the specifications of the Java Community Process, and guess who was pressing to get that this way, yeh, Oracle. Really, they are going to have a hard time on court about that.
As far as I know, the only thing Google has to fear is the distribution of Java definitions with unauthorized changes (changing license or removing of copyright notices). And as big as it may end, it will never be close to what Oracle hoped for as would be hard to claim deprivation of property from something made public.
I am in agreement with you here. I think this is a very difficult case for Oracle to win on these grounds alone. I just object to the idea that programming languages are somehow fundamentally uncopyrightable material.
It may be arguable whether Java is distinctive enough to qualify as a derivative work. It is probably a question of fact rather than law though.
I remember when Java came out, it certainly seemed different from everything else. It had the promise of write once run anywhere, but the syntax was also quite different. I had to port a lot of code from C/C++ to Java. It was, to my educated eyes, very different. It felt much like smalltalk, yet different from even that. Different enough to give me anger management issues while pounding 20oz bottles of cherry coke at 3am in the lab… those were good times… never had so much fun in my life.
Still, this is probably a moot point because I doubt the case will get past the open source issue.
Unfortunately I find it very common for lawyers to go through a wierd lengthy drawn out process to rationalize away any beliefs they may have had. I have yet to meet a lawyer who is able to directly answer simple direct questions.
That is funny, I know a lot of lawyers who say that nobody ever asks a simple direct question.
Machines understand Java, but it was designed for people writing it, to tell machines what to do. You could argue that only people speak Java.
If you designed any language and you start publishing expressions of it for other people to read and understand, freedom of speech gives those readers the right to also use that language to express ideas. Once they have learned that language and thoughts in that language start forming in their minds, freedom of speech allows them to express those thoughts. If you would be allowed to obtain copyright on a language, that would be very cruel if it led to other people not being allowed to express themselves in it.
Assume authored languages are protected by freedom of speech. Freedom of speech can be limited by time, place, and manner. Copyright, which is also in the Constitution remember, does nothing more than limit the manner of speech that may be allowed for a specific period of time. So even if authored languages are protected speech, they can still be constrained. Otherwise, copyright (which is in the constitution) would be unconstitutional.
Copyright might be cruel, but cruelty is not unconstitutional except as a punishment (8th Amend.). Besides, you would get really divided answers as to what is more cruel, copyright or letting the masses take an author’s work because they want to express themselves in a medium they knew (or should have known) was not theirs when they decided they wanted to use it.
And yes, I know, Oracle made it public and free and told people to use it and blah, blah, blah. That might make a difference in this specific case. Heck, it probably makes a big difference in this case because of how they did it.
But beware na~Avet~A(c). It is dangerous to extend the facts in this case to a much more sweeping condemnation of copyrightability in general. Copyright specifically protected publishing expressions of ideas. How can you refuse protection because an author is “publishing expressions of it for other people to read and understand?”
A) First of all, whenever a person says “I really do” that means that they don’t.
B) Copyrights aren’t granted via standards bodies. Let alone, your should learn what is JCP(Java Community Process). It’s not your average standards body.
C) “Patents protect original ideas.” – Actually not ideas, specific implementations.
D) If you wrote your language, then you know that you write the compiler that understands the input. Oracle claiming copyright over Java programming language would be the worst kick in the nuts for all software developers. Their Sun’s intention was that people would use the language and even the specification from JCP comes with a license to use as you wish by clicking a simple “Yes, I agree”. Now if Oracle claims copyright over the compiler, that is a very strange assertion. Google most definitely did not copy Oracle’s IP by creating their own compiler that understands Java’s grammar.
E) Your argument that it takes a lot of effort to create an API is meaningless. It took a lot of time to discover all of the laws of nature. It takes a lot of time for a builder to build a house. It does not imply copyright protection, per se. Copyrights are also not applicable to insufficiently creative works. For example, I can write an class to contain address information and no logic and looking at it someone could write an exact same copy. That “copy” would not violate copyright, simply because the original was only expressing facts. APIs are like that. Not that they don’t take a lot of time to think about the future, but it’s the same level of creativity as a reporter selecting particular words to describe facts(witch will not be copyrightable). Implementing the APIs will be a different story, though.
F) I agree on freedom of speech.
H) A programming language is a medium, not an expression in itself. In fact, syntax in itself is not even an expression. The compiler is. Word and Photoshop is the tool for expressing yourself, just like Java is, If Java language is copyrighted then your document is copyrighted by Microsoft not you.
Actually, I really do. Take it to mean whatever you want.
I never said that copyright was granted via a standards body, simply that some standards bodies have requirements of anything submitted to them for acceptance.
This is the strongest reason that Oracle has very little case. But only because of Sun’s intention for the language and what they told people to do with it. However, the very fact that there is a license, and that you have to accept it, seems to imply that there is copyright.
Actually, it probably would, but the case is really simple and it would be difficult to prove that the copier had done an exact copy of the original. Independent recreation does not violate copyright, but straight copying does, even for simple things.
News articles absolutely have copyright. I cannot remember the case name off the top of my head, but that was established a long, long time ago.
The language is both a medium and an expression. The compiler and runtime engine are also expressions. Are you saying that Microsoft has no copyright in Word and Adobe has no copyright in Photoshop.
Wasn’t there a case recently where Microsoft argued that the XML language for the new Word/Excel/etc. programs has copyright? Didn’t they win/settle favorably?
I am not arguing that they own the copyright of things you write in their languages. They just own the copyright for the language itself. In the case of both Word and Photoshop you are licensed to use the product to create works under your own copyright.
If you copy either of those two programs without the author’s permission, you are breaking copyright.
“After hours of slaying zombie lawyers in a dark crypt, which I could only access by sacrificing all of my loved ones and making a deal with several demons, I have finally gotten my hand on the patent for the “if” keyword. Now, none shall be able to escape my wrath !”
— Teaser for the solo campaign of Diablo 10. “for”, “in” and “to” are (C) 2012-2112 Oracle corporation Inc.
Edited 2012-04-13 19:29 UTC
as far as I can tell the Klingon “analogy” is something dreamed up for the theverge piece this article links to: http://www.theverge.com/2012/4/13/2944440/google-oracle-lawsuit-pro…
Note it says: “It’s not a perfect analogy, but let’s take the Klingon language as an example …”
I believe that’s the journalist writing: No lawyer is going to start an argument like that!
But I agree with the points made here that Oracle’s claim of copyright protection for computer language is both wrong and undesirable, and if it were true then Java (and C#) would clearly be in violation of the C language.
A pyrrhic victory for Mr Ellison at best then should they succeed.
Klingon is The Verge’s example, not Oracle’s. Oracle’s submission does not contain the word ‘Klingon’ at all, according to Evince ctrl-f.