Re: The GPL and you
Quoting Daniel Isacc Walker ([EMAIL PROTECTED]): It appears that way .. However, something else comes to mind. Some PHP applications are closed source. My code would facilitade a close source PHP script to _use_ talkfilters . But that's not a violated either, even if I keep it closed? I'm not sure I understand that last sentence. Let's recap: You've written some PHP code supporting PHP access to a GPLed library, talkfilters. (You haven't specified what licence you have placed your PHP glue code under.) You are absolutely free to issue your PHP glue code under -- the code you're asking about -- under any licence whatsoever. A person who uses that code does not thereby violate any licence. (Neither the GPL nor the PHP licence regulates software usage.) Third parties who assemble the three codebases (the PHP interpreter, your code, and talkfilters) apparently thereby create a derivative work that is in licence conflict. _Their_ redistribution of the three codebases together would thus seem to constitute (technically) copyright violation, to which the talkfilters author could in theory object. Ok, so the GPL doesn't apply unless I distribute something. Let's be more precise, please: If you accept someone else's work subject to the terms of GPLv2, then your subsequent redistribution of that work (or one that is derivative of it) entails obligations -- about which you can read in the GPL text. Then it only applies if I create a derivative work, being source or binary? Once again: Read the GPL. If you simply distribute your own glue code, you incur no obligations from anyone else's licensing. (Why would you?) If you merely put your glue code on (e.g.) the same CD-ROM as talkfilters and/or the PHP interpreter, and redistribute that, the combination imposes no obligation on you: As GPLv2 puts it, that is mere aggregation. If you distribute your glue code linked with the PHP interpreter and talkfilters, that derivative work's distribution (given that it is in licence conflict) very minimally violates the talkfilters author's copyright. He could theoretically object. Clearer? -- Cheers, Dogs may have kept us company on the hunt, but it was Rick Moen the cats who insisted we invent houses and discover fire. [EMAIL PROTECTED]-- Khiem Tran
Re: The GPL and you
Quoting Daniel Isacc Walker ([EMAIL PROTECTED]):
License conflict? Like it's not illegal but it's a taboo?
No. {sigh}
Once again: Read the GPL.
Under GPLv2 clause 6, your permission to redistribute a covered work
(including derivative works) is conditioned on your imposing no further
restrictions on recipients than GPLv2's own terms. The PHP licence
includes one term that trivially restricts users (regarding the names
of their projects). Thus, derivative works that incorporate PHP and
GPLv2-licensed code have requirements that conflict: We speak of such
works as being in licence conflict.
Since your permission to distribute GPL-encumbered derivative works (such
as your PHP glue code linked to the PHP interpreter and to the
talkfilters library) rests on your NOT imposing further restrictions
beyond those of GPLv2, since you cannot avoid one such restriction, you
may not distribute the derivative work. Your doing so -- or anyone else
doing so who received his copy of talkfilters subject to GPLv2[1] --
constitutes (technically) the tort of copyright violation against the
talkfilters author.
By contrast, distributing your PHP glue code by itself violates nobody
else's rights[2]. Distributing talkfilters by itself violates nobody's
rights. Distributing the PHP interpreter by itself violates nobody's
rights. Distributing all three of them on the same CD-ROM violates
nobody's rights (mere aggregation).
Somebody _using_ any or all of those codebases violates nobody's rights.
(Neither GPLv2 nor the PHP licence regulates code usage.)
[1] You can receive and use a codebase that's GPL-licensed without
accepting that licence: If you've lawfully received any piece of
programming code, you have certain implied rights by statute, including
the right to run it. However, as GPLv2 clause 5 points out, nothing
then gives you the right to redistribute or to create derivative works,
rights that are reserved to the copyright owner by default.
[2] Assuming your permission grant for that code allows that. You've
not mentioned what licence terms you specify for that code.
--
Cheers,Cthulhu loves me, this I know; because the High Priests tell me so!
Rick Moen He won't eat me, no, not yet. He's my Elder God, dank and wet!
[EMAIL PROTECTED]
GNU/LinEx, Debian, and the GNU FDL
On Sun, Aug 31, 2003 at 10:22:33PM -0400, Richard Stallman wrote: I am the one in the FSF who has made statements about Debian and non-free software, and the statements I have made are true as far as I know. If you think anything I said is not true, please show me the statement and the relevant facts. If something I said is incorrect, I will change it. This is a bit of tangent, but your remarks here reminded me of a story I read recently. http://www.ofb.biz/modules.php?name=Newsfile=articlesid=260 RMS provided us with this update on GNU/LinEx: The developers told me that GNU/LinEx included only free software, but after this interview was published various people have checked it and found non-free programs in it. I therefore cannot recommend GNU/LinEx at present. I hope that this problem will be corrected. I have seven questions for you based on this episode: * Wasn't it Debian developers who brought the non-free components of GNU/LinEx to your attention? * Shouldn't you familiarize yourself better with a product before endorsing it (than you did with GNU/LinEx)? * What distribution do you recommend today, given that GNU/LinEx doesn't even segregate non-free into a separate distribution as Debian does? * Have you stopped endorsing Debian GNU/Linux because we're giving you static about the GNU FDL? * If Debian stopped distributing non-free software from its mirror network (and made no mention of it during its install process, and so forth), would you begin endorsing Debian again? * Is your endorsement of Debian GNU/Linux more contingent upon our shipping GNU Manuals in main than it is upon our ceasing to distribute non-free software in non-free? * Isn't it reasonable, in the absence of any statements from you to the contrary, for people to assume that the answer to the preceding question is yes, given that GNU/LinEx doesn't distinguish between free and non-free software anymore than most distributions do, and since the status quo hasn't actually changed in the distribution market, Debian hasn't actually done anything *except* challenge the freeness of GNU FDL to lose your endorsement? -- G. Branden Robinson| There is no gravity in space. Debian GNU/Linux | Then how could astronauts walk [EMAIL PROTECTED] | around on the Moon? http://people.debian.org/~branden/ | Because they wore heavy boots. pgpCqAlJFmDku.pgp Description: PGP signature
Re: A possible GFDL compromise
On Sun, Aug 31, 2003 at 10:22:42PM -0400, Richard Stallman wrote: This is an illuminating comparison, because the practical problems of the GFDL (and I won't claim there are none) are basically of the same kind (though of a lower magnitude) than those of the 4-clause BSD license. (I explained a couple of months ago why they are smaller.) None of us have ever considered saying that the 4-clause BSD license is non-free, or suggesting that programs under such licenses should be removed from Debian main. I've considered it. -- G. Branden Robinson|Somebody once asked me if I thought Debian GNU/Linux |sex was dirty. I said, It is if [EMAIL PROTECTED] |you're doing it right. http://people.debian.org/~branden/ |-- Woody Allen pgpIjxQAeWlG5.pgp Description: PGP signature
UnrealIRCd License (Click-Through issue)
Hi, all! [Full-quote for debian-legal] [Please CC me and [EMAIL PROTECTED] Though I no longer intend to package UnrealIRCd I would like to help the developers in regard to the licensing issues. The main issue for the developers is AFAIK the no warranty clause and how to make it legally binding. Here's the message: * codemastr [EMAIL PROTECTED] [2003-08-31 16:57]: A while ago you had asked about the UnrealIRCd license with regard to Debian. We have a new license that we intend to implement in the very near future. This new license is available at http://www.codemastr.com/EULA.html. I hope this new version will solve all problems with the original license. This is basicaly the following: ---snip--- In order to continue with the download and installation of UnrealIRCd you must accept the following license agreement: [Full copy of GPL] The UnrealIRCd Team reserves the right to modify this agreement at anytime as long as notice is given on the unrealircd.com mainpage at least 24 hours before changes take effect. The new version will only apply to new downloads; those who downloaded UnrealIRCd prior to the changes will still be subject to the version of the EULA that was available at the time of download. By installing, modifying, distributing, downloading, or otherwise using the provided software you agree to be bound by this license agreement. If you do not agree to any/all of the terms stated in this agreement then must not accept this license. ---snip--- And that seems OK to me. Although you can probably restrict yourself to the TERMS AND CONDITIONS part. As for the click-through license, there is a reason for this, in around 2000, the California Supreme Court (at least I believe it was the Supreme Court, I might be remembering wrong), made a decision that most browse to EULAs are not legally binding. The reason was, the user was not actually forced to accept the agreement. The way the judge made it sound, the user could say I do not accept this agreement and still download the software, because there was no actual Do you accept yes/no type of thing. Basically, the ruling said a browse to license is the digital equivilent of trying to tell someone they are bound by a contract they did not sign. The solution to this issue is a click though license which actually tells the user click here to accept, and only then can you download if no is clicked, then the download does not occur. This decision has recently been upheld to some degree by other courts, in October of 2002, the Court of Appeals for the Second Circuit ruled against Netscape in regards to a license similar to this. I found the information at http://caselaw.lp.findlaw.com/scripts/getcase.pl?court=2ndnavby=caseno=017860v2exact=1 An interpretation of this decision I found states: In contrast to downloading programs such as [Netscape] Communicator, downloading SmartDownload is not accompanied by a clickwrap presentation, which requires a user to expressly manifest his or her assent to the terms of the license agreement before downloading a software program. Instead, a user must scroll down below the Download button to view licensing terms governing its use. The appellate court found that having selected to download the program, users were neither required to unambiguously manifest assent to the program's license agreement nor to view the license terms, or even become aware of their existence. After examining California case law and relevant law governing contract agreements, the appellate court held that the act of clicking a download button does not signify assent to a contract's terms if the consumer is not aware of the them. http://www.2000law.com/news_archive.htm Those court rulings basically make it sound as though if the license is not a click through license, then it isn't worth the proverbial paper it is written on. Any comments you or any of the Debian legal team might have would be welcomed. I don't know much about licensing issues. As far as I understand it, nobody has the right to copy, modify or redistibute your software without a license to do so. Thus it doesn't make sense not to be bound by a license agreement because then you don't have a right to copy the software. I hope the debian-legal people can enlighten both of us. :) Cheers, Mika
Re: A possible GFDL compromise
On 2003-09-01 03:22:33 +0100 Richard Stallman [EMAIL PROTECTED] wrote: sections, and spread misinformation about non-free software we distribute. To accuse someone of dishonesty is a grave accusation. To accuse someone of accusing someone of dishonesty is pretty serious, too. You appear to be arguing against an accusation that was not made. Maybe it will be made now, by someone who wishes to create another tangent. Why give them the opportunity? How does it serve your purpose? I really admire your efforts in correcting any inaccurate past statements, but you do sometimes make them and you cannot deny that, surely? Else why would you correct them? It may have happened purely because some of those informing you do not give you correct information at times, but that is misinformation too, even though it is not dishonesty. Yes, misinformation can be honest in a way. None are perfect. Humanity, bah. -- MJR/slef My Opinion Only and possibly not of any group I know. http://mjr.towers.org.uk/ jabber://[EMAIL PROTECTED] Creative copyleft computing services via http://www.ttllp.co.uk/
Re: Decision GFDL
On Sun, Aug 31, 2003 at 01:47:01AM -0400, Walter Landry wrote: Steve Langasek [EMAIL PROTECTED] wrote: On Sun, Aug 31, 2003 at 12:26:04AM -0400, Walter Landry wrote: Based on faulty information, the Release Manager told them not to bother. Now they should bother. Where was this said? The only statement I've seen is that these bugs will not be considered blockers for sarge. Do you mean to say that the maintainers of all the affected packages only fix the RC bugs against their packages, and ignore everything else? That is why they are called release-critical. They have to be fixed before the release. Other bugs do not. The release manager has some discretion to decide that a bug isn't _really_ release critical, but I didn't think he could just ignore the Social Contract. That is my big question, which no one seems to want to answer. Is it ok for the Release Manager to ignore the Social Contract? These documents are not going to become free in the forseeable future. Is it ok for the maintainers of the packages that contain GFDL documentation to ignore the Social Contract? They have also agreed to uphold it; it shouldn't require a mandate from the release manager to get these bugs fixed. (Even treating the bug as RC does not guarantee the Social Contract has been upheld, as it only guarantees the bug will not exist in the release -- possibly by removing the package from testing and leaving it, bugs and all, in unstable.) The severity of these bugs has not been changed; they are still considered serious bugs, and they still need to be fixed. And yet they are somehow not really serious bugs, since serious bugs are, by definition, release-critical. No. Serious bugs are, by definition, violations of must requirements in Debian Policy. This makes them release-critical by *default*, but this is not the first time that a serious bug has been ignored for a release. Nothing stops the maintainers from working on them between now and the freeze date if they have the time for it. Nothing stops you from working on them, if you feel this is important to resolve prior to release. But if no one is willing to work on them, your claim that there won't be a significant delay seems rather ephemeral. The problem is not manpower, it is willpower. gcc, for example, has already done most of the work to fix this bug. All that really needs to be done is for someone with authority to tell them to apply it. Does this mean that the gcc maintainers don't agree with this list's interpretation of the GFDL, or that they don't regard this as a high priority between now and the release? Does the patch have negative side effects that leave the maintainers reluctant to apply it (such as leaving sarge without any gcc manual at all, even in non-free)? -- Steve Langasek postmodern programmer pgpIOw7TcuY2Q.pgp Description: PGP signature
Re: The GPL and you
O Domingo, 31 de Agosto de 2003 ás 13:51:13 -0700, Daniel Isacc Walker escribía: [...] under the GPL . What this means is that my software is automatically GPL'd even though it has no GPL'd source in it. The GPL doesn't distinguish [...] incorporated directly into PHP that means that PHP automatically becomes GPL'd. Even if I made some kind of external module for PHP, PHP would No; that's a common misunderstanding of how the GPL's copyleft works. Linking a work with a GPL-licensed work does not make the first work GPL-licensed. What it really means is that the combination of both works, if it is distributed, it must be under the terms of both licenses simultaneously (each work retains its original license, but the combination...). Now, the GPL has a clause that says you may not impose further restrictions than those imposed by this license (my wording), so if the other work's license has any restrictions not in the GPL, the resulting license is internally inconsistent, so per the GPL, you cannot distribute the resulting work at all. Note how I wrote about distribution. Use is no problem, since the GPL universally allows use (the only restrictions would be those imposed by the other work's license). A common trick which is used to distribute such undistributable combined works consists in distributing the components separately and leaving to the user the task of combining them. -- Tarrío (Compostela)
Re: UnrealIRCd License (Click-Through issue)
Mika Fischer [EMAIL PROTECTED]: The main issue for the developers is AFAIK the no warranty clause and how to make it legally binding. I'm not really sure what it means to make a no warranty clause legally binding. If you are trying to avoid getting sued then you might be better off if you make a sincere effort to inform users of the potential risks rather than rely on some legal gobbledygook to give you magical protection. I don't know which jurisdictions you're interested in, but I've been told that disclaimers don't help much under German law, for example.[*] Common sense says that it is more useful to warn people in ordinary natural language of the real risks specific to a particular product rather than attach the same legalese that nobody reads to every product. Remember, also, that users may be children, who don't understand legalese and cannot make a contract with you. In order to continue with the download and installation of UnrealIRCd you must accept the following license agreement: [Full copy of GPL] Which includes, presumably, the following words: 5. You are not required to accept this License, since you have not signed it. I don't know much about licensing issues. As far as I understand it, nobody has the right to copy, modify or redistibute your software without a license to do so. Thus it doesn't make sense not to be bound by a license agreement because then you don't have a right to copy the software. Yes, but if you download a program, are you copying it, or is the web site copying it, or both? I rather suspect that if you can obtain the data by following a URL (even one containing the string accept=yes) then it might be the web site that is copying, as someone can always claim to have obtained the data by following a URL they saw in IRC rather than by filling in your form. I'm just speculating here. Edmund [*] Still, it's funny how German signs tend to use the word verboten where a British sign would use the word danger. I rather like this style of sign as I once saw on a country walk in England: Danger. Uncleared military target area. Don't touch anything. It might explode and kill you.
Re: A possible GFDL compromise
MJ Ray [EMAIL PROTECTED]: To accuse someone of accusing someone of dishonesty is pretty serious, too. He was probably in a hurry and misunderstood. A polite correction would do. There's no need to start accusing RMS of accusing people of accusing him of dishonesty, not that I want to accuse you of ... anything. :-)
Re: UnrealIRCd License (Click-Through issue)
Hi, Edmund! * Edmund GRIMLEY EVANS [EMAIL PROTECTED] [2003-09-01 12:12]: I'm not really sure what it means to make a no warranty clause legally binding. If you are trying to avoid getting sued then you might be better off if you make a sincere effort to inform users of the potential risks rather than rely on some legal gobbledygook to give you magical protection. While I agree with that I assume they are trying to prevent to be sued when a company uses their software and it stops working because of a bug. cf. GPL clauses 11 and 12. Another point I remember are import/export restrictions which may be in place (for example in France) because the software contains cryptographic code. Common sense says that it is more useful to warn people in ordinary natural language of the real risks specific to a particular product rather than attach the same legalese that nobody reads to every product. Agreed. [Full copy of GPL] Which includes, presumably, the following words: 5. You are not required to accept this License, since you have not signed it. And what does that mean? May I download Emacs, not accept the GPL, use it, run into problems with my business because of using it and then sue the FSF? Yes, but if you download a program, are you copying it, or is the web site copying it, or both? I rather suspect that if you can obtain the data by following a URL (even one containing the string accept=yes) then it might be the web site that is copying, as someone can always claim to have obtained the data by following a URL they saw in IRC rather than by filling in your form. I'm just speculating here. IANAL and I suspect such questions can only be answered by a lawyer or a person with much experience in this area. Cheers, Mika
Some licensing questions regarding celestia
Hi! [Please CC me] I'm the new maintainer of celestia which is a space simulation program. As such it contains a lot of data, numerical data such as positions of stars as well as 3D models and textures. The copyright status of all this data is a real mess and we (the authors and me) are trying to clarify it. What I got out of the whole FDL debate is that data in Debian has to be DFSG-free (with which I agree). celestia is released under the terms of the GPL. Now the questions: 1) If one includes public-domain material in a GPL work, does one have to state what material is in the public domain? 2) Are there any GPL-compatibility issues when the data is licensed differently from the GPL? So if an author grants the rights to copy, modify and redistribute is it enough to basicaly say: This software is GPLed but file xyz is licensed according to the following statement: ...? Does a list like the following suffice? ---snip--- Filename: xyz Author: Name Terms: blah... Filename: abc ... ---snip--- 3) What rights do need to be granted for data to be included in non-free. In particular what about the following: ---snip--- JPL Image Policy JPL images are available for use by the public free of charge. However, by electing to download images from this web site the user agrees that Caltech makes no warranties or representations with respect to its ownership of copyrights for the images, does not represent others who may claim to be owners of rights in the images, and makes no warranties as to the quality of the images. Commercial users (excluding journalistic uses) are required to copy the JPL Image Release document and return a signed copy to the Caltech's Intellectual Property Counsel, California Institute of Technology M/C 201-85, Pasadena, California 91125, who will countersign document and return a copy to you. Copies may be faxed to (626) 577-2528. This document will become effective when it is countersigned by Caltech. ---snip--- TIA, Mika
Re: Is the Nokia Open Source License DFSG compliant?
Kevin Rosenberg [EMAIL PROTECTED] wrote: I believe this license is DFSG compliant, Sections 3.1 and 3.2 are similar to some GPL sections. I wonder about section 3.6 as well. The one thing that worries me is 3.2 Availability of Source Code. Any Modification which You create or to which You contribute must be made available in Source Code form under the terms of this License either on the same media as an Executable version or via an accepted Electronic Distribution Mechanism to anyone to whom you made an Executable version available; and if made available via Electronic Distribution Mechanism, must remain available for at least twelve (12) months after the date it initially became available, or at least six (6) months after a subsequent version of that particular Modification has been made available to such recipients. You are responsible for ensuring that the Source Code version remains available even if the Electronic Distribution Mechanism is maintained by a third party. So if Debian puts the source and binaries up on its ftp site, it is required to keep the source there for at least 12 months, and at least 6 months after a particular version. We can speculate about whether this is DFSG-free or not, but it certainly can't be distributed by Debian. Debian does not keep around the source to all of the old versions for at least six months. Also, in relation to the patent stuff. There was a similar kind of problem in an Intel driver license http://lists.debian.org/debian-legal/2002/debian-legal-200202/msg00032.html Basically, if there was a patent that applied to the Intel driver, then the whole thing wouldn't be GPL incompatible (which is a problem, since the driver links into the kernel). Intel has a lot of patents, but the conclusion I got from the discussion was that we don't worry unless they start enforcing patents related to the driver. This is basically what Andrew Suffield is saying. Regards, Walter Landry [EMAIL PROTECTED]
Re: Some licensing questions regarding celestia
On 2003-09-01 14:50:40 +0100 Mika Fischer [EMAIL PROTECTED] wrote: 1) If one includes public-domain material in a GPL work, does one have to state what material is in the public domain? I'm not sure, but would say yes. 2) Are there any GPL-compatibility issues when the data is licensed differently from the GPL? This is covered in the GPL FAQ about aggregation I think. 3) What rights do need to be granted for data to be included in non-free. I'm not sure. That it can be distributed by the ftp mirrors, I guess. Does it have to be configured specifically for use with Debian or could users just go get it themselves, instead of using our ftp mirror sponsors' connectivity? I didn't see anything in the licence about derived works: is that covered by use? -- MJR/slef My Opinion Only and possibly not of any group I know.
Re: UnrealIRCd License (Click-Through issue)
Mika Fischer [EMAIL PROTECTED]: 5. You are not required to accept this License, since you have not signed it. And what does that mean? May I download Emacs, not accept the GPL, use it, run into problems with my business because of using it and then sue the FSF? Of course. You have the right to sue anyone for anything at any time! However, in their defence the FSF will probably use the following elements of the GPL as evidence that they were not negligent: 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, ... If someone tries to sue you for distributing FSF software, you can point at the same parts of the GPL and also at the warning on your web site, if you have one. I suspect that in general it isn't particularly helpful to make people agree to the GPL before downloading GPL software. Since the GPL isn't a contract it's not clear what you expect people to agree to, and since the GPL allows people to redistribute the software the person who sues you might not have downloaded the software themselves. Even if they did download it themselves, unless you get a signature from them they can always claim to have received it second-hand. If an aeroplace manufacturer deliberately missed out half the screws in order to save money and as a result the wings fell off, would you expect them to be any less liable for negligence just because you bought the aeroplace second-hand?
Re: UnrealIRCd License (Click-Through issue)
Hi, Edmund! * Edmund GRIMLEY EVANS [EMAIL PROTECTED] [2003-09-01 17:03]: Of course. You have the right to sue anyone for anything at any time! Oh well, I think you know what I meant. :) However, in their defence the FSF will probably use the following elements of the GPL as evidence that they were not negligent: 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, ... However I decided not to accept the license and may not even have read it. If someone tries to sue you for distributing FSF software, you can point at the same parts of the GPL and also at the warning on your web site, if you have one. The whole point of my hypothetical example is that I don't accept the license and use the software nevertheless. How could the GPL then be of any help? I suspect that in general it isn't particularly helpful to make people agree to the GPL before downloading GPL software. Since the GPL isn't a contract it's not clear what you expect people to agree to, and since the GPL allows people to redistribute the software the person who sues you might not have downloaded the software themselves. Even if they did download it themselves, unless you get a signature from them they can always claim to have received it second-hand. If an aeroplace manufacturer deliberately missed out half the screws in order to save money and as a result the wings fell off, would you expect them to be any less liable for negligence just because you bought the aeroplace second-hand? But what does that mean for the case we're discussing? Is what you're saying esantially: 1) Click-Through or just GPL doesn't make a difference. 2) The GPL no-warranty statement is useless because one does not have to agree to the license. ? 2) would seem strange to me because the same would be true about M$-EULAs and certainly someone would have sued... So the question basically remains: How can the UnrealIRCd team best disclaim warranty? Does a click-through license help with that? If not what about the rulings in my first mail? Cheers, Mika
Re: Some licensing questions regarding celestia
Hi, MJ! Thanks for the pointer to the GPL FAQ. I've read it and found some things... * MJ Ray [EMAIL PROTECTED] [2003-09-01 16:49]: On 2003-09-01 14:50:40 +0100 Mika Fischer [EMAIL PROTECTED] wrote: 1) If one includes public-domain material in a GPL work, does one have to state what material is in the public domain? I'm not sure, but would say yes. http://www.fsf.org/licenses/gpl-faq.html#CombinePublicDomainWithGPL indirectly implies that public domain code does not have to be marked as such. 2) Are there any GPL-compatibility issues when the data is licensed differently from the GPL? This is covered in the GPL FAQ about aggregation I think. No it isn't, I'm afraid. http://www.fsf.org/licenses/gpl-faq.html#MereAggregation only deals with code. The impression I get from reading the FAQ is that data does not have to be free at all. http://www.fsf.org/licenses/gpl-faq.html#TOCIfInterpreterIsGPL states: a free software license like the GPL, based on copyright law, cannot limit what data you use the interpreter on. Cheers, Mika
Re: UnrealIRCd License (Click-Through issue)
* Mika Fischer [EMAIL PROTECTED] [2003-09-01 10:53]: And that seems OK to me. Although you can probably restrict yourself to the TERMS AND CONDITIONS part. I'm sorry but this is wrong. You have to include the whole GPL as you did. cf. http://www.fsf.org/licenses/gpl-faq.html#GPLOmitPreamble Cheers, Mika
Re: Some licensing questions regarding celestia
On 2003-09-01 17:00:57 +0100 Mika Fischer [EMAIL PROTECTED] wrote: http://www.fsf.org/licenses/gpl-faq.html#CombinePublicDomainWithGPL indirectly implies that public domain code does not have to be marked as such. Yes, as long as you are comfortable with the appearance of GPLing it, which most people are. 2) Are there any GPL-compatibility issues when the data is licensed differently from the GPL? This is covered in the GPL FAQ about aggregation I think. No it isn't, I'm afraid. http://www.fsf.org/licenses/gpl-faq.html#MereAggregation only deals with code. Bah. It also applies in general to all software. The start of the answer is unnecessarily specific to programs. Anyway, 2 down, 1 to go. Best of luck.
Re: A possible GFDL compromise
On Sun, Aug 31, 2003 at 08:52:29PM -0700, Don Armstrong wrote: Luckily, we only have a few (one?) large components of Debian that are under a 4-clause BSD license [OpenSSL]. I for one, will be glad when gnutls completely supplants the use of openssl. This may change, if ftpmasters accept netbsd-kernel-source-* packages once they've stabilized enough to be worth packaging. Conversely, however, I'm currently putting in a great deal of manual effort to hunt down holders of copyrights in the source tree that have a 4-clause license, try to convince them to relicense, and get those changes committed to the NetBSD core tree. So far, there are approximately 5000 fewer files (out of around 15,000; mostly pure UCB text that never got updated) with a 4-clause BSD license, if you consider all pending patches as well as those already confirmed and applied. Some large portion of the rest are held under the copyright of The NetBSD Foundation (rather than individual authors), and can be removed in one fell swoop if and when TNF officially decides to permit it. That question has been under discussion for many months, at this point, and for all I know may be discussed for many more, but it *is* at least under discussion (and some number of NetBSD developers that I contacted about other occurances have said they're refusing to assign copyright unless and until TNF switches to a 3-clause license). RMS left CC'ed largely because he might find it intriguing, given that NetBSD was the concrete example given in one of the origional arguments. -- Joel Baker [EMAIL PROTECTED],''`. Debian GNU NetBSD/i386 porter: :' : `. `' `- pgpE6bvrtZStG.pgp Description: PGP signature
Re: Is the Nokia Open Source License DFSG compliant?
On Mon, Sep 01, 2003 at 10:09:00AM -0400, Walter Landry wrote: 3.2 Availability of Source Code. Any Modification which You create or to which You contribute must be made available in Source Code form under the terms of this License either on the same media as an Executable version or via an accepted ^^ Electronic Distribution Mechanism to anyone to whom you made an Executable version available; and if made available via Electronic Distribution Mechanism, must remain available for at least twelve (12) months after the date it initially became available, or at least six (6) months after a subsequent version of that particular Modification has been made available to such recipients. You are responsible for ensuring that the Source Code version remains available even if the Electronic Distribution Mechanism is maintained by a third party. So if Debian puts the source and binaries up on its ftp site, it is required to keep the source there for at least 12 months, and at least 6 months after a particular version. We can speculate about whether this is DFSG-free or not, but it certainly can't be distributed by Debian. Debian does not keep around the source to all of the old versions for at least six months. This problem is in the MPL too. So far we've been assuming that on the same media also covers FTP sites that host both source and binaries, and that this clause only kicks in if you distribute binary-only versions offline. One thing I haven't thought of before is that, apparently, you *must* supply an Executable version. If you supply only the source then you must keep it available for 6-12 months. That's probably a bug in the wording. Hmm, it also seems that you _must_ publish source code even if you only intend to use a modification privately. None of this might matter because as far as I can tell, 2.1(a) grants enough permissions to render the rest of the license meaningless. I don't think the MPL was ever properly reviewed here :( Richard Braakman
Re: UnrealIRCd License (Click-Through issue)
Mika Fischer [EMAIL PROTECTED]: The whole point of my hypothetical example is that I don't accept the license and use the software nevertheless. How could the GPL then be of any help? People don't have to accept the GPL unless they are redistributing. I would guess that the GPL helps because, by including the disclaimer in the licence and making permission to redistribute conditional on including the licence, it shows that you have made a valiant effort to make certain that everyone understands that there is no warranty. But what does that mean for the case we're discussing? Is what you're saying esantially: 1) Click-Through or just GPL doesn't make a difference. 2) The GPL no-warranty statement is useless because one does not have to agree to the license. ? What I'm saying is that I'm not convinced that click-through is any better than just putting a one-sentence warning with a reference to the GPL, and the latter might even be better, as people are more likely to read and understand the one-sentence warning than they are to read and understand the GPL. 2) would seem strange to me because the same would be true about M$-EULAs and certainly someone would have sued... I don't think EULAs have any validity as contracts. If I buy a second-hand computer, how can I be bound by a document that even the previous owner of the computer didn't sign? And even for the person who clicks through, where is the consideration? Assuming it is even possible for them to surrender their right to sue for damages, what is the software publisher giving them in return? The right to use their own property? Anyway, you can also ask, if EULAs really work, why don't they use them in cars? You could put a load of small print and the words I agree next to the ignition. Brilliant. I should patent the idea. So the question basically remains: How can the UnrealIRCd team best disclaim warranty? Does a click-through license help with that? If not what about the rulings in my first mail? The click-through licence probably doesn't hurt, but it might be better to give a plaintext warning along the lines of: This free software is provided with no warranty. Download it and use it at your own risk. See the alicence/a for details. However, if you want to be sure, it's best to ask a lawyer in the jurisdictions where you own property ...
Re: Some licensing questions regarding celestia
On 2003-09-01 17:42:28 +0100 Mika Fischer [EMAIL PROTECTED] wrote: Bah. It also applies in general to all software. Is data a subset of software? In general, no. In this case, yes, assuming we are only talking about things that will be uploaded to Debian. [...] Thanks, but it's probably going to be a long way because of all the user-contributed data... I'm sure some will appreciate the effort.
Re: The GPL and you
On Sun, 31 Aug 2003, Anthony DeRobertis wrote: On Sunday, Aug 31, 2003, at 18:54 US/Eastern, Daniel Isacc Walker wrote: I use talkfilters data structures and API in my code. Is that a derivate work? Everything else I've read indicates that it isn't.. We had a rather recent discussion about this, I believe in the context of perl's dbd mysql stuff. In short, I happen to agree with you; others on this list may agree or disagree, but Debian's policy seems to be paranoia: Assume that is a derivative work, in part because the FSF says so. That's interesting. In the Wine community the majority opinion is that the Win32 API and data structures are not copyrightable. My opinion is derived mainly from my work with Wine. Since the Win32 API is so vast (100's of functions and structures) I would think if header files are copyrightable then the Win32 headers must be copyrighted by Microsoft. As far as I know, Microsoft hasn't attacked Wine for using the API . I could be mixing two distinct ideas here. Copyrighting a header file could be completely different from copyrighting an API .. Ever though they seem like the same thing. Daniel Walker
Re: A possible GFDL compromise
On Sun, 31 Aug 2003 22:22:42 -0400 Richard Stallman [EMAIL PROTECTED] wrote: I, and, to a large extent, other members of this list, are concerned that, beyond the non-trivial freedom aspects, texts under the GFDL will begin to suffer the same fate that code licensed under the 4-clause BSD license has. This is an illuminating comparison, because the practical problems of the GFDL (and I won't claim there are none) are basically of the same kind (though of a lower magnitude) than those of the 4-clause BSD license. (I explained a couple of months ago why they are smaller.) None of us have ever considered saying that the 4-clause BSD license is non-free, or suggesting that programs under such licenses should be removed from Debian main. That begs the question, are the practical problems of the GFDL basically the same (though of lesser magnitude) than those of the 4-clause BSD license? :) It's pretty apparent that the majority of the people here disagree with you about that, and have presented many, many, many points to support their position. pgpt1xWu8r9zH.pgp Description: PGP signature
Re: Some licensing questions regarding celestia
Hi, again! * Mika Fischer [EMAIL PROTECTED] [2003-09-01 15:50]: I'm the new maintainer of celestia which is a space simulation program. As such it contains a lot of data, numerical data such as positions of stars as well as 3D models and textures. The copyright status of all this data is a real mess and we (the authors and me) are trying to clarify it. OK, this is getting better and better. How far does one have to go in regard to data? A few examples. - Data published on the web: http://www.obspm.fr/encycl/cat1.html lists stars with possible planets around them. Is one allowed to use this data in a program? Basically for me this is just information and it doesn't make sense to restrict that. - Data announced on the web: http://www.eso.org/outreach/press-rel/pr-2001/pr-07-01.html which leads to: http://obswww.unige.ch/~udry/planet/new_planet.html which has the data. - If one creates a 3D Model from published data, what is the legal status of this work? Can it be GPLed? Can it be put in the public domain? These are all for now. I'm sure more esoteric examples will eventually show up. Cheers, Mi 'And I haven't even started with the images!' ka
Re: GPL preamble removal
* Glenn Maynard [EMAIL PROTECTED] [2003-09-01 19:16]: You can remove the preamble. http://www.fsf.org/licenses/gpl-faq.html#ModifyGPL Yes, but that creates a different license and that is not wanted in this case. CC RMS: could this please be clarified? It's causing repeat confusion, and nobody knows if we're really allowed to remove the preamble and/or create derivative licenses of the GPL. I think the following is quite clear: http://www.fsf.org/licenses/gpl-faq.html#GPLOmitPreamble ---snip--- The preamble and instructions are integral parts of the GNU GPL and may not be omitted. In fact, the GPL is copyrighted, and its license permits only verbatim copying of the entire GPL. (You can use the legal terms to make another license but it won't be the GNU GPL.) ---snip--- which links to: http://www.fsf.org/licenses/gpl-faq.html#ModifyGPL Cheers, Mika
Re: UnrealIRCd License (Click-Through issue)
IANAL, TINLA On Mon, Sep 01, 2003 at 05:28:41PM +0200, Mika Fischer wrote: Hi, Edmund! * Edmund GRIMLEY EVANS [EMAIL PROTECTED] [2003-09-01 17:03]: Of course. You have the right to sue anyone for anything at any time! Oh well, I think you know what I meant. :) However, in their defence the FSF will probably use the following elements of the GPL as evidence that they were not negligent: 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, ... However I decided not to accept the license and may not even have read it. That doesn't change anything. Quote from the GPL: 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. [...] 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. So, even if you do not accept the license but you do copy, modify, and/or distribute the Program, you're still bound by the License. If someone tries to sue you for distributing FSF software, you can point at the same parts of the GPL and also at the warning on your web site, if you have one. The whole point of my hypothetical example is that I don't accept the license and use the software nevertheless. Not accepting the GPL is not a way to avoid it. You would be using a copy of the program without the right to do so; it would be the Free Software-equivalent of software piracy. If you want to be 100% sure, I suggest you go to a solicitor with a copy of the GPL; but I don't think a click-through license is necessary. [...] -- Wouter Verhelst Debian GNU/Linux -- http://www.debian.org Nederlandstalige Linux-documentatie -- http://nl.linux.org Stop breathing down my neck. My breathing is merely a simulation. So is my neck, stop it anyway! -- Voyager's EMH versus the Prometheus' EMH, stardate 51462. pgpuoIIMKM4mv.pgp Description: PGP signature
Re: UnrealIRCd License (Click-Through issue)
Hi! * Wouter Verhelst [EMAIL PROTECTED] [2003-09-01 20:39]: So, even if you do not accept the license but you do copy, modify, and/or distribute the Program, you're still bound by the License. What about use? I think that's the most important one here. Not accepting the GPL is not a way to avoid it. You would be using a copy of the program without the right to do so; it would be the Free Software-equivalent of software piracy. Could you then comment on the quoted comment on a ruling in my first email? ---snip--- In order to resolve the question of arbitration, the key issue the Second Circuit had to decide was whether plaintiffs, by downloading the free software, agreed to be bound to Netscape's license agreement, which included the arbitration clause. [...] The appellate court found that having selected to download the program, users were neither required to unambiguously manifest assent to the program's license agreement nor to view the license terms, or even become aware of their existence. After examining California case law and relevant law governing contract agreements, the appellate court held that the act of clicking a download button does not signify assent to a contract's terms if the consumer is not aware of the them. Therefore, users are not bound by the SmartDownload license agreement since a reasonably prudent consumer would not have been aware of its existence. ---snip--- Full text:http://www.2000law.com/news_archive.htm Ruling: http://caselaw.lp.findlaw.com/scripts/getcase.pl?court=2ndnavby=caseno=017860v2exact=1 Thanks, Mika pgpgK2dTi61Rg.pgp Description: PGP signature
Re: UnrealIRCd License (Click-Through issue)
On Mon, Sep 01, 2003 at 08:39:19PM +0200, Wouter Verhelst wrote: That doesn't change anything. Quote from the GPL: 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. [...] 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. So, even if you do not accept the license but you do copy, modify, and/or distribute the Program, you're still bound by the License. s/you're still bound by the License/you can be countersued for\ copyright infringement *or* license violation if you try to sue under\ warranty law/ If someone tries to sue you for distributing FSF software, you can point at the same parts of the GPL and also at the warning on your web site, if you have one. The whole point of my hypothetical example is that I don't accept the license and use the software nevertheless. Not accepting the GPL is not a way to avoid it. You would be using a copy of the program without the right to do so; it would be the Free Software-equivalent of software piracy. In fact, use rights are not within the scope of copyright in most jurisdictions; so even if you received no license text with a copy of a given work, you would still have the right to use the work -- and in many jurisdictions, you would also have the right to sue the author under warranty law if something went horribly wrong while using the software. So the GPL itself does not directly protect against lawsuits from disgruntled users, which is actually the strongest argument in favor of GPL 2(c). In any case, the GPL is not well suited for use as a click-through, because most of the information contained in the GPL does not pertain to end users at all. What you would really want is something much more limited, such as a disclaimer as described in 2(c) together with an 'accept' button. FWIW, to date I'm not aware of any end user suing the author of a GPL work under warranty law. -- Steve Langasek postmodern programmer pgpTj6g2VdFaU.pgp Description: PGP signature
Re: GPL preamble removal
On Mon, Sep 01, 2003 at 08:35:24PM +0200, Mika Fischer wrote: CC RMS: could this please be clarified? It's causing repeat confusion, and nobody knows if we're really allowed to remove the preamble and/or create derivative licenses of the GPL. I think the following is quite clear: http://www.fsf.org/licenses/gpl-faq.html#GPLOmitPreamble ---snip--- The preamble and instructions are integral parts of the GNU GPL and may not be omitted. In fact, the GPL is copyrighted, and its license permits only verbatim copying of the entire GPL. (You can use the legal terms to make another license but it won't be the GNU GPL.) ---snip--- which links to: http://www.fsf.org/licenses/gpl-faq.html#ModifyGPL This is not clear at all. The preamble is integral and may not be removed, and the license permits only verbetim copying, but you can make changes as long as you remove the preamble, even though it just said you can do neither. What? Is it trying to say that you can remove the preamble as long as you also make some kind of modification to the license text? That would be strange (and pointless, I think; change some whitespace.) -- Glenn Maynard
Re: The GPL and you
Quoting Daniel Isacc Walker ([EMAIL PROTECTED]): That's interesting. In the Wine community the majority opinion is that the Win32 API and data structures are not copyrightable. My opinion is derived mainly from my work with Wine. Since the Win32 API is so vast (100's of functions and structures) I would think if header files are copyrightable then the Win32 headers must be copyrighted by Microsoft. As far as I know, Microsoft hasn't attacked Wine for using the API . I could be mixing two distinct ideas here. Copyrighting a header file could be completely different from copyrighting an API .. Ever though they seem like the same thing. I think you may have a point. During the ATT vs. UC Regents lawsuit, part of UC's defence against charges that BSD incorporated ATT UNIX 32V's header files was that was functional code required for compatibility, so that code written for 32V could run on BSD without having to replicate all the header files' definitions and declarations. Copyright law has traditionally given considerable leeway for compatibility requirements, and generally only expressive code qualifies for copyright coverage, not functional code. I would tend to think that the talkfilters data structures and API are not copyrightable. (But IANAL, TINLA.) -- Cheers, find / -user your -name base -print | xargs chown us:us Rick Moen [EMAIL PROTECTED]
Re: UnrealIRCd License (Click-Through issue)
IANAL, TINLA. On Mon, 01 Sep 2003, Mika Fischer wrote: Could you then comment on the quoted comment on a ruling in my first email? This particular ruling is rather interesting, because the main issue was the ability of the license agreement to compel arbitration, rather than the generic warranty disclaimer. That is, the issue of the applicability of Communicator's no warranty clause to the issues at hand was not or had not been raised upon appeal. Regardless, the question still has to be asked: Is the notice of license used in a package or program sufficient to place consumers on notice of the terms of said license? I'd personally argue that the inclusion of a LICENSE file in the source distribution, or a /usr/share/doc/pacakgename/copyright is enough, at the very least, to put consumers on notice of the no warranty clause pertaining to use of the software. Furthermore, the absence of monetary exchange (eg, payment for the software) tends to lend credence to the warranty waiver. Finally, it could rather trivially be argued that the use of the program establishes a contractual relationship between a consumer and the developer or distributor. If the consumer were to argue that he/she had not consented to the license, then a countersuit under copyright law is a logical conclusion. If you are seriously concerned about the warranty clause, etc. please don't hesitate to talk to an attorney or solicitor. Don Armstrong -- If you wish to strive for peace of soul, then believe; if you wish to be a devotee of truth, then inquire. -- Friedrich Nietzsche http://www.donarmstrong.com http://www.anylevel.com http://rzlab.ucr.edu pgpUfNH7k1BhD.pgp Description: PGP signature
Re: Is the Nokia Open Source License DFSG compliant?
On Mon, Sep 01, 2003 at 07:29:42PM +0300, Richard Braakman wrote: I don't think the MPL was ever properly reviewed here :( I don't think it was, but IMO it is not a DFSG-compatible license. -- G. Branden Robinson|The basic test of freedom is Debian GNU/Linux |perhaps less in what we are free to [EMAIL PROTECTED] |do than in what we are free not to http://people.debian.org/~branden/ |do. -- Eric Hoffer pgprdWQorZbBu.pgp Description: PGP signature
Re: GPL preamble removal
On Mon, Sep 01, 2003 at 03:13:14PM -0400, Glenn Maynard wrote: http://www.fsf.org/licenses/gpl-faq.html#GPLOmitPreamble [...] http://www.fsf.org/licenses/gpl-faq.html#ModifyGPL This is not clear at all. The preamble is integral and may not be removed, and the license permits only verbetim copying, but you can make changes as long as you remove the preamble, even though it just said you can do neither. What? Is it trying to say that you can remove the preamble as long as you also make some kind of modification to the license text? That would be strange (and pointless, I think; change some whitespace.) I asked for clarification on this point months ago. http://lists.debian.org/debian-legal/2003/debian-legal-200305/msg00091.html From: Branden Robinson [EMAIL PROTECTED] To: [EMAIL PROTECTED] Cc: debian-legal@lists.debian.org Subject: the GPL FAQ on the copyrightability of the GPL's text Date: Wed, 7 May 2003 01:07:06 -0500 Message-ID: [EMAIL PROTECTED] On Thu, 12 Jun 2003 13:09:52 -0400, I received the following response from Brett Smith: I am very sorry for the late response to this message. Since we are a non-profit with very limited resources, messages to this address often get backlogged, and we are always struggling to catch up. Thank you for taking the time to write to us with your concerns about this. I've brought the ambiguity to the attention of Richard Stallman, Bradley Kuhn, and Dave novalis Turner; hopefully we will be able to resolve the matter soon, and I will let you know when we do. I've heard nothing since then. -- G. Branden Robinson| Yesterday upon the stair, Debian GNU/Linux | I met a man who wasn't there. [EMAIL PROTECTED] | He wasn't there again today, http://people.debian.org/~branden/ | I think he's from the CIA. pgplfFgVBUxtq.pgp Description: PGP signature
Re: UnrealIRCd License (Click-Through issue)
On Mon, Sep 01, 2003 at 09:01:35PM +0200, Mika Fischer wrote: Hi! * Wouter Verhelst [EMAIL PROTECTED] [2003-09-01 20:39]: So, even if you do not accept the license but you do copy, modify, and/or distribute the Program, you're still bound by the License. What about use? I think that's the most important one here. Section 0. Read it, for a change :-) (Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, [...]) Not accepting the GPL is not a way to avoid it. You would be using a copy of the program without the right to do so; it would be the Free Software-equivalent of software piracy. Could you then comment on the quoted comment on a ruling in my first email? ---snip--- In order to resolve the question of arbitration, the key issue the Second Circuit had to decide was whether plaintiffs, by downloading the free software, agreed to be bound to Netscape's license agreement, which included the arbitration clause. [...] The appellate court found that having selected to download the program, users were neither required to unambiguously manifest assent to the program's license agreement nor to view the license terms, or even become aware of their existence. After examining California case law and relevant law governing contract agreements, the appellate court held that the act of clicking a download button does not signify assent to a contract's terms if the consumer is not aware of the them. Therefore, users are not bound by the SmartDownload license agreement since a reasonably prudent consumer would not have been aware of its existence. ---snip--- That may be true; however, such a ruling should not imply that a user is free to do whatever he likes with the software. If the user does not choose to accept the license, he still is bound by copyright law. Copyright law does not give a user many options. Depending on his jurisdiction, he may have the right to use the program; he may have the right to install it. He probably does not have the right to copy or modify it. He most likely does not have the right to redistribute or sublicense it. These rights are in most jurisdictions by default only available to the author (you, in this case); only if you, as an author, explicitely give permission to copy, modify, redistribute or sublicense, only then can a user reasonably assume he has the right to perform these actions. Therefore, if he does not accept the license, he can do next to nothing with it. His only real option is to accept the license, and live by it. This also excludes ignorance about the license text, since if the user does not know about the license, he should assume his default rights, as specified by copyright law, apply, as opposed to anything he would want to, and technically could, do with the software. Of course, this does not protect you against the ruling of a braindead judge who should not have been a judge in the first place, but nothing can. Of course, IANAL, TINLA. Contact a solicitor if you want a certain level of competence. -- Wouter Verhelst Debian GNU/Linux -- http://www.debian.org Nederlandstalige Linux-documentatie -- http://nl.linux.org Stop breathing down my neck. My breathing is merely a simulation. So is my neck, stop it anyway! -- Voyager's EMH versus the Prometheus' EMH, stardate 51462. pgp5pqYDQi6qe.pgp Description: PGP signature
Re: Some licensing questions regarding celestia
On Mon, Sep 01, 2003 at 03:50:40PM +0200, Mika Fischer wrote: Now the questions: 1) If one includes public-domain material in a GPL work, does one have to state what material is in the public domain? At least in the U.S., to my knowledge, this is not *required*. I can quote entire plays' worth of Shakespeare without even hinting as to their source or copyright status, and need fear no copryight infrigement claims. However, when mixing copyrighted and public domain materials, it might be *wise* to indicate which materials are in the public domain. But this is often not done -- just look at the Disney corporation. 2) Are there any GPL-compatibility issues when the data is licensed differently from the GPL? So if an author grants the rights to copy, modify and redistribute is it enough to basicaly say: This software is GPLed but file xyz is licensed according to the following statement: ...? Does a list like the following suffice? ---snip--- Filename: xyz Author: Name Terms: blah... Filename: abc ... ---snip--- Ideally, the answer to this question would be simple; any *data* a GPLed *program* operates on can be licensed however one likes. However, certain transformations of data by a GPLed program can make what should be simple more complex. For instance, libgcc.a gets linked into object files produced by GCC by default, so the FSF had to add special rider conditions ensuring that while GCC itself is GPLed, it is permissible to link libgcc against anything. (It has never been, to my knowledge, the FSF's intention to ensure that one could only produce GPLed software with GCC.) In the general case, I'd say the license of the data doesn't matter. However, whether this holds for celestia depends on what the program does. 3) What rights do need to be granted for data to be included in non-free. In particular what about the following: ---snip--- JPL Image Policy JPL images are available for use by the public free of charge. However, by electing to download images from this web site the user agrees that Caltech makes no warranties or representations with respect to its ownership of copyrights for the images, does not represent others who may claim to be owners of rights in the images, and makes no warranties as to the quality of the images. Commercial users (excluding journalistic uses) are required to copy the JPL Image Release document and return a signed copy to the Caltech's Intellectual Property Counsel, California Institute of Technology M/C 201-85, Pasadena, California 91125, who will countersign document and return a copy to you. Copies may be faxed to (626) 577-2528. This document will become effective when it is countersigned by Caltech. ---snip--- I'm not sure we have a written what makes a work distributable in non-free policy. A significant amount of the stuff in non-free would be DFSG-free if it weren't for a discriminatory clause like for non-commercial use only. I *think*, as long as Debian and its mirror network can redistribute the work without charge or royalty, it meets the fundmental test for distributability in non-free -- patent, trademark, trade secret, and other issues notwithstanding. I'm sorry these answers are more gray than you were probably hoping for. -- G. Branden Robinson|I have a truly elegant proof of the Debian GNU/Linux |above, but it is too long to fit [EMAIL PROTECTED] |into this .signature file. http://people.debian.org/~branden/ | pgpc0vZ0cigA9.pgp Description: PGP signature
Re: A possible GFDL compromise
Richard Stallman wrote: This is an illuminating comparison, because the practical problems of the GFDL (and I won't claim there are none) are basically of the same kind (though of a lower magnitude) than those of the 4-clause BSD ^^^ Replace this with greater magnitude, and you'll have my opinion on the practical problems (as opposed to the freeness problems). This is because the 4-clause BSD license only directly imposed a burden on people who chose to advertise, while the GFDL imposes corresponding burdens on all users. Perhaps you feel that the advertising burden is the greater burden? license. (I explained a couple of months ago why they are smaller.) Your explanations were unconvincing. As far as I could tell, they basically amounted to People don't need to modify documentation as much as they do software, which most of this list vehemently disagrees with, and People will behave themselves, which is unlikely. If I've missed an aspect of your explanation, please repeat it.
Re: Some licensing questions regarding celestia
On Mon, Sep 01, 2003 at 06:42:28PM +0200, Mika Fischer wrote: Is data a subset of software? For the purposes of testing its license terms (if any) against the Debian Free Software Guidelines and distributing it from our archives, yes. For other purposes, please retain a philosopher. :) -- G. Branden Robinson| I suspect Linus wrote that in a Debian GNU/Linux | complicated way only to be able to [EMAIL PROTECTED] | have that comment in there. http://people.debian.org/~branden/ | -- Lars Wirzenius pgpCvnh8blham.pgp Description: PGP signature
Why does Debian's GCC still have GFDL components in main? (was Re: Decision GFDL)
Steve Langasek wrote (in http://lists.debian.org/debian-legal/2003/debian-legal-200309/msg7.html): Does this mean that the gcc maintainers don't agree with this list's interpretation of the GFDL, or that they don't regard this as a high priority between now and the release? I believe that the maintainers want to have a document they can point to when ignorant users say WHY ISN'T THE GCC MANUAL IN MAIN? WHAT'S WRONG WITH YOU? Perhaps they're also waiting for the consensus on manuals with no Invariant Sections and no Cover Texts, which took longer to come to consensus. (The majority opinion now is that the GFDL is not a free license even then, due to the overly broad technical measures clause among other things.) The GCC mantainers can correct me if I'm wrong or verify that I'm right which is why I've cross-posted this to debian-gcc. Could some Debian developer who is a debian-legal regular perhaps write such a document and put in on some Debian website? Somewhere on people.debian.org would quite likely satisfy the desire for something 'official'. You can lift mine if you like. :-) Perhaps even a nice summarizing post on debian-legal would do. Does the patch have negative side effects that leave the maintainers reluctant to apply it (such as leaving sarge without any gcc manual at all, even in non-free)? Quite likely. Probably this could be fixed with little effort by uploading a non-free-gcc source package though...
Re: Some licensing questions regarding celestia
On Mon, Sep 01, 2003 at 08:24:57PM +0200, Mika Fischer wrote: How far does one have to go in regard to data? A few examples. - Data published on the web: http://www.obspm.fr/encycl/cat1.html lists stars with possible planets around them. Is one allowed to use this data in a program? Basically for me this is just information and it doesn't make sense to restrict that. In the U.S., mere facts are not subject to copyright protection, and there are no separate laws extending copyright-like protection to databases of facts. In many European jurisdictions, copyright-like protections to extend to databases of facts. One possible ass-covering maneuver might be to perform some sort of transform on the format of the data before distributing it. For all I know, this is necessary for the data to be useful to celestia in the first place. - Data announced on the web: http://www.eso.org/outreach/press-rel/pr-2001/pr-07-01.html which leads to: http://obswww.unige.ch/~udry/planet/new_planet.html which has the data. What's the question here? - If one creates a 3D Model from published data, what is the legal status of this work? Can it be GPLed? Can it be put in the public domain? In the U.S., if your 3D model of published factual information is not original in its expressiveness, it doesn't warrant copyright protection. The application of an algorithmic, transformative process to factual data that results in a 3D model historically has no real grounds for copyright protection in the U.S., but I imagine we'll see that principle come more and more under attack as Hollywood studios use CGI more and more. On the other hand, if you used a 3D model so generated as inspiration for an oil painting you produces, your oil painting would, in all likelihood, be under your copyright in the U.S. (Unless it was a work for hire, in which case it belongs to your employer, etc.) Regarding the public domain question: * If the facts upon which the model was based were in the public domain, AND the transformation that produced the 3D model introduced no original expressiveness through which (in the U.S.) copyright can attach, then the 3D model not just *can* be put in the public domain, it automatically *is* in the public domain. * To my knowledge, in the U.S, a statement from all the copyright holders of a work is sufficient to place it in the public domain, if they want to do so before it would otherwise pass into the public domain through expiration of copyright (now somewhere between 1 and 2 million years after the death of the last suriving author if the copyright is held by individuals, and 100 billion years in the case of corporations[1]). [1] This parenthetical is a facetious exaggeration. I think. -- G. Branden Robinson| No math genius, eh? Then perhaps Debian GNU/Linux | you could explain to me where you [EMAIL PROTECTED] | got these... PENROSE TILES! http://people.debian.org/~branden/ | -- Stephen R. Notley pgpzqosOhad9V.pgp Description: PGP signature
Re: A possible GFDL compromise
On Sun, Aug 31, 2003 at 10:22:42PM -0400, Richard Stallman wrote: None of us have ever considered saying that the 4-clause BSD license is non-free, or suggesting that programs under such licenses should be removed from Debian main. I've said it at least once. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | pgprQ7rgvvc1R.pgp Description: PGP signature
Re: The GPL and you
On Mon, Sep 01, 2003 at 09:58:00AM -0700, Daniel Isacc Walker wrote: On Sun, 31 Aug 2003, Anthony DeRobertis wrote: On Sunday, Aug 31, 2003, at 18:54 US/Eastern, Daniel Isacc Walker wrote: I use talkfilters data structures and API in my code. Is that a derivate work? Everything else I've read indicates that it isn't.. We had a rather recent discussion about this, I believe in the context of perl's dbd mysql stuff. In short, I happen to agree with you; others on this list may agree or disagree, but Debian's policy seems to be paranoia: Assume that is a derivative work, in part because the FSF says so. That's interesting. In the Wine community the majority opinion is that the Win32 API and data structures are not copyrightable. My opinion is derived mainly from my work with Wine. Since the Win32 API is so vast (100's of functions and structures) I would think if header files are copyrightable then the Win32 headers must be copyrighted by Microsoft. As far as I know, Microsoft hasn't attacked Wine for using the API . The common interpretation runs like this: An API is not copyrightable. However, you can't have a hypothetical license either. There must be _at least one_ implementation of the API which will permit your program to function usefully, and which has an appliable license, in order for you to use it. So, you can't create a proprietary program that links to a GPLed library and distribute the result - this is clearly a derivative work of the GPLed library - but if there's a MIT/X11 implementation of the same library available, you can use that instead. Which one people decide to actually use on their systems is not your concern. Needless to say, you cannot play silly buggers with this and argue your way around the GPL. It's either a derivative work (because it needs *that* library) or it isn't (because this other one will do instead). -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | pgpwJezFzqBeZ.pgp Description: PGP signature
Re: UnrealIRCd License (Click-Through issue)
On Mon, Sep 01, 2003 at 04:01:59PM +0100, Edmund GRIMLEY EVANS wrote: Of course. You have the right to sue anyone for anything at any time! Only in the US. Most countries require the plaintiff actually have a case before letting them in the courtroom. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | pgpD4RuHKVM2J.pgp Description: PGP signature
Re: Some licensing questions regarding celestia
On Mon, Sep 01, 2003 at 04:16:36PM -0500, Branden Robinson wrote: On Mon, Sep 01, 2003 at 08:24:57PM +0200, Mika Fischer wrote: How far does one have to go in regard to data? A few examples. - Data published on the web: http://www.obspm.fr/encycl/cat1.html lists stars with possible planets around them. Is one allowed to use this data in a program? Basically for me this is just information and it doesn't make sense to restrict that. In the U.S., mere facts are not subject to copyright protection, and there are no separate laws extending copyright-like protection to databases of facts. In many European jurisdictions, copyright-like protections to extend to databases of facts. I don't believe that database law applies here, due to the small size of the data set. However, I'm only familiar with the UK implementation; other parts of Europe may differ. It's the French version in particular which matters here anybody? -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | pgptKg73lTJCI.pgp Description: PGP signature
Re: UnrealIRCd License (Click-Through issue)
On Mon, 01 Sep 2003, Andrew Suffield wrote: On Mon, Sep 01, 2003 at 04:01:59PM +0100, Edmund GRIMLEY EVANS wrote: Of course. You have the right to sue anyone for anything at any time! Only in the US. Most countries require the plaintiff actually have a case before letting them in the courtroom. This isn't really true in the US as well. Frivolous lawsuits in the US often are thrown out very rapidly in the pre trial phase.[1] Many of the egregious cases which are thrown out often involve the awarding of fees and/or damages to the defendant. Now, that doesn't mean that you still wouldn't have to retain an attorney, but assuming you've got a reasonably competent one, they should be able to make the case go away relatively rapidly, often without bankrupting you. Don Armstrong 1: Of course, you do hear about rather rediculous judgements from time to time. That's because there are quite a few moronic lower court judges out there. Most of those settlements (the Mc-D's coffee one for instance) are often overturned or reduced in the appeals process. -- What I can't stand is the feeling that my brain is leaving me for someone more interesting. http://www.donarmstrong.com http://www.anylevel.com http://rzlab.ucr.edu pgpuoiKruYuy0.pgp Description: PGP signature
Re: Why does Debian's GCC still have GFDL components in main? (was Re: Decision GFDL)
[ Disclaimer: I supposedly have CVS access, last I was told, and I] [ certainly do most of the work to ensure that GCC will work on the ] [ proto-port to NetBSD; apart from that, and reading both debian-gcc ] [ and debian-legal, you probably have to ask Matthias Klose for a final ] [ answer, since he initially replied to the GFDL bug. ] On Mon, Sep 01, 2003 at 04:54:21PM -0400, Nathanael Nerode wrote: Steve Langasek wrote (in http://lists.debian.org/debian-legal/2003/debian-legal-200309/msg7.html): Does this mean that the gcc maintainers don't agree with this list's interpretation of the GFDL, or that they don't regard this as a high priority between now and the release? I believe that the maintainers want to have a document they can point to when ignorant users say WHY ISN'T THE GCC MANUAL IN MAIN? WHAT'S WRONG WITH YOU? This seems to be the primary issue. Removing the manuals isn't all that hard (okay, maybe not trivial, but not *hard*), but it will result in the sound of a million developers crying out, and then suddenly being silenced... er, okay, maybe not. But *I* certainly would like to have a canonical answer to give folks (like the proposed GFDL FAQ) when they come beating on the lists, railing against fate and Debian that the manuals have vanished to the abyss of non-free. Perhaps they're also waiting for the consensus on manuals with no Invariant Sections and no Cover Texts, which took longer to come to consensus. (The majority opinion now is that the GFDL is not a free license even then, due to the overly broad technical measures clause among other things.) One which I share, though I believe the actual statement by Matthias was (paraphrased) Call me when y'all make up your minds. I'd say the survey more or less accomplished demonstrating this, for the moment. I'm not sure if that suffices for Matthias, or whether he wants a statement from the (RM/DPL/ftpmaster/High Pooba) sanctioning it. The GCC mantainers can correct me if I'm wrong or verify that I'm right which is why I've cross-posted this to debian-gcc. See the disclaimer at the beginning of this message. :) Could some Debian developer who is a debian-legal regular perhaps write such a document and put in on some Debian website? Somewhere on people.debian.org would quite likely satisfy the desire for something 'official'. You can lift mine if you like. :-) Perhaps even a nice summarizing post on debian-legal would do. Does the patch have negative side effects that leave the maintainers reluctant to apply it (such as leaving sarge without any gcc manual at all, even in non-free)? Quite likely. Probably this could be fixed with little effort by uploading a non-free-gcc source package though... Certainly it would require splitting things out and juggling a bunch of things to get things to get it all sorted out. Not impossible, but I don't blame Matthias for wanting a clear ruling on it before going to that much effort. A patch won't suffice, since the *sources* for main must be Free, and thus, the entire manual set must be moved to a separate source package, to be maintained in Debian. -- Joel Baker [EMAIL PROTECTED],''`. Debian GNU NetBSD/i386 porter: :' : `. `' `- pgpsSb3UMpH64.pgp Description: PGP signature
Re: Some licensing questions regarding celestia
On Mon, Sep 01, 2003 at 11:45:05PM +0100, Andrew Suffield wrote: On Mon, Sep 01, 2003 at 04:16:36PM -0500, Branden Robinson wrote: On Mon, Sep 01, 2003 at 08:24:57PM +0200, Mika Fischer wrote: How far does one have to go in regard to data? A few examples. - Data published on the web: http://www.obspm.fr/encycl/cat1.html lists stars with possible planets around them. Is one allowed to use this data in a program? Basically for me this is just information and it doesn't make sense to restrict that. In the U.S., mere facts are not subject to copyright protection, and there are no separate laws extending copyright-like protection to databases of facts. In many European jurisdictions, copyright-like protections to extend to databases of facts. I don't believe that database law applies here, due to the small size of the data set. It is not really the size of the data set that matters; the amount of work required to create the dataset is. -- Wouter Verhelst Debian GNU/Linux -- http://www.debian.org Nederlandstalige Linux-documentatie -- http://nl.linux.org Stop breathing down my neck. My breathing is merely a simulation. So is my neck, stop it anyway! -- Voyager's EMH versus the Prometheus' EMH, stardate 51462.
Re: Decision GFDL
Steve Langasek [EMAIL PROTECTED] wrote: On Sun, Aug 31, 2003 at 01:47:01AM -0400, Walter Landry wrote: That is my big question, which no one seems to want to answer. Is it ok for the Release Manager to ignore the Social Contract? These documents are not going to become free in the forseeable future. Is it ok for the maintainers of the packages that contain GFDL documentation to ignore the Social Contract? Of course not, but they keep getting conflicting signals. J. Random debian-legal-eagle says it is a bug, but the esteemed Release Manager tells them it might not be. So the problem is the statement that the Release Manager made. Once again, is it ok for the Release Manager to ignore the Social Contract? I understand that the Release Manager has some flexibility with policy, but I thought that he didn't have that flexibility with the Social Contract. Please let me know if I'm wrong. They have also agreed to uphold it; it shouldn't require a mandate from the release manager to get these bugs fixed. (Even treating the bug as RC does not guarantee the Social Contract has been upheld, as it only guarantees the bug will not exist in the release -- possibly by removing the package from testing and leaving it, bugs and all, in unstable.) Unstable packages often have RC bugs. That is why it is called unstable. Regards, Walter Landry [EMAIL PROTECTED]
Re: The GPL and you
On Sun, 31 Aug 2003, Rick Moen wrote: Quoting Daniel Isacc Walker ([EMAIL PROTECTED]): I use talkfilters data structures and API in my code. Is that a derivate work? Everything else I've read indicates that it isn't.. Copyright doesn't cover ideas. But I believe you're asking the wrong question. The reason Debian doesn't _redistribute_ GPLed PHP extensions is that they're believed to be in licence conflict (when linked with the PHP interpreter). But as long as you're NOT seeking to create something that everyone will feel comfortable redistributing, you're fine. License conflict? Like it's not illegal but it's a taboo? Daniel Walker
