| |||||||||||||
RMS on Copyrightability of APIsRMS on Copyrightability of APIsPosted May 1, 2012 3:14 UTC (Tue) by gmaxwell (subscriber, #30048)In reply to: RMS on Copyrightability of APIs by drag Parent article: Fair use or "first excuse"? Oracle v. Google goes to the jury (ars technica)
This is a common bad argument. It's seductive because it's made out of true facts, but that doesn't prevent it from being wrong in most places where its applied.
When you are distributing a work you must abide by the license. Imagine some CrazyPublicLicense which says that any piece of software distributed within a 5ft radius is considered a derivative for the purpose of the license and must also be CPL licensed work.
The authors of the CPL don't get to "define what derivative works are", of course, but they have broad freedom to set whatever terms they like on the distribution of _their own code_— including spatial-proximity-during-distribution, a requirement to only copy onto stone tables, or what have you. If you distribute non-CPL work along with CPL work in violation of the license you're infringing the copyright of the CPL work you're distributing.
If you aren't actually distributing (or otherwise engaging in activities which would be unlawful absent the license) the covered work, then indeed its terms don't apply— including its theories of what constitute a derivative for the purpose of the license. But this is a far more clear criteria than that generic argument "does not define what derivative works", which is prone to being applied when the covered work _is_ also being distributed.
(Log in to post comments)
RMS on Copyrightability of APIs Posted May 1, 2012 4:08 UTC (Tue) by drag (subscriber, #31333) [Link]
> When you are distributing a work you must abide by the license. Imagine some CrazyPublicLicense which says that any piece of software distributed within a 5ft radius is considered a derivative for the purpose of the license and must also be CPL licensed work.
No they can't. 'Derivative work' is a legal term. It is defined by the court precedent and legislative law. You don't get to choose the definitions of this phrase. While it's true you can agree to a EULA that places additional restrictions that go beyond the scope of copying/distributing, that is entirely besides the point.
I am guessing that that abusing 'derivative works' term would put the license in jeopardy of being declared invalid... which means that everybody loses their rights to copy the software except the original author.
Needless to say: the GPL limits itself to controlling your copy actions of the code and the copying actions of derivative works in the specific case of distribution.
In all actuality it is very likely that some cases linking GPL libraries by closed source software would be 100% legal (and distributing it), while in other cases linking would not be legal. Also it's very likely that some types of proprietary kernel modules would be legal and others would not. It just depends on the specific circumstances and what the judges decide it to be.
In fact that I am willing to bet that if you tried to claim that your GPL license version would prevent 100% of all use of APIs (or kernel modules, or whatever) by closed source software then that would be actually placing additional restrictions beyond what a true GPL license would have and would make your software subtly incompatible, license-wise, with other software using the true GPL license.
RMS on Copyrightability of APIs Posted May 1, 2012 7:21 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link]
>No they can't. 'Derivative work' is a legal term. It is defined by the court precedent and legislative law. You don't get to choose the definitions of this phrase.
You can _loosen_ it by declaring that you don't consider certain uses to be derivative works. Basically, it's the 'author is [almost] always right' principle.
RMS on Copyrightability of APIs Posted May 1, 2012 7:37 UTC (Tue) by scientes (guest, #83068) [Link]
> Basically, it's the 'author is [almost] always right' principle.
It is not just a principle, its codified as estoppel. This is one reason why Alan Cox is so clear on this issue.
RMS on Copyrightability of APIs Posted May 18, 2012 1:30 UTC (Fri) by steffen780 (guest, #68142) [Link]
> and would make your software subtly incompatible, license-wise, with other software using the true GPL license.
How do I put this.. please read the GPL. Any restriction beyond the GPL itself is voided, you cannot just place additional restrictions on it. Look at, for example, GPLd software in the Apple jailstore - it's illegal. Of course, by definition, the copyright owner cannot break the license (or if she can, it is entirely irrelevant as she cannot sue herself for infringement..), so the owner can put her software in the Apple store anyways.
RMS on Copyrightability of APIs Posted May 1, 2012 13:59 UTC (Tue) by nevets (subscriber, #11875) [Link] The authors of the CPL don't get to "define what derivative works are", of course, but they have broad freedom to set whatever terms they like on the distribution of _their own code_This isn't entirely true. In fact, I would argue that what you just suggested would be struck down in court, as it would most likely be declared as "unconscionable". You can't make a EULA state ownership of other things. Like everything in a 5ft radius. There's limits to what EULA contracts can actually demand. Have a look at this wikipedia article.
RMS on Copyrightability of APIs Posted May 1, 2012 14:49 UTC (Tue) by marcH (subscriber, #57642) [Link]
> There's limits to what EULA contracts can actually demand.
If the licence of a library is struck down in court as unreasonable, that surely makes no one able to use the library until a new licence is issued.
"Judge: since you missed your chance at a licence, sorry you just fell in the public domain". Uh?
RMS on Copyrightability of APIs Posted May 2, 2012 6:42 UTC (Wed) by kevinm (guest, #69913) [Link] This is a common bad argument. It's seductive because it's made out of true facts, but that doesn't prevent it from being wrong in most places where its applied. You are correct, but in this case it isn't wrong - the scenario under discussion is when the GPLd library is not distributed by the author of the closed-source application. Only the closed-source application itself is distributed.
RMS on Copyrightability of APIs Posted May 2, 2012 6:45 UTC (Wed) by gmaxwell (subscriber, #30048) [Link]
Indeed, I was aware of this when I responded— and probably should have been more clear about that— but the lure of someone being wrong on the Internet was too strong for me. The followups appear to have confirmed my assumptions that people were confused about this, sadly.
|
|||||||||||||