Re: GPL + question
On Sun, May 31, 2015 at 11:32:57AM +0100, Ian Jackson wrote: I think the problem here is the notion that a file necessarily has exactly one licence. Totally agree. [snip] So it is true that a downstream redistributor who does not change F cannot change the licence, because the only permission needed (in copyright law) is that from upstream. If upstream grant GPLv2+ then even if the downstream writes GPLv3+ then everyone can still rely on upstream's permission, putting the upstream GPLv2+ notice back. They can do it because the license never changed, it was *just* distributed under a different set of terms (the GPLv2+ says you can distribute it as if it were GPLv3+ and everything is tidy -- it does *not* say you can yell hocus pocus and the license changes into GPLv3+ and can never be turned back.) However, if a downstream redistributor modifies the file, they can narrow the set of permissions. This is because in copyright law, they have their own copyright in the modifications. So if a downstream changes GPLv2+ to GPLv3+ _and modifies the file_ then the GPLv2+ is no longer applicable. To the combined / derived work, yes. If the modifications were reverted, there's no reason the file would be GPLv3+, it'd go back to the original terms, since you (as *not* the copyright holder) can't relicense the work. [snip] (So I think Paul is wrong if he thinks, as he seems to, that it is a violation of copyright law to change GPLv2+ to GPLv3+ when merely redistributing.) This is again confusing the issue. Let me be clear: A) You can *not* relicense a work that you don't hold. B) The GPL*+ grants you the right to redistribute under future terms and still be complient with the version that its license under. You can use B to distribute under the GPLv3 if you would like, but *you do not relicense the work*. It's still GPLv2. You can take GPLv2+ and *distribute* it as if it was GPLv3, but you *do not relicense* the work. Given that A is a thing, you have no way to do that anyway. But to do so would be rude and we should try not to do it. So I think that where practical if we can determine that a package is dual (or triple) licenced, we should document all the permissions - that is, all the licences. I think that's sensable, I have a hard time believing upstreams are so good, honestly, we should be thankful to the author of this package we're talking about, since they took the time to document it was derived (hopefully modified!) from a LGPL work. It appears that in this particular case package has been modified by the GPLv3+-preferring downstream. In that case there is no permission to distribute under GPLv2+ any more. Totally agree. I don't think it is sensible to insist that the Debian maintainer do a lot of work to try to discover whether some files in the package have /not/ been modified by the GPLv3+-preferring author (or by other people who contributed to that author's version and never saw a GPLv2+ licence). I agree we should trust upstream in this case, but I'm generally interested. I don't like the idea people think they can just magically relicense files rather then use the right to redistribute. That work is not necessary in copyright law, and we aren't really doing the Free Software world much of a service by performing it - at least, once we have decided to package the GPLv3+ fork at all. If someone wants to try to strip the GPLv3+ parts out of the fork then that's fine of course, but I don't think we should insist that the Debian maintainer do the necessary archaeology. I don't think that's needed; I love the GPLv3 and I don't see a problem here :) I also wouldn't have the maintainer add it to the copyright file, but there is a lot of misunderstanding on this thread on the mechanism by which the GPLv3 works. The proof is on you -- where does it say you can relicense someone else's copyrighted work / IP? Not *redistribute*, *relicense*. The answer to this question is very clear. The proof is in the original licensing notice: either version 2 of the License, or (at your option) any later version. The downstream distributor may choose a later version (3, say, or any version 3 or later) and comply with its terms. No, wait Ian :) Full quote: | This program is free software; you can redistribute it and/or modify | it under the terms of the GNU General Public License as published by | the Free Software Foundation; either version 2 of the License, or | (at your option) any later version. You may *redistribute* or *modify*. You may not *relicense*. I can't come along to gettext and say it's now GPLv3+, because it will forever be GPLv2+ until the copyright holder(s) relicense. I can merely redistribute it under the terms of the GPLv3+, until I modify it, where I can change the terms of the derived work, but *not* the original works. If I revert my changes, there's no reason it'd still be GPLv3+ I strongly think we
Re: GPL + question
On Sun, 31 May 2015 13:10:14 -0400 Paul Tagliamonte wrote: [...] They can do it because the license never changed, it was *just* distributed under a different set of terms (the GPLv2+ says you can distribute it as if it were GPLv3+ and everything is tidy -- it does *not* say you can yell hocus pocus and the license changes into GPLv3+ and can never be turned back.) But here we are not talking about GPL v2 or later; we are talking about files which (before being modified) were originally under LGPL v2 or later... And the LGPL v2 says you can yell hocus pocus (more or less). Please see my previous message: https://lists.debian.org/debian-legal/2015/05/msg00045.html -- http://www.inventati.org/frx/ There's not a second to spare! To the laboratory! . Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE pgpPxlKencabs.pgp Description: PGP signature
Re: GPL + question
Paul Tagliamonte writes (Re: GPL + question): They *can* since the work as modified *can* be distributed under the terms of the GPLv3+, *without* changing the original work's license, but the *file* can be distributed as GPLv3+, since that's the minimum license needed to comply with all parts. I think the problem here is the notion that a file necessarily has exactly one licence. The only actually accurate statements one can make are there is [not] permission to distribute file F under licence L. If you want to know what licence does a file have what you really ean is for what set of licences L1,L2,L3... is it the case that there is permission to distribute. So it is true that a downstream redistributor who does not change F cannot change the licence, because the only permission needed (in copyright law) is that from upstream. If upstream grant GPLv2+ then even if the downstream writes GPLv3+ then everyone can still rely on upstream's permission, putting the upstream GPLv2+ notice back. However, if a downstream redistributor modifies the file, they can narrow the set of permissions. This is because in copyright law, they have their own copyright in the modifications. So if a downstream changes GPLv2+ to GPLv3+ _and modifies the file_ then the GPLv2+ is no longer applicable. For Debian, there is the question of what to put in debian/copyright. Obviously we need to put in debian/copyright some applicable Ln (that is, an Ln which applies to all the files F). If there is permission to distribute under L1,L2,... then it would be legitimate in a copyright law sense to write in debian/copyright only L1. To do so would not be a breach of the copyright, because we would be acting in accordance with L1. (So I think Paul is wrong if he thinks, as he seems to, that it is a violation of copyright law to change GPLv2+ to GPLv3+ when merely redistributing.) But to do so would be rude and we should try not to do it. So I think that where practical if we can determine that a package is dual (or triple) licenced, we should document all the permissions - that is, all the licences. It appears that in this particular case package has been modified by the GPLv3+-preferring downstream. In that case there is no permission to distribute under GPLv2+ any more. I don't think it is sensible to insist that the Debian maintainer do a lot of work to try to discover whether some files in the package have /not/ been modified by the GPLv3+-preferring author (or by other people who contributed to that author's version and never saw a GPLv2+ licence). That work is not necessary in copyright law, and we aren't really doing the Free Software world much of a service by performing it - at least, once we have decided to package the GPLv3+ fork at all. If someone wants to try to strip the GPLv3+ parts out of the fork then that's fine of course, but I don't think we should insist that the Debian maintainer do the necessary archaeology. The proof is on you -- where does it say you can relicense someone else's copyrighted work / IP? Not *redistribute*, *relicense*. The answer to this question is very clear. The proof is in the original licensing notice: either version 2 of the License, or (at your option) any later version. The downstream distributor may choose a later version (3, say, or any version 3 or later) and comply with its terms. Thanks, Ian. -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/21866.58201.37278.496...@chiark.greenend.org.uk
Re: GPL + question
On Sat, 30 May 2015 23:24:53 +0200 Ángel González keis...@gmail.com wrote: On 30/05/15 03:30, Riley Baird wrote: Only the copyright holder can change what a *work* is licensed as. Unless the copyright holder grants the permission to do so, I would say... Let's say I hold copyright on a work, and I grant someone else permission to change the license of a work. Who would enforce the second license? Only a copyright holder can enforce their copyrights. IMHO you would be the one responsible for enforcing the license... Exactly. So, if a work is originally licensed under GPL-2+ and Person A makes a copy and gives it to Person B under GPL-3. Now consider that Person B gives a copy to Person C under GPL-2+. Person A can't enforce the original copyright holder's copyrights. I find it difficult to believe that the original copyright holder would enforce Person A's license. unless you also granted (delegated?) the right of enforcing the work license to someone else. I'm not sure that you can grant the right of enforcing the license to someone else, otherwise why wouldn't copyleft authors just let give everyone the right to enforce their license? pgp0cuqkC5oY0.pgp Description: PGP signature
Re: GPL + question
I'm not sure that you can grant the right of enforcing the license to someone else, I suspect that for legal litigation you may need to represent the copyright owner. That's what I meant; I probably didn't word it clearly, though. pgp4w78cg1zYD.pgp Description: PGP signature
Re: GPL + question
On 30/05/15 03:30, Riley Baird wrote: Only the copyright holder can change what a *work* is licensed as. Unless the copyright holder grants the permission to do so, I would say... Let's say I hold copyright on a work, and I grant someone else permission to change the license of a work. Who would enforce the second license? Only a copyright holder can enforce their copyrights. IMHO you would be the one responsible for enforcing the license... unless you also granted (delegated?) the right of enforcing the work license to someone else. -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/556a2aa5.90...@gmail.com
Re: GPL + question
On 31/05/15 00:10, Riley Baird wrote: On Sat, 30 May 2015 23:24:53 +0200 Ángel Gonzálezkeis...@gmail.com wrote: IMHO you would be the one responsible for enforcing the license... Exactly. So, if a work is originally licensed under GPL-2+ and Person A makes a copy and gives it to Person B under GPL-3. Now consider that Person B gives a copy to Person C under GPL-2+. Person A can't enforce the original copyright holder's copyrights. I find it difficult to believe that the original copyright holder would enforce Person A's license. unless you also granted (delegated?) the right of enforcing the work license to someone else. I'm not sure that you can grant the right of enforcing the license to someone else, Copyright collecting societies do so. otherwise why wouldn't copyleft authors just let give everyone the right to enforce their license? I suspect that for legal litigation you may need to represent the copyright owner. Also note that if authors gave that power to everyone, anyone attempting to exercise that right would still need the author in order to prove that the author didn't also provide a propietary license to the infringer,¹ so it wouldn't be that useful. Usually, copyright collecting societies are the only ones entitled to license that author's work, and their position when detecting infringement is thus quite different. ¹ unless they also gave up the right to ever license it under different terms, perhaps. A very bad idea IMHO. -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/556a50c1.2070...@gmail.com
Re: GPL + question
Charles Plessy ple...@debian.org writes: If it were me, I would give the benefit of the doubt to the upstream author of missfits, and trust him that if he added a GPLv3+ header, it is because he modified the files, as he says in the README. When I adopted the first package from this author (sextractor), I asked him per E-mail about this subdirectory -- more since a newer version of the library in question (wcslib) is already in Debian, and I wanted to avoid using a convienience copy. He convinced me then that his version is slightly changed and therefore not just linkable with the original. I didn't ask for his other packages (missfits, scamp, swarp, psfex; some of them are already in Debian, others are now in NEW), but since they all follow the same structure, I am quite sure that his arguments are valid there as well. In that case, the license to be indicated in debian/copyright should be GPLv3+. I re-uploaded the package with a (hopefully) clarifying comment in debian/copyright; let's see how he decides now. Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/87h9quebju@debian.org
Re: GPL + question
On Fri, May 29, 2015 at 09:32:12AM +0200, Ole Streicher wrote: Hi, I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. Upstream authors can't change licensing of any files, under any conditions, ever. If I say a file is GPLv2+, it is forever GPLv2+, even if it's combined with a GPLv3 work, in that case the *files* are still GPLv2+, that other file is a GPLv3 work, and the *combined work* is distributed under the terms of the GPLv3, since it satisfies the license of every file in the combined / derived work. debian/copyright currently mentions only GPL-3+ for the whole package. Yeah, debian/copyright isn't what the binary is distributed under, it's what the source licenses are. If it had MIT/Expat code, you'd still need it in debian/copyright if the other files are GPLv2+ The ftp-master now asked me to add GPL-2+ for these files to debian/copyright. But I think that this would be wrong, since the files under src/wcs are not distributable under GPL-2+ (because they contain GPL-3+ code from Emmanuel Bertin). Nah, it's wrong because you said LGPLv2+, adding it sounds right. Just because files are being combined in such a way that they're distributed under different terms than some of the files doesn't mean we exclude them. Just like Expat is contained within BSD-3. Or ISC is contained in Expat. You still need all three, since that's the licese for the file. Only the copyright holder can change what a *work* is licensed as. Anyone can distribute a derived work inline witht he terms of their license. That may also contains other terms as well. Do I miss an important point here? Best regards Ole Cheers, Paul -- :wq signature.asc Description: Digital signature
Re: GPL + question
Maximilian maximil...@actoflaw.co.uk writes: and this seems to imply that the end user can choose which licence suits them. Not only the end user -- also (in our case) the upstream author. So, he can choose to redistribute the files under GPL-3+. Being them modified or not. However, if Emmanuel Bertin's code is specifically licensed as GPLv3 only then it needs to be made clear that this is the case where applicable - the fact that this code is GPLv3 only ought not affect the fact that the other original files may be GPLv2+. True. However, the original files are not in the upstream tarball and therefore do not need to be documented in debian/copyright. This is even the case if original and redistributed filed differ only by their license. Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/ytza8wnmsm1@news.ole.ath.cx
Re: GPL + question
That's literally what I said. d/copyright is for source not binary. On May 29, 2015 8:42 AM, Riley Baird bm-2cvqnduybau5do2dfjtrn7zbaj246s4...@bitmessage.ch wrote: I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. Upstream authors can't change licensing of any files, under any conditions, ever. If I say a file is GPLv2+, it is forever GPLv2+, even if it's combined with a GPLv3 work, in that case the *files* are still GPLv2+, that other file is a GPLv3 work, and the *combined work* is distributed under the terms of the GPLv3, since it satisfies the license of every file in the combined / derived work. But there are multiple works being combined into the one file. So some parts of the file are GPLv2+ and other parts of the file are GPLv3. The file as a whole can only be distributed under GPLv3.
Re: GPL + question
Paul Tagliamonte paul...@debian.org writes: On Fri, May 29, 2015 at 09:32:12AM +0200, Ole Streicher wrote: Hi, I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. Upstream authors can't change licensing of any files, under any conditions, ever. Generally spoken, this is wrong (or please point me to the source). For example, I can give you a file saying do what you want with it, then this is the license. However, do what you want with it includes that you can republish the file as GPL. For GPL. GPL-2+ contains a statement | you can redistribute it and/or modify it under the terms of the GNU | General Public License as published by the Free Software Foundation; | either version 2 of the License, or (at your option) any later | version. which means: the upstream author is allowed to redistribute the (changed or even unchanged) files under version 2, or also under version 3. Since he redistributed them under version 3 or later, the license of these files is GPL-3+. The statement above explicitely allows him to do so. If I say a file is GPLv2+, it is forever GPLv2+ Could you strengthen this with a reference? even if it's combined with a GPLv3 work, in that case the *files* are still GPLv2+, The files are modified. The author of the modification applied GPL-3+ to the changed file. He has the right to do so (see above), and since original and change are glued together (neither the change nor the original are separately distributed by upstream), the modified file cannot be distributed by GPL-2 anymore. debian/copyright currently mentions only GPL-3+ for the whole package. Yeah, debian/copyright isn't what the binary is distributed under, it's what the source licenses are. I speak about sources. The ftp-master now asked me to add GPL-2+ for these files to debian/copyright. But I think that this would be wrong, since the files under src/wcs are not distributable under GPL-2+ (because they contain GPL-3+ code from Emmanuel Bertin). Nah, it's wrong because you said LGPLv2+, adding it sounds right. It is wrong. The files in src/wcs are not distributable under GPL-2, since they contain changes that are GPL-3+. Only the copyright holder can change what a *work* is licensed as. Anyone can distribute a derived work inline witht he terms of their license. That may also contains other terms as well. If the original license allows, then anyone can redistribute the files under a different license. And (L)GPL has a paragraph that allows this under certain conditions (namely LGPL - GPL, and version upgrades). Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/ytziobbmtsw@news.ole.ath.cx
Re: GPL + question
I'm probably wrong, but the code that was originally GPLv2+ remains licensed under the GPLv2 *in addition* to the GPLv3 that the overall package is licensed under. The GPLv2 states that: 'if the Program specifies a version number of this License which applies to it and any later version, you have the option of following the terms and conditions *either* of that version *or* of any later version published by the Free Software Foundation' (my emphasis) and this seems to imply that the end user can choose which licence suits them. However, if Emmanuel Bertin's code is specifically licensed as GPLv3 only then it needs to be made clear that this is the case where applicable - the fact that this code is GPLv3 only ought not affect the fact that the other original files may be GPLv2+. As said above I'm probably wrong, but at least that's the way I see it! Regards, Max On 29 May 2015 08:32, Ole Streicher oleb...@debian.org wrote: Hi, I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. debian/copyright currently mentions only GPL-3+ for the whole package. The ftp-master now asked me to add GPL-2+ for these files to debian/copyright. But I think that this would be wrong, since the files under src/wcs are not distributable under GPL-2+ (because they contain GPL-3+ code from Emmanuel Bertin). Do I miss an important point here? Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/ytzmw0nn8jn@news.ole.ath.cx
Re: GPL + question
Paul Tagliamonte paul...@debian.org writes: On Fri, May 29, 2015 at 10:41:58PM +1000, Riley Baird wrote: But there are multiple works being combined into the one file. So some parts of the file are GPLv2+ and other parts of the file are GPLv3. The file as a whole can only be distributed under GPLv3. I don't see the point in adding LGPL, *IFF* the works *ARE* modified and derived works. Not just straight copy-paste. I'd be interested in what changes took place, I don't see any marking of it. Same for me. However: the (L)GPL allows even an unmodified redistribution under a later license. It is up to upstream to decide whether he chooses the original or a later one. And since I take these files from upstream, not from the original author, I am bound to his decision, independently whether the files are modified or not. Therefore, if he chooses to redistribute the files in src/wcs/ under GPL-3+, than this is the license for these file, and it should be documented as such under debian/copyright. And in this case, the redistribution under a GPL-3+ is clear (by adding the according statement to the file headers). This doesn't appear to be the case, this looks like LGPLv2.1+ files were modified by someone licensing their changes under GPLv3+, which is legit. I believe treating this file as GPLv3+ is fine / good enough. The reason here is not modification (although it makes this case clear), but redistribution. Upstream has chosen to redistribute the files under GPL-3+, and if we want to use these files, we have to respect this. Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/ytzeglzmsxd@news.ole.ath.cx
Re: GPL + question
I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. debian/copyright currently mentions only GPL-3+ for the whole package. The ftp-master now asked me to add GPL-2+ for these files to debian/copyright. But I think that this would be wrong, since the files under src/wcs are not distributable under GPL-2+ (because they contain GPL-3+ code from Emmanuel Bertin). From the facts you have given, I think that your view is correct. pgpa6taNbDWeb.pgp Description: PGP signature
Re: GPL + question
I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. Upstream authors can't change licensing of any files, under any conditions, ever. If I say a file is GPLv2+, it is forever GPLv2+, even if it's combined with a GPLv3 work, in that case the *files* are still GPLv2+, that other file is a GPLv3 work, and the *combined work* is distributed under the terms of the GPLv3, since it satisfies the license of every file in the combined / derived work. But there are multiple works being combined into the one file. So some parts of the file are GPLv2+ and other parts of the file are GPLv3. The file as a whole can only be distributed under GPLv3. pgp_WyX02xrY3.pgp Description: PGP signature
Re: GPL + question
On Fri, May 29, 2015 at 10:41:58PM +1000, Riley Baird wrote: But there are multiple works being combined into the one file. So some parts of the file are GPLv2+ and other parts of the file are GPLv3. The file as a whole can only be distributed under GPLv3. the terminology being thrown around was so confusing I had to look at the source to see what was actually going on here :) There was *one* work, which *was* LGPL. By an author. They published it on their own. This work will forver be LGPL. The author of this package took that source, and *modified* it (modified, *not* combined). This modified work is distributed as GPLv3. I don't see the point in adding LGPL, *IFF* the works *ARE* modified and derived works. Not just straight copy-paste. I'd be interested in what changes took place, I don't see any marking of it. Defer to the ftp-master who processed it. Ask them for clarification (feel free to point to this mail) In the case where two works are combined into one file - this is functionally compilation (at least not the preferred form of modification, which means it's *not* source) This doesn't appear to be the case, this looks like LGPLv2.1+ files were modified by someone licensing their changes under GPLv3+, which is legit. I believe treating this file as GPLv3+ is fine / good enough. Cheers, Paul -- .''`. Paul Tagliamonte paul...@debian.org | Proud Debian Developer : :' : 4096R / 8F04 9AD8 2C92 066C 7352 D28A 7B58 5B30 807C 2A87 `. `'` http://people.debian.org/~paultag `- http://people.debian.org/~paultag/conduct-statement.txt signature.asc Description: Digital signature
Re: GPL + question
If I say a file is GPLv2+, it is forever GPLv2+, even if it's combined with a GPLv3 work, in that case the *files* are still GPLv2+, that other file is a GPLv3 work, and the *combined work* is distributed under the terms of the GPLv3, since it satisfies the license of every file in the combined / derived work. But there are multiple works being combined into the one file. So some parts of the file are GPLv2+ and other parts of the file are GPLv3. The file as a whole can only be distributed under GPLv3. That's literally what I said. You gave the impression that each *file* could only be under one license, which would be good for the purposes of d/copyright, but isn't the case. pgpgjisQHmj6s.pgp Description: PGP signature
Re: GPL + question
On Fri, May 29, 2015 at 03:09:34PM +0200, Ole Streicher wrote: Same for me. However: the (L)GPL allows even an unmodified redistribution under a later license. This is key -- redistribution. It doesn't change the license. If I get this file after you say it's GPLv3, it's still LGPLv2.1+ to me if I remove it from other works that change the distribution terms (unless it's been modified, in which case the licensing of the work on the whole changes, and yadda yadda yadda) I originally thought there was a different question being asked; sorry about that (the terms used and not looking at the source didn't help :)) It is up to upstream to decide whether he chooses the original or a later one. And since I take these files from upstream, not from the original author, I am bound to his decision, independently whether the files are modified or not. Unmodified, the license of the works is unchanged, even if we *distribute* under a different one. Therefore, if he chooses to redistribute the files in src/wcs/ under GPL-3+, than this is the license for these file, and it should be documented as such under debian/copyright. And in this case, the redistribution under a GPL-3+ is clear (by adding the according statement to the file headers). This doesn't appear to be the case, this looks like LGPLv2.1+ files were modified by someone licensing their changes under GPLv3+, which is legit. I believe treating this file as GPLv3+ is fine / good enough. The reason here is not modification (although it makes this case clear), but redistribution. Upstream has chosen to redistribute the files under GPL-3+, and if we want to use these files, we have to respect this. Best regards Ole -- .''`. Paul Tagliamonte paul...@debian.org | Proud Debian Developer : :' : 4096R / 8F04 9AD8 2C92 066C 7352 D28A 7B58 5B30 807C 2A87 `. `'` http://people.debian.org/~paultag `- http://people.debian.org/~paultag/conduct-statement.txt signature.asc Description: Digital signature
Re: GPL + question
Or a CLA. Or breaking copyright law. Or modified the work and distribute it under a superset of the old terms. Or or or :) For the record; I don't believe Apple is breaking copyright law, and I didn't mean to imply that :) -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/CAO6P2QQOZvg4eP2Rrcbf2XRwVZY_c-bDhRaSWVE-+g=mfp-...@mail.gmail.com
Re: GPL + question
On Fri, May 29, 2015 at 04:06:52PM +0200, Ole Streicher wrote: Paul Tagliamonte paul...@debian.org writes: On Fri, May 29, 2015 at 03:09:34PM +0200, Ole Streicher wrote: Same for me. However: the (L)GPL allows even an unmodified redistribution under a later license. This is key -- redistribution. It doesn't change the license. It does. Just look into the license (resp. the header, for simplicity): No, it doesn't. | Each version is given a distinguishing version number. If the | Library as you received it specifies that a certain numbered version | of the GNU Lesser General Public License or any later version | applies to it, you have the option of following the terms and | conditions either of that published version or of any later version | published by the Free Software Foundation. Note this says you have the option of following the terms and conditions of the version noted, *or* any later version, *not* that you relicense, you can just follow different terms. This means you can redistribute under the terms of whatever, but not relicense. To relicense implies you hold copyright, since only the copyright holder can license their works, even copylefted works. | you can redistribute it and/or modify it under the terms of the GNU | General Public License as published by the Free Software Foundation; | either version 2 of the License, or (at your option) any later version. So, redistribution may change the license. No, you may redistribute it under different terms, *not* relicense. You may *use* GPLv2+ as GPLv3+, *BUT* the original work is *STILL* GPLv2+, since you can't relicense works. To relicense implies you hold copyright, since only the copyright holder can license their works, even copylefted works. Snipping the rest, this seems to be your major point of confusion. Cheers, Paul -- .''`. Paul Tagliamonte paul...@debian.org | Proud Debian Developer : :' : 4096R / 8F04 9AD8 2C92 066C 7352 D28A 7B58 5B30 807C 2A87 `. `'` http://people.debian.org/~paultag `- http://people.debian.org/~paultag/conduct-statement.txt signature.asc Description: Digital signature
Re: GPL + question
Paul Tagliamonte paul...@debian.org writes: No, you may redistribute it under different terms, *not* relicense. You may *use* GPLv2+ as GPLv3+, *BUT* the original work is *STILL* GPLv2+, since you can't relicense works. Sorry, but I still think release under the terms of the General Public License v3+ means that the file has the license GPLv3+. To relicense implies you hold copyright, since only the copyright holder can license their works, even copylefted works. Again: please provide a reference for this. The copyright holder has surely the initial right to license his work, but I don't see a reason why he can't transfer this. It is also wrong for the changed case that we have: If only the copyright holder (Mark Calabretta) had the right to change the license, then the files in question could not have been modified and distributed under the GPL-3+ license by the upstream author (Emmanuel Bertin) -- since even the modified files are still copyrighted by Mark, so the Emmanuel alone could not change their license. This is, however, against the idea of the + in the GPL versions. Therefore, please show a proof that only the copyright holder can change the license. Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/87twuve7vz@debian.org
Re: GPL + question
Miriam Ruiz mir...@debian.org writes: So in my opinion, if you modify a code which was released under GPL2+ and you license your modifications as GPL3+, the resulting work has to also be GPL, and the terms or conditions that apply are those of the version 3 of the lincense, or later, but you're not effectively relicensing the code that is not yours, so that part would be still licensed as GPL2+ by the author and copyright holder. I may give to others the permission to use the modified/redistributed file under GPL-3+. This permission is what is usually called License. In that sense, the license is changed. So if you later removed the part of code that was covered by a different license, the resulting code would be still under the original license, The license is usually granted to a file as a whole, not to specific lines. If got got a changed file from me, and you revert my changes, then you are still bound to the conditions that we agreed about when you got the file -- these conditions are the license. If we agreed on GPL-3, then you are bound to GPL-3. because you were never the copyright holder, and you never had permission to relicense it. I seriously doubt that any judge would rule otherwise. Just again this example: http://www.opensource.apple.com/source/xnu/xnu-1456.1.26/bsd/sys/msg.h This is a file that is initially copyrighted by Daniel Boulet (and licensed under BSD-2-Clause). However, without any other change, it also has the header | Copyright (c) 2000-2007 Apple Inc. All rights reserved. | [...] | This file contains Original Code and/or Modifications of Original Code | as defined in and that are subject to the Apple Public Source License | Version 2.0 (the 'License'). You may not use this file except [...] So, Apple puts another license to this file, probably without having the permission of Daniel Boulet. Would you accept such a file in Debian? It is clearly not BSD-licensed, even if an unchanged BSD-licensed version exists. When trusting the Apple Lawyers a bit, then this contradicts your argumentation. Best regards Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/87pp5je701@debian.org
Re: GPL + question
Paul Tagliamonte paul...@debian.org writes: I don't know any jurisdiction where I can take a work of yours and now claim I have the rights to it under a different license. Apple did, as I have shown. I think they have good lawyers. Best Ole -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/87lhg7e6ee@debian.org
Re: GPL + question
On Fri, May 29, 2015 at 05:43:21PM +0200, Ole Streicher wrote: Paul Tagliamonte paul...@debian.org writes: I don't know any jurisdiction where I can take a work of yours and now claim I have the rights to it under a different license. Apple did, as I have shown. I think they have good lawyers. Or a CLA. Or breaking copyright law. Or modified the work and distribute it under a superset of the old terms. Or or or :) (FWIW, BSD-alikes deal perfectly fine with further restrictions so long as their terms are met; GPL does not.) Best Ole For further clarification, I'd suggest asking the FSF about the differences in relicensing vs redistributing under the GPL/LGPL. Cheers, Paul -- #define sizeof(x) rand() /paul :wq signature.asc Description: Digital signature
Re: GPL + question
On Fri, May 29, 2015 at 05:11:12PM +0200, Ole Streicher wrote: Again: please provide a reference for this. The copyright holder has surely the initial right to license his work, but I don't see a reason why he can't transfer this. Via copyright asignment, not licensing, unless the license includes a copyright asignment to an entity. It is also wrong for the changed case that we have: If only the copyright holder (Mark Calabretta) had the right to change the license, then the files in question could not have been modified and distributed under the GPL-3+ license by the upstream author (Emmanuel Bertin) -- They *can* since the work as modified *can* be distributed under the terms of the GPLv3+, *without* changing the original work's license, but the *file* can be distributed as GPLv3+, since that's the minimum license needed to comply with all parts. since even the modified files are still copyrighted by Mark, so the Emmanuel alone could not change their license. This is, however, against the idea of the + in the GPL versions. No, it's really not. Therefore, please show a proof that only the copyright holder can change the license. Wat? Copyright statute? What jurisdiction? If you want to fight this, I suggest you get a lawyer, I don't know any jurisdiction where I can take a work of yours and now claim I have the rights to it under a different license. The proof is on you -- where does it say you can relicense someone else's copyrighted work / IP? Not *redistribute*, *relicense*. Cheers, Paul -- .''`. Paul Tagliamonte paul...@debian.org | Proud Debian Developer : :' : 4096R / 8F04 9AD8 2C92 066C 7352 D28A 7B58 5B30 807C 2A87 `. `'` http://people.debian.org/~paultag `- http://people.debian.org/~paultag/conduct-statement.txt signature.asc Description: Digital signature
Re: GPL + question
2015-05-29 16:06 GMT+02:00 Ole Streicher oleb...@debian.org: Paul Tagliamonte paul...@debian.org writes: On Fri, May 29, 2015 at 03:09:34PM +0200, Ole Streicher wrote: Same for me. However: the (L)GPL allows even an unmodified redistribution under a later license. This is key -- redistribution. It doesn't change the license. It does. Just look into the license (resp. the header, for simplicity): | you can redistribute it and/or modify it under the terms of the GNU | General Public License as published by the Free Software Foundation; | either version 2 of the License, or (at your option) any later version. So, redistribution may change the license. It is indeed quite a grey area, and quite confusing, in my opinion. According with the (simple but enough for my purposes) definition in Wikipedia (Copyright is a form of intellectual property, applicable to any expressed representation of a creative work. It is often shared among multiple authors, each of whom holds a set of rights to use or license the work, and who are commonly referred to as rightsholders. These rights frequently include reproduction, control over derivative works, distribution, public performance, and moral rights such as attribution.) [1], it is the author[s] the one[s] who has the rights to license the work. GPL2 [2] says: This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The Program, below, refers to any such program or work, and a work based on the Program means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language., so the unmodified program is explicitly defined by the license as a work based on the Program. It also says that Activities other than copying, distribution and modification are not covered by this License; they are outside its scope, so the license does explicitly not apply to relicensing. You can't relicense other person's work released under GPL2. What the license says is that You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License. It is also said that If the Program specifies a version number of this License which applies to it and any later version, you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. So in my opinion, if you modify a code which was released under GPL2+ and you license your modifications as GPL3+, the resulting work has to also be GPL, and the terms or conditions that apply are those of the version 3 of the lincense, or later, but you're not effectively relicensing the code that is not yours, so that part would be still licensed as GPL2+ by the author and copyright holder. So if you later removed the part of code that was covered by a different license, the resulting code would be still under the original license, because you were never the copyright holder, and you never had permission to relicense it. I seriously doubt that any judge would rule otherwise. That's just my two cents. Greetings, Miry [1] http://en.wikipedia.org/wiki/Copyright [2] http://www.gnu.org/licenses/gpl-2.0.html -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/CAFotxVMyvYNxbg+asOJHDNNMopMVfLVeEeKW4=cazhoedvr...@mail.gmail.com
Re: GPL + question
Please end this thread, it's getting nuts. Ask the FSF if you're still unclear. Thanks, Paul On Fri, May 29, 2015 at 11:41 AM, Paul Tagliamonte paul...@debian.org wrote: On Fri, May 29, 2015 at 05:11:12PM +0200, Ole Streicher wrote: Again: please provide a reference for this. The copyright holder has surely the initial right to license his work, but I don't see a reason why he can't transfer this. Via copyright asignment, not licensing, unless the license includes a copyright asignment to an entity. It is also wrong for the changed case that we have: If only the copyright holder (Mark Calabretta) had the right to change the license, then the files in question could not have been modified and distributed under the GPL-3+ license by the upstream author (Emmanuel Bertin) -- They *can* since the work as modified *can* be distributed under the terms of the GPLv3+, *without* changing the original work's license, but the *file* can be distributed as GPLv3+, since that's the minimum license needed to comply with all parts. since even the modified files are still copyrighted by Mark, so the Emmanuel alone could not change their license. This is, however, against the idea of the + in the GPL versions. No, it's really not. Therefore, please show a proof that only the copyright holder can change the license. Wat? Copyright statute? What jurisdiction? If you want to fight this, I suggest you get a lawyer, I don't know any jurisdiction where I can take a work of yours and now claim I have the rights to it under a different license. The proof is on you -- where does it say you can relicense someone else's copyrighted work / IP? Not *redistribute*, *relicense*. Cheers, Paul -- .''`. Paul Tagliamonte paul...@debian.org | Proud Debian Developer : :' : 4096R / 8F04 9AD8 2C92 066C 7352 D28A 7B58 5B30 807C 2A87 `. `'` http://people.debian.org/~paultag `- http://people.debian.org/~paultag/conduct-statement.txt -- :wq -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cao6p2qqdbfbxp-w4amfwn0q0n_reh4zs+aqwygggevfzfs+...@mail.gmail.com
Re: GPL + question
On 29/05/15 16:30, Ole Streicher wrote: Miriam Ruiz mir...@debian.org writes: So in my opinion, if you modify a code which was released under GPL2+ and you license your modifications as GPL3+, the resulting work has to also be GPL, and the terms or conditions that apply are those of the version 3 of the lincense, or later, but you're not effectively relicensing the code that is not yours, so that part would be still licensed as GPL2+ by the author and copyright holder. I may give to others the permission to use the modified/redistributed file under GPL-3+. This permission is what is usually called License. In that sense, the license is changed. I think you're mixing up the license of a work, and the effective license of a combined work. A work is an abstract legal thing, dating back to when the most advanced computer available was a monk with an abacus. A file is a computing concept. The law says nothing about files. If various people have contributed bits of a file, my understanding is that the file is a combined work consisting of individual works by those people. To distribute a file that contains one or more works in a way that copyright would not normally allow without the result being illegal, you must get permission from all the copyright holders. A copyright license is just pre-emptive permission from a copyright holder - if you follow these conditions, the answer is yes - so the most common way to get permission from all the copyright holders is to comply with all the conditions imposed by all the copyright holders, simultaneously. For instance, if the file combines a GPL-2+ work with a GPL-3+ work, you must simultaneously comply with both GPL-2 or GPL-3 or some future version and GPL-3 or some future version In this simple case, the second condition implies the first, so you might use a shorthand: this is effectively the same as the whole thing being GPL-3+. However, this is just a shorthand, and the real status is somewhat more complicated. However, the general case is not this simple: http://www.opensource.apple.com/source/xnu/xnu-1456.1.26/bsd/sys/msg.h This is a file that is initially copyrighted by Daniel Boulet (and licensed under BSD-2-Clause). However, without any other change, it also has the header | Copyright (c) 2000-2007 Apple Inc. All rights reserved. | [...] | This file contains Original Code and/or Modifications of Original Code | as defined in and that are subject to the Apple Public Source License | Version 2.0 (the 'License'). You may not use this file except [...] So what you have here is (claimed to be) a file containing a combination of a work by Daniel Boulet, licensed under BSD-2-Clause, and a work by Apple, licensed under APSL-2.0. To distribute that file, assuming that the claim is true, you must simultaneously comply with the conditions of the BSD-2-Clause license (because if you don't, you are infringing Daniel Boulet's copyright), and with the conditions of the APSL-2.0 (because if you don't, you are infringing Apple's copyright). If Apple have not, in fact, modified the file (except to add their license boilerplate, which might not be sufficiently creative to be considered to be a copyrightable work), then their assertion that they hold copyright on the file might not actually be true. If that is the case, then it might be possible (at your own legal risk) to disregard that part, then defend yourself on that basis if they sue you. I wouldn't want to try it myself, because Apple's lawyers are more expensive than I can afford; and there's no real point anyway, because you can presumably just get Daniel Boulet's original version of the file from FreeBSD (?) and avoid the whole issue. If the two licenses are contradictory (one says you must do something and the other says you must not, e.g. GPL-2 and GPL-3, or GPL and OpenSSL) then the combined work is non-distributable. I suspect these two licenses are not contradictory, though - the BSD-2-clause license doesn't require much. Would you accept such a file in Debian? It is clearly not BSD-licensed, even if an unchanged BSD-licensed version exists. If we can exercise the rights demanded by the DFSG while simultaneously complying with both the applicable licenses, then the work is Free. I don't know the precise status of the APSL-2.0 (neither do I particularly want to), but if the two licenses were a pair that I know to be Free and non-contradictory (e.g. BSD-2-clause and GPL), then the ftpmasters would (and frequently do) accept files like that in Debian. S -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/55689e39.5020...@debian.org
Re: GPL + question
Le Fri, May 29, 2015 at 09:32:12AM +0200, Ole Streicher a écrit : I just had a discussion with an ftp-master who rejected one of my packages. The package in question is missfits. It contains a directory, src/wcs/ with files that were originally released by Mark Calabretta under LGPL-2+, but changed by the upstream author (Emmanuel Bertin) and released in the package under GPL-3+. debian/copyright currently mentions only GPL-3+ for the whole package. The ftp-master now asked me to add GPL-2+ for these files to debian/copyright. But I think that this would be wrong, since the files under src/wcs are not distributable under GPL-2+ (because they contain GPL-3+ code from Emmanuel Bertin). Do I miss an important point here? Hi Ole, I am also surprised by this request (isn't there a typo with a L missing in front of GPL-2+ ?). The README in src/wcs contains: This directory contains a modified version of the WCSlib V2.2 library by Mark Calabretta mcala...@atnf.csiro.au, released under the GNU Lesser General Public License. The original version was downloaded from ftp://ftp.cv.nrao.edu/fits/src/wcs/. See http://www.atnf.csiro.au/people/mcalabre/WCS/wcslib for more details. Here, the author of missfits says that he modified the copy of the WCSlib that he redistributes with the sources of missfits. In addition, he added a GPLv3+ header on top of each file. Unfortunately, WCSlib version 2.2 is so old that I could not find a pristine copy on the Internet to confirm that each file was really modified. If it were me, I would give the benefit of the doubt to the upstream author of missfits, and trust him that if he added a GPLv3+ header, it is because he modified the files, as he says in the README. In that case, the license to be indicated in debian/copyright should be GPLv3+. Have a nice week-end, Charles -- Charles Plessy Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150529231946.gd4...@falafel.plessy.net
Re: GPL + question
Only the copyright holder can change what a *work* is licensed as. Unless the copyright holder grants the permission to do so, I would say... Let's say I hold copyright on a work, and I grant someone else permission to change the license of a work. Who would enforce the second license? Only a copyright holder can enforce their copyrights. pgpEAoZd5rFUD.pgp Description: PGP signature
Re: GPL + question
Paul Tagliamonte paul...@debian.org writes: On Fri, May 29, 2015 at 03:09:34PM +0200, Ole Streicher wrote: Same for me. However: the (L)GPL allows even an unmodified redistribution under a later license. This is key -- redistribution. It doesn't change the license. It does. Just look into the license (resp. the header, for simplicity): | you can redistribute it and/or modify it under the terms of the GNU | General Public License as published by the Free Software Foundation; | either version 2 of the License, or (at your option) any later version. So, redistribution may change the license. If I get this file after you say it's GPLv3, it's still LGPLv2.1+ to me if I remove it from other works that change the distribution terms No: We got the files from upstream, and upstream granted us certain permissions for them (this is what the license actually does). Then we are bound to this conditions. We could, ofcourse, get the same files from somewhere else (f.e. from the original author), and these files then can be used under his conditions. But this is a different story. For example, I can give you a file that I got under MIT license, and ask you not to distribute this file -- then you are bound to this, independently whether the file is MIT licensed or not. The only thing you could to is to take the same file from somewhere else and distribute that. Or, as an other example (which is closer to Debian): when I packaged eso-midas, I found that it contained a file sys/msg.h that is originally from NetBSD (with the appropriate license). The specific file in the upstream tarball, however, was relicensed by Apple with the APSL (DFSG incompatible) [1]. Even if the file was the same as the free version, I think that I did it right to remove the file from the tarball and replace it with the original version. I originally thought there was a different question being asked; sorry about that (the terms used and not looking at the source didn't help :)) In my case, the files are modified; so I think there is no doubt that the files are under src/wcs are GPL-3+. Unmodified, the license of the works is unchanged, even if we *distribute* under a different one. Could you put a reference on this? Best regards Ole [1] http://www.opensource.apple.com/source/xnu/xnu-1456.1.26/bsd/sys/msg.h -- To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/ytz617bmq9v@news.ole.ath.cx
Re: GPL + question
On Fri, 29 May 2015 14:50:39 +0200 Ole Streicher wrote: Paul Tagliamonte paul...@debian.org writes: [...] Only the copyright holder can change what a *work* is licensed as. Unless the copyright holder grants the permission to do so, I would say... [...] If the original license allows, then anyone can redistribute the files under a different license. And (L)GPL has a paragraph that allows this under certain conditions (namely LGPL - GPL, and version upgrades). The GNU LGPL v2 (which I understand is the original license for the original files, before they were modified and released under the terms of the GNU GPL v3 or later) states, in section 3: |3. You may opt to apply the terms of the ordinary GNU General Public | License instead of this License to a given copy of the Library. To do | this, you must alter all the notices that refer to this License, so | that they refer to the ordinary GNU General Public License, version 2, | instead of to this License. (If a newer version than version 2 of the | ordinary GNU General Public License has appeared, then you can specify | that version instead if you wish.) Do not make any other change in | these notices. | |Once this change is made in a given copy, it is irreversible for | that copy, so the ordinary GNU General Public License applies to all | subsequent copies and derivative works made from that copy. | |This option is useful when you wish to copy part of the code of | the Library into a program that is not a library. So it seems to me that it is indeed possible to redistribute those files under the terms of the GNU GPL v3 or later, by altering the permission notices so that they refer to the GPL-3+. And this operation is irreversible. Hence, I cannot understand why the FTP masters asked you to add GPL-2+ for these files to debian/copyright, when GPL-2+ seems to have never described the licensing status for those files, before or after their adaptation into the missfits package... I hope this helps to shed some light on this garbled issue. Bye. -- http://www.inventati.org/frx/ There's not a second to spare! To the laboratory! . Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE pgpdIKtqjc06D.pgp Description: PGP signature
Re: GPL question [Was: Re: cdrtools]
On Friday 11 August 2006 18:10 pm, Goswin von Brederlow wrote: I believe that the totaly interchangable option of specifying -static or not should not change the free-ness of the source or resulting binary. So if you link static and you agree that it is a violation that way then you should not be able to get away with it by linking dynamically. The GPL is viral in nature and specificaly made to work across linking boundaries. People should not be able to add non-free portitons to the source by hiding them in libraries. I agree, but then should and is sometimes disagree. But after thinking about it some more, I believe a dynamically linked binary together with the corresponding shared libraries should be considered as a distribution method for the complete program that gets assembled in a common address space. Consider for example the case of EvilCo, back before dynamic linking was widespread, trying to use a GPL'd library in their non-free program. They try to get around the GPL by distributing their compiled program code in a single .o file in a mere aggregate along with the GPL library .a file, and ask users to link the program themselves. This is obviously bogus; they've just created an alternate means of distribution of the resulting binary, and so the binary itself must be distributable under the terms of the GPL, which it isn't. And the case of a dynamically linked executable with shared libraries is almost exactly the same as this scenario, only it's the system dynamic linker doing the work instead of the user doing it manually. Anyway, as somebody else pointed out, this is off-topic for debian-devel, and I apologize. Please direct any replies to debian-legal (too bad kmail doesn't let me set Followup-To afaik). -- Daniel Schepler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL question
Samuel Hocevar [EMAIL PROTECTED]: However, if your printing server component is a library and is GPLed, then every work linked to it has to be GPLed (or have an even less restrictive license). Also, is it relevant that at the moment the whole app. comes on a single CD? This is considered mere aggregation of software by the GPL, and thus the different parts of the work do not need to have the same license, even if there is one GPLed app there. Sorry if this is off-topic, but I'm just checking that I understand the GPL properly. As I understand it, it is relevant that the whole application comes on a single CD, because this is what prevents you from linking a non-GPL program with a GPL library. If you distribute a CD with a GPL library, and a separate CD with a non-GPL program as a separate work, and someone gets both CDs and links the program with the library, then the GPL has been obeyed, because: (i) the GPL library is being distributed according to the GPL; (ii) the non-GPL program doesn't contain any code from the library and is therefore not a derivative work under copyright law; (iii) the GPL only restricts copying, distribution and modification; it does not and could not restrict linking. So my impression is that the GPL is basically equivalent to the LGPL modulo (a significant amount of) inconvenience. If this is wrong, I would like to know why. If it's off-topic, is there another list I could use? Edmund
Re: GPL question
-- Forwarded Message -- Subject: Re: GPL question Date: Tue, 5 Sep 2000 16:13:30 +0100 From: Mike Cunningham [EMAIL PROTECTED] On Tue, 05 Sep 2000, you wrote: snipped my stuff Um.. debian-legal doesn't engage in handing out legal advice. We're focussed on whether something would cause legal problems for debian -- we have no real experience dealing with other legal issues. You might want to contact the author of Ghostscript. Sorry, -- Raul OK Raul, fair enough. I just thought you guys might have thrashed this one out already and have a view you might share if you were asked politely. Please accept my apologies. Mike *** Attachments in this message have been swept by NAI's TVD (version 4.0.4093) for the presence of known computer viruses. * _ This message has been checked for all known viruses by Star Internet delivered through the MessageLabs Virus Control Centre. For further information visit http://www.star.net.uk/stats.asp
Re: GPL question
On Tue, Sep 05, 2000, Mike Cunningham wrote:
I work for a company which sells a proprietary closed-source call centre
application. We are looking to write a central printing server component which
would [hopefully] make use of Ghostscript. I understand that we would need to
release the printing server under the GPL and we have no problem with doing
that.
It depends on how ghostscript is called. If it is just called with
system(); or popen(); then you don't need to make it GPL.
My question is: would the rest of our product need to be re-licensed
under the GPL too?
Again, it depends on how the rest of your product communicates with
the printing server. If they are completely separate programs (ie. one
calling the other with system() or through a pipe), then both can have
their separate license.
However, if your printing server component is a library and is GPLed,
then every work linked to it has to be GPLed (or have an even less
restrictive license).
Also, is it relevant that at the moment the whole app. comes on a single CD?
This is considered mere aggregation of software by the GPL, and
thus the different parts of the work do not need to have the same
license, even if there is one GPLed app there.
I.e. if we added the new print server to the CD then have we just formed a
distribution (as described in the license) and ...aaagh.
Don't worry, as I said, just have a look at the very last sentence of
section 2 of the GPL.
Regards,
Sam.
--
Samuel Hocevar [EMAIL PROTECTED]
http://www.via.ecp.fr/~sam/
1024D/29499F61 1999-04-221155 4B19 A50F 1136 6E60 A499 7CF3 F5AF 2949 9F61
dig goret.org @zoy.org axfr \
| perl -e 'for(sort()){print pack(H32,$1) if(/^c..\.(\w+)/)}' | gzip -d
Re: GPL question
Scripsit Mike Cunningham [EMAIL PROTECTED] I work for a company which sells a proprietary closed-source call centre application. We are looking to write a central printing server component which would [hopefully] make use of Ghostscript. I understand that we would need to release the printing server under the GPL and we have no problem with doing that. Not necessarily. If your printing saver simply generates postscript output, and Ghostscript is just one of several configurable options for what to do with the output, I do not think that the copyright for Ghostscript migrates to the postscript creator. After all, there do exists other quite legitimate options for handling a postscript file than Ghostscript - not least among which is transmitting it to a genuine PostScript(R) printer. Also, is it relevant that at the moment the whole app. comes on a single CD? Not if your application is clearly functional and useful without using Ghostscript (which would be the case if it offered to pipe its output directly to the printer). I.e. if we added the new print server to the CD then have we just formed a distribution (as described in the license) and ...aaagh. I think that would be mere aggregation as described at the end of GPL clause 2. -- Henning MakholmVi skal nok ikke begynde at undervise hinanden i den store regnekunst her, men jeg vil foreslå, at vi fra Kulturministeriets side sørger for at fremsende tallene og også give en beskrivelse af, hvordan man læser tallene. Tak for i dag!
Re: GPL Question
On Thu, 14 Oct 1999, Matthew Simpson wrote: You are free to use and distribute any command string in the Printer Technical Reference. I double checked this with my manager. The only That seems like a pretty straightforward answer to me. What aspect of the law are you worried about violating? They don't need to put their commands under the GPL. Information such as command strings are not subject to copyright law. The only thing that is copyrighted in this case is the actual manual itself. It's much like if you buy a math textbook, you don't have to get permission from the author of the textbook to publish a scientific paper based on formulas you learned out of the book. The only way you would be restricted is if you signed an NDA regarding the information in the manual, which it appears that you did not.
