Re: Clarification of GPL
Ben Reser wrote: The problem here is exactly that. Assignment is a double edged sword. Assignment makes it easier for one individual to litigate against people who violate the license (which means violating the copyright). But it also permits the assignee to change the license for future releases in any manner which they please. Including proprietary licenses that perhaps the majority of contributors may not be inclined to agree to. Correct. Personally, I would not assign my copyright to someone else unless that person made a promise not to switch to a proprietary license. GNU projects require assignment of copyright to the FSF. In this case it is very unlikely (arguably impossible) that code would be relicensed in any manner that is inconsistent with the GPL. Though some people might even argue this point. I'm not arguing it, but keep in mind the discussions we've seen with the GFDL. The FSF believes this license is in accordance with its principles, but many others disagree. We will have to wait and see what happens with GPL version 3 (which is what the FSF will relicense its code to once this license is written). Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Ben Reser said on Mon, Dec 15, 2003 at 10:27:35PM -0800,: He may be hired by a commercial software firm who pays him a large sum of money to turn the application closed source and work on it Ah, well. You are right. Bu the loss is not for ever. But, other persons can always take the code which is already available under the GPL, and work on it. The original authors' subsequent modifications are no longer available though. Ditto about the fears about GPL 3 going the GFDL way. If existing software under the present GPL is relicensed under a (unjustifiably feared non free) new GPL, users can continue to use the code base available under the older (present version) -- remember, the GPL vests the option to use a later version of the GPL in the user; not the copyright holder. That is the safety of the GPL. -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://in.geocities.com/paivakil +~+ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Ben Reser [EMAIL PROTECTED] writes: The problem here is exactly that. Assignment is a double edged sword. Assignment makes it easier for one individual to litigate against people who violate the license (which means violating the copyright). But it also permits the assignee to change the license for future releases in any manner which they please. Including proprietary licenses that perhaps the majority of contributors may not be inclined to agree to. The copyright assignment forms used by the FSF avoid this problem. The copyright assignment form itself specifies that the FSF is required to follow certain restrictions on the assigned code. A typical example is: The Foundation promises that all distribution of the Work, or of any work based on the Work, that takes place under the control of the Foundation or its assignees, shall be on terms that explicitly and perpetually permit anyone possessing a copy of the work to which the terms apply, and possessing accurate notice of these terms, to redistribute copies of the work to anyone on the same terms. These terms shall not restrict which members of the public copies may be distributed to. These terms shall not require a member of the public to pay any royalty to the Foundation or to anyone else for any permitted use of the work they apply to, or to communicate with the Foundation or its agents in any way either when redistribution is performed or on any other occasion. The Foundation promises that any program based on the Work offered to the public by the Foundation or its assignees shall be offered in the form of machine-readable source code, in addition to any other forms of the Foundation's choosing. However, the Foundation is free to choose at its convenience the media of distribution for machine-readable source code. Ian -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Gream, Matthew wrote: This is the case in the UK under the CDPA 1988, for both cases of copyright assignment (s.90) and exclusive licenses (s.92): they must be in writing and signed. Whether any interpretation, in light of other legal instruments or case law, recognises digital signatures as having equivalent effect to this is question better answered elsewhere. There is a Directive (99/93/EG) which mandates that EC member states recognize electronic signatures as equivalent to paper signatures. I am not sure whether it also states that digital files are equivalent to in writing. It would seem to be the case that submitting a patch constitutes granting a perpetual non-exclusive implied license for the reasonable purposes of incorporation of the the patch into the project under the terms of the license of the project - the patch being used to modify the work (the I think it is debatable in many cases whether a patch by itself is sufficiently original to qualify for copyright. But in any case you'd be best off insisting on an explicit copyright and license statement with the patch. project) and create a derived work (the new project). Interestingly the individual portions (i.e. files) of the project could lapse from copyright (and, therefore, GPL protection), even while copyright subsists in the entire collection as a whole, unless the project could be claimed to be a database, and subject to a relatively perpetual protection under a sui generis database right (which exists in the EU). [1] That's Directive 96/9/EC. I do not think the sui generis database protection can be applied to computer programs. There has to be qualitatively and/or quantitatively a substantial investment in either the obtaining, verification or presentation of the contents (art. 7(1)). European caselaw seems to focus on the principle that the investment has to be primarily aimed at these activities. If the database is a spin-off, a byproduct of something else, it's not protected. http://www.ivir.nl/publications/hugenholtz/fordham2001.html Note that this right is not available for producers of databases who live outside the EC member states (art. 11(2)). Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Clarification of GPL
Each source file is tagged with a header naming him as copyright followed by a GPL header. For anybody to submit a patch to the original distribution, you agree that he gets copyright of it. In most countries, an assignment of copyright has to be in writing and on paper. So an e-mail may well be insufficient. This is the case in the UK under the CDPA 1988, for both cases of copyright assignment (s.90) and exclusive licenses (s.92): they must be in writing and signed. Whether any interpretation, in light of other legal instruments or case law, recognises digital signatures as having equivalent effect to this is question better answered elsewhere. It would seem to be the case that submitting a patch constitutes granting a perpetual non-exclusive implied license for the reasonable purposes of incorporation of the the patch into the project under the terms of the license of the project - the patch being used to modify the work (the project) and create a derived work (the new project). Interestingly the individual portions (i.e. files) of the project could lapse from copyright (and, therefore, GPL protection), even while copyright subsists in the entire collection as a whole, unless the project could be claimed to be a database, and subject to a relatively perpetual protection under a sui generis database right (which exists in the EU). [1] If fact, we could probably split chips further: you in fact may not be modifying and creating the derivative work (i.e. the project) yourself, but perhaps are are giving the patch to a person who will exercise the acts that give real effect to the implied license - this may bring in other complications such as liability of individual who applies the patch, and additional equitable issues between yourself and that individual. Under UK copyright law, moral rights are not applicable to computer programs, nor are there any allowances for revocation (I believe that there are under French droit d'auteur) apart from those that would be related to the terms of the license that are contractual, equitable and estoppel in nature. It would seem not possible to revoke that original implied license unless it could be argued that the terms of it were breached (e.g. in the chain of the appplication of the patch, or changing overall project license, etc). Please correct me if I am wrong - I'm an IPR student and may not yet have the whole picture worked out. Matthew [1] Not a real issue given the duration of copyright and the relative youth of any software, however I find this recent issue with Roland MT-32 and reverse engineering (cf. http://www.artworxinn.com/alex/history.htm) interesting as it suggests that some classes of works that failed to achieve restoration may now be in the public domain in the united states. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Gream, Matthew wrote: That's Directive 96/9/EC. I do not think the sui generis database protection can be applied to computer programs. There has to be qualitatively and/or quantitatively a substantial investment in either the obtaining, verification or presentation of the contents (art. 7(1)). European caselaw seems to focus on the principle that the investment has to be primarily aimed at these activities. If the database is a spin-off, a byproduct of something else, it's not protected. http://www.ivir.nl/publications/hugenholtz/fordham2001.html I'm not sure either that it could be applied to computer programs as such, but for example a large scale continually evolving project such as an operating system residing in a CVS repository seems to fall in the scope of the protection. Was there a substantial investment in creating the CVS repository of code? It seems to me the investment was in writing the code, not in putting the code in the repository. That seems to disqualify the repository from database protection under the spin-off doctrine. Here in the Netherlands, the national association of realtors put the information on houses for sale online in a web database. Previously every realtor had his own private little database. This web database was held to be a spin-off without substantial investment, since the investments made by the realtors were primarily aimed at their own databases, not the web edition. Anyway, I agree that these rights are not fully fleshed out, I was trying to make a point about how the original copyright in a patch could expire, but the patch may still be protected from extraction of the overall work. That may indeed be the case, although I think it may be difficult to encounter such a situation in practice. Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Mahesh, The nearest analogy from literature I can think of at the moment is X being a grammar text book and Y my essay, which conforms to grammar in that text book. Is my essay a derivative of the grammar book? Example is too far-fetched. What if Y were a separate book with extensive treatment of the exercises presented in X ?? Y could not exist without X - the prior publication. Yet copyright AFAIK treats it as an independent work. Certainly as long is it is being distributed as a separate volume. Abe Kornelis. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Abe Kornelis scripsit: The nearest analogy from literature I can think of at the moment is X being a grammar text book and Y my essay, which conforms to grammar in that text book. Is my essay a derivative of the grammar book? Example is too far-fetched. What if Y were a separate book with extensive treatment of the exercises presented in X ?? Indeed, such a book exists: the _C Answer Book_, by Tondo and Gimpel, provides answers to the exercises in _The C Programming Language_, by Kernighan and Richie. -- John Cowan [EMAIL PROTECTED] www.reutershealth.com www.ccil.org/~cowan The exception proves the rule. Dimbulbs think: Your counterexample proves my theory. Latin students think 'Probat' means 'tests': the exception puts the rule to the proof. But legal historians know it means Evidence for an exception is evidence of the existence of a rule in cases not excepted from. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
On Sun, Dec 14, 2003 at 10:24:55AM +0100, Arnoud Engelfriet wrote: That's a realistic worry, although if all those people license their code under GPL, they cannot revoke that license and stop distribution of the program. A bigger issue is if in the future the project wants to change the license. Then they have to ask everyone permission. The problem here is exactly that. Assignment is a double edged sword. Assignment makes it easier for one individual to litigate against people who violate the license (which means violating the copyright). But it also permits the assignee to change the license for future releases in any manner which they please. Including proprietary licenses that perhaps the majority of contributors may not be inclined to agree to. This is true because the GPL does not apply to the owner of a copyrighted work as long as they own the entire work. If they don't own the entire work it becomes rather complex. They'd more than likely have to receive permissions from all the people who do own the rights. They might be able to remove the parts that they didn't own. But even this wouldn't be easy, as parts of the code they have written may be considered derivative works of the code they don't own. Unfortunately, what is and is not a derivative work is a pretty fuzzy line. GNU projects require assignment of copyright to the FSF. In this case it is very unlikely (arguably impossible) that code would be relicensed in any manner that is inconsistent with the GPL. Though some people might even argue this point. However, in this particular case the assignee is not the FSF but the primary author of the application. There is no way to be sure of the author's future motives as you can with the FSF. He may be hired by a commercial software firm who pays him a large sum of money to turn the application closed source and work on it for them. But even if you assume or trust the author to only have good motives you can not assume that his successors in interest will be. Copyrights live on past death and as a result pass to the authors heirs. Those heirs could do anything they want with the software. Including selling it. While it is true that the license provided by the GPL can not be revoked unless you do not comply with the terms, i.e. the released code would be available to the world to use forever, including continuing to make new derivatives licensed under the GPL. It is also true that most people using the GPL do so because the wish to disallow the commercial use of their code without the reciprocal release of source code. Assignment in this case subverts this wish and expectation. The author in this case could will the software to the FSF upon their death. That would certainly avoid the issue of successors (presuming the will is properly done). However, if they're going to will it to the FSF why not just assign all the code to them now? Or possibly some other organization if they don't like the politics of the FSF. I personally don't think assignment is necessary. I find find the motives of people requiring assignment to them personally to be highly suspect. I wouldn't contribute to a project that required that. -- Ben Reser [EMAIL PROTECTED] http://ben.reser.org Conscience is the inner voice which warns us somebody may be looking. - H.L. Mencken -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
ti EMAIL wrote: A piece of software I regularly use is released under the GPL. My concern is how the original writer and maintainer accepts patches. Each source file is tagged with a header naming him as copyright followed by a GPL header. For anybody to submit a patch to the original distribution, you agree that he gets copyright of it. In most countries, an assignment of copyright has to be in writing and on paper. So an e-mail may well be insufficient. Can you transfer copyright to somebody after editing source under GPL, or are these two things unrelated? He claims this is to avoid stupid situations later on where the copyright is split out over a million people, each of which could stop further distribution of his program. That's a realistic worry, although if all those people license their code under GPL, they cannot revoke that license and stop distribution of the program. A bigger issue is if in the future the project wants to change the license. Then they have to ask everyone permission. For a code module/library I?ve written and released under the GPL, is it possible to be incorporated into this previous program giving the previous stipulation based on only the information I?ve stated here? If two works are under GPL, it is always permissible to combine them into a third work under GPL. Can you add requirements to a license in a source file such that your name must always be included as writing the file if anybody decides to use your code? (Questioning if this can be legally binding.) That's certainly legally binding. Note that the GPL already requires people to keep notices in the source intact (see section 1, which must be followed whenever someone distributes original or modified source code). Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Mahesh T. Pai scripsit: If you do not like assigning copyright to the original author, you are free to create your own fork by adding your modifications, and distribute the whole thing yourselves. People did it to GNU Emacs by creating Xemacs. You are indeed free to do this, but it rarely happens and is generally considered a Bad Thing. Forks which remain split and where both branches have substantial life are very few: GNU Emacs/XEmacs, bison/byacc, FreeBSD/NetBSD, NetBSD/OpenBSD. -- John Cowan [EMAIL PROTECTED] http://www.ccil.org/~cowan http://www.reutershealth.com Thor Heyerdahl recounts his attempt to prove Rudyard Kipling's theory that the mongoose first came to India on a raft from Polynesia. --blurb for _Rikki-Kon-Tiki-Tavi_ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
ti EMAIL said on Sat, Dec 13, 2003 at 03:38:59AM -0500,: Each source file is tagged with a header naming him as copyright followed by a GPL header. For anybody to submit a patch to the original distribution, you agree that he gets copyright of it. Requiring assignment of copyright in patches is an issue of project management. Linux, the kernel does not require such assignment. The FSF requires assignment for packages in the GNU project. It is an issue of individual perception. you transfer copyright to somebody after editing source under GPL, or are these two things unrelated? Will you please clarify this?? situations later on where the copyright is split out over a million people, each of which could stop further distribution of his program. Depending on how you look it. See above. For a code module/library I?ve written and released under the GPL, is it possible to be incorporated into this previous program If you do not like assigning copyright to the original author, you are free to create your own fork by adding your modifications, and distribute the whole thing yourselves. People did it to GNU Emacs by creating Xemacs. Can you add requirements to a license in a source file such that your name must always be included as writing the file if anybody decides to use your code? (Questioning if this can be legally binding.) This is what the GPL and several other free/libre licenses do. They go a bit further; and also require that the modifications, if any, made by the (re)distributor also should be mentioned. Regarding legal binding -- In all these years, only the SCO has been silly enough to question its bindingness. -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://in.geocities.com/paivakil +~+ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Mahesh T. Pai wrote: [...] Regarding legal binding -- In all these years, only the SCO has been silly enough to question its bindingness. OTOH, SCO is probably in full agreement with Linus on this: groups.google.com/groups?selm=ZhWT-39U-3%40gated-at.bofh.it quote Yes, but they will cite the prohibition against *creating* derived works. So? The same prohibition exists with the GPL. You are not allowed to create and distribute a derived work unless it is GPL'd. I don't see what you are arguing against. It is very clear: a kernel module is a derived work of the kernel by default. End of story. You can then try to prove (through development history etc) that there would be major reasons why it's not really derived. /quote Now replace kernel with SysV UNIX and GPL with confidential (OCO or something like that). How nice. regards, alexander. P.S. www.byte.com/documents/s=8276/byt1055784622054/0616_marshall.html quote GPL GPL has the same derivative rights concept [as UNIX], according to Sontag... /quote -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Clarification of GPL
Alexander Terekhov said on Sat, Dec 13, 2003 at 07:06:40PM +0100,: Now replace kernel with SysV UNIX and GPL with confidential (OCO or something like that). How nice. I consider this as a bug with the law - silliness of treating programs as analogous to `literary, artistic and dramatic works' as I find from the treaties, and the Indian law. AFAIK, at least few EU countries treat the too treat software as a part of the category `literary, dramatic and artistic' work. The situation is this - I have program X, and you wrote Y which depends on X. `Derivative work' in traditional copyright law was a work which modified X. Here, though Y is depending on X; and will not work without X, (a plug in; kernel loadable module) traditionally, it ought to be treated as an independent work. The nearest analogy from literature I can think of at the moment is X being a grammar text book and Y my essay, which conforms to grammar in that text book. Is my essay a derivative of the grammar book? But, in computer programs, the analogy will not hold good; and therefore, the law is buggy. Why waste bandwidth trying to work within the framework of buggy laws? We should try to change the law. But that is work for another list ... -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://in.geocities.com/paivakil +~+ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3