Re: Open Software License v2.1
Josh Triplett [EMAIL PROTECTED] writes: MJ Ray wrote: On 2004-09-13 03:39:39 +0100 Glenn Maynard [EMAIL PROTECTED] wrote: This License shall terminate automatically and You may no longer exercise any of the rights granted to You by this License as of the date You commence an action, including a cross-claim or counterclaim, against Licensor or any licensee alleging that the Original Work infringes a patent. The new clause may still not satisfy everyone, but it's much better: it no longer forbids all patent action against Licensor; in fact, it no longer makes a special case of the Licensor at all. It is a great step forwards: it no longer contaminates other software. Sadly, as written, it still seems to terminate a copyright licence as a consequence of patent-based action, even in self-defence. Until I'm shown harder facts about copyright misuse and trademark misuse in each law this licence is used in, I'm still uncomfortable with this idea. Does that really matter, if the condition for termination is acceptable? If the patent license is terminated, the only reason to care whether the copyright license terminates as well is if you intend to ignore the lack of a patent license. (Granted, Debian tends to do that in many cases. :) ) Furthermore, if you *sue claiming that the work infringes your patent*, I see absolutely no reason why you should have any rights to the work, since you are trying to eliminate the rights of others to the work. I can understand the objection to terminating the license over unrelated lawsuits, but not the objection to termination when you actually sue over the software in question. The alternative would be that *no one except you* would have rights to the software, which means you have now essentially made it your own proprietary software. No, you did that when you invented it and filed for a patent. It's *already* your own proprietary software, and you're going to the courts to get that enforced. Consider a copyright-only case: Alice and Bob each release some software under a copyleft, with a clause mentioning that any lawsuit claiming copyright infringement on the work or any derivative forfeits all right to the original work and any derivative. Alice and Bob each use each other's software extensively, though they don't actually like each other much at all. Now Charlie comes along and derives a new work from Alice's and Bob's software. He violates the copyleft. They'd each like to sue him for copyright infringement, but if either one sues to defend his property rights, he loses his rights to the other's software. Is this free? I don't think it is, though I'm not so sure of myself to think that reasonable people can't disagree. But the patent case seems very similar: A and B each develop some software and distribute it under a copyleft with an attached patent license with a termination clause for any suit against the licensor or any licensee claiming patent infringement in the associated code. They each use the other's software and patented techniques. C then distributes an illicit derived work, in such a way that he does not violate the copyright license but does violate the patent license. Neither A nor B can sue him without losing their rights to the other's software. So if the copyright version is non-free, this is non-free, right? -Brian -- Brian Sniffen [EMAIL PROTECTED]
Re: Open Software License v2.1
On Wed, Sep 15, 2004 at 12:01:15AM -0400, Brian Thomas Sniffen wrote: Consider a copyright-only case: Alice and Bob each release some software under a copyleft, with a clause mentioning that any lawsuit claiming copyright infringement on the work or any derivative forfeits all right to the original work and any derivative. Alice and Bob each use each other's software extensively, though they don't actually like each other much at all. Now Charlie comes along and derives a new work from Alice's and Bob's software. He violates the copyleft. They'd each like to sue him for copyright infringement, but if either one sues to defend his property rights, he loses his rights to the other's software. Is this free? This isn't claiming that the works of Alice or Bob are infringing copyright; it's claiming that Charlie is infringing copyright. Neither Alice nor Bob face license termination for each other's work for suing Charlie over Charlie's use of those works; they'd only lose the license to Charlie's derivative work. Or at least they shouldn't, if this type of license is implemented properly. Take a simpler case. Alice writes a program. Bill contributes somewhat to it--enough to have a copyright claim. John takes the result, and violates the license. Bill sues John for violating his part of the copyright. Does Bill lose his license to Alice's work? No; he's not saying that Alice's work is in violation, he's saying that John is in violation (through his act of distributing without eg. offering source). I'm a little mixed up, though: I'd expect that an action, not a work, violates a copyright or patent license--by copying in a way that isn't allowed, etc.--whereas the work itself infringes on a patent if there's no license at all. I'm not sure, though, and I'm getting confused just thinking about it ... I don't think it is, though I'm not so sure of myself to think that reasonable people can't disagree. But the patent case seems very similar: A and B each develop some software and distribute it under a copyleft with an attached patent license with a termination clause for any suit against the licensor or any licensee claiming patent infringement in the associated code. They each use the other's software and patented techniques. C then distributes an illicit derived work, in such a way that he does not violate the copyright license but does violate the patent license. Neither A nor B can sue him without losing their rights to the other's software. So if the copyright version is non-free, this is non-free, right? I don't agree with the copyright example, so this becomes mostly irrelevant, but anyway: I like generality, but I don't think I'd generalize here. I consider software copyright enforcement to be legitimate, and potentially beneficial. I do not consider any kind of software patent enforcement to be beneficial, neither to free software nor anything else. If C is violating patents and not copyrights, then I really don't care if it's made harder for A and B to sue him over it. I simply don't consider right to use software patents to be legitimate or worth protecting. Software patents have no redeeming value. -- Glenn Maynard
Re: Debian and Mozilla Trademarks
Glenn Maynard wrote: On Thu, Sep 09, 2004 at 06:18:08AM -0700, Josh Triplett wrote: Andrew Suffield wrote: On Wed, Sep 08, 2004 at 10:30:55PM -0700, Paul C. Bryan wrote: Are there any other examples of restrictions placed on open-source licenses that Debian has had to deal with in the past? When a package is resricted by trademark usage, does Debian have a policy to effectively deal with it? There's no difference between trademark law and anything else, and nothing special about copyright law. We are interested in what you can do with the work, for *whatever* legal reasons. It is significant that the DFSG does not talk about copyrights. Agreed entirely; unfortunately, this logic does not seem to be applied in the case of the discussions of the license for the Debian logo. See the Free Debian logos? thread. The Official Use logo is not, as far as I understand, intended to be DFSG-free at all, and is not intended to be used in main; I don't think this is considered a bug. The Open Use logo should be, but isn't; this is a bug, but I think it's an acknowledged one. I understand that. The thread in question was primarily discussing the Open Use logo license. I don't think many people are seriously advocating that the DFSG only applies to restrictions made under copyright law. In that thread, several people suggested that a restriction such as You may not use this logo, or any confusingly similar logo, to refer to anything else in a way which might cause confusion with Debian. was Free, and furthermore that it was as free as a trademark license could be. - Josh Triplett signature.asc Description: OpenPGP digital signature
Re: Debian and Mozilla Trademarks
On Tue, Sep 14, 2004 at 11:33:59PM -0700, Josh Triplett wrote: I don't think many people are seriously advocating that the DFSG only applies to restrictions made under copyright law. In that thread, several people suggested that a restriction such as You may not use this logo, or any confusingly similar logo, to refer to anything else in a way which might cause confusion with Debian. was Free, and furthermore that it was as free as a trademark license could be. With a straight face? -- Glenn Maynard
Re: Debian Hardened project (question about use of the Debian trademark)
On Tue, Sep 14, 2004 at 04:40:53PM +0100, Martin Michlmayr wrote: * Lorenzo Hernandez Garcia-Hierro [EMAIL PROTECTED] [2004-09-08 16:26]: I want to know if i can use the trademark Debian on the name of a project that i've started , Debian Hardened which i want to see as an official Debian sup-project. I personally feel that this name has the same problems that Trusted Debian has - it suggests that normal Debian is not secure. In any case, I think you should post your question to debian-project rather than -legal since -project is more appropriate and might get more feedback. Notice that unlike the Trusted Debian case, Lorenzo seem to be willing for it to be an officially recognized sub project, like the custom debians are. Friendly, Sven Luther
Re: most liberal license
[ Please keep me on cc as I'm not subscribed ] Hi! Thanks, for your response: Is there some other as free as public domain license? I don't like to reinvent the wheel, but I haven't found one yet.\ I ususally recommend and use the MIT-Licence for that, it essentially says the same stuff as yours, is the shortest of all on opensource.org, and is well known and widely used. Yes, I know the MIT-License and it is the option if there are any objections against my draft. However there are some things I dislike about the MIT-License: * You are forced to include the original copyright notice, in whatever substantial portions of the Software are. * Even worse, you are required to include the permission notice, thus it is half way towards copyleft. (I.e. it doesn't affect other software, but still you can't sell it in a proprietary way.) * It is an enumerate style license, which means that - you might forget something - it is water on the mills of those who write wired legal text saying you might do everything, but afterwards try to define what everything is. - it is based upon US copyright law and the rights enumerated therein, but there might exist other juristdictions with additional/other rights. Ideally I would put my software in the public domain, but I've been told, that this isn't possible in all jurisdictions (I don't even know about my own), so I thought to circumwent this by licensing it to give the same rights *as* public domain. Harald -- http://www.unet.univie.ac.at/~a0300802/
Re: Open Software License v2.1
On 2004-09-15 04:14:40 +0100 Josh Triplett [EMAIL PROTECTED] wrote: Does that really matter, if the condition for termination is acceptable? If the patent license is terminated, the only reason to care whether the copyright license terminates as well is if you intend to ignore the lack of a patent license. (Granted, Debian tends to do that in many cases. :) ) I think so. There are at least two interesting cases which terminating copyright licence on patent action hurts free software users. The first is the case where you were licensed no patents to use the software. Hopefully this will be the most common case, as free software developers reject software patents. If only the patent licence terminates, including the software in a counter-claim defending some patent accusation does not remove your freedom to use the software. Why should a licensor be allowed to use copyright to their advantage when attacking others with patents? Of course, if your counter-claim is successful, I suspect you must grant a RF patent licence for the software to remain free. The second is when your use of the software only occurs in a jurisdiction which does not have software patents. A similar argument applies. Furthermore, if you *sue claiming that the work infringes your patent*, Of course, but I see no reason to unnecessarily harm free software developers who wish to use software patents to defend software patent accusations. It's not a tactic I like, but it seems valid. Further, some have claimed that copyright-based enforcement of patents may be a type of misuse so these all terminates licence terms are useless anyway, but I'm not sure about that. I consider myself fortunate that my work is outside the reach of software patents... so far. -- MJR/slefMy Opinion Only and not of any group I know Creative copyleft computing - http://www.ttllp.co.uk/ http://www.thewalks.co.uk stand 13,Lynn Carnival,12 Sep
Re: Open Software License v2.1
On Wed, Sep 15, 2004 at 09:06:18AM +0100, MJ Ray wrote: The first is the case where you were licensed no patents to use the software. Hopefully this will be the most common case, as free software developers reject software patents. If only the patent licence terminates, including the software in a counter-claim defending some patent accusation does not remove your freedom to use the software. Why should a licensor be allowed to use copyright to their advantage when attacking others with patents? Of course, if your counter-claim is successful, I suspect you must grant a RF patent licence for the software to remain free. So the claim here, then, is that an aggressor could obtain the copyright to some software that employs these clauses, make offensive patent suits on entirely unrelated grounds, and use the software they obtained as a wall against patents related to that software being used defensively in reaction to the offensive patent suits. (This much doesn't seem too convincing.) The offensive patent suits must be unrelated, since related patents owned by the licensor are granted by Grant of Patent License. On the other hand, that probably wouldn't apply if the copyright changes hands; if Apache is under this license, and someone who owns lots of web-related patents buys Apache ... Of course, but I see no reason to unnecessarily harm free software developers who wish to use software patents to defend software patent accusations. It's not a tactic I like, but it seems valid. Further, some have claimed that copyright-based enforcement of patents may be a type of misuse so these all terminates licence terms are useless anyway, but I'm not sure about that. I don't think this is very interesting, because--to my understanding-- patent enforcement is extremely expensive, well beyond the reach of the vast majority of free software authors, while patent-defense clauses aren't. I'm not sure how effective these clauses would be, though, unless they become very widely used. -- Glenn Maynard
Re: most liberal license
[I think we may be saying the same thing here, but I thought some clarification was necessary.] On Wed, 15 Sep 2004, Glenn Maynard wrote: You can never take someone else's work, place restrictions on it and sell it. You can if the license allows it. if a work is in the public domain, nobody can. Again, you can do the same with a public domain work. That being said, nothing is stoping people from returning the work to its previous status (by removing whatever you did, if anything) and redistributing that under the original terms. [If that's what you meant, I'm sorry for being obtuse. ;-)] Don Armstrong -- Three little words. (In decending order of importance.) I love you -- hugh macleod http://www.gapingvoid.com/graphics/batch35.php http://www.donarmstrong.com http://rzlab.ucr.edu
Re: Real names in a football game
O Martes, 14 de Setembro de 2004 ás 22:18:46 +0200, Isaac Clerencia escribía: I think this can be illegal (also team names?). Yes, it falls under trade mark protection laws. Since team names and logos, and players' names are big assets for their teams and national leagues (put Beckam's name in a 5-euro t-shirt and now it's worth 50 euros), they're defended very aggresively. Some football (soccer) games have been released with players with names like José García and Roberto da Silva because they couldn't get the rights to the actual names. I'd remove the names (i.e. change them to other, innocuous names) even without asking as I know the answer beforehand. I already have a version without player names ready to be uploaded, removing team names should take a little more effort. Use city names. Or common prefix + city name + common suffix (Sporting Club de A Coruña, Atlético de Valencia, Madrid S.A.D., Berlin 89, etc.), but this would possibly re-create actual teams' names. Or turn the teams into national selections. Country names (or any geographical names) aren't protected by trademark laws. -- Jacobo Tarrío | http://jacobo.tarrio.org/
Re: Real names in a football game
On Wednesday 15 September 2004 11:56, Jacobo Tarrio wrote: O Martes, 14 de Setembro de 2004 ás 22:18:46 +0200, Isaac Clerencia escribía: I'd remove the names (i.e. change them to other, innocuous names) even without asking as I know the answer beforehand. I've uploaded a new package without the player names. I already have a version without player names ready to be uploaded, removing team names should take a little more effort. Use city names. Or common prefix + city name + common suffix (Sporting Club de A Coruña, Atlético de Valencia, Madrid S.A.D., Berlin 89, etc.), but this would possibly re-create actual teams' names. I started doing this, but I got tired as there are lots of team names. I'll do it ASAP. Best regards pgpQd93bqg2XX.pgp Description: PGP signature
Re: Open Software License v2.1
On Wed, Sep 15, 2004 at 05:42:45PM +0200, Bernhard R. Link wrote: If a software discriminates against people wanting to sell or even only those selling free software as binary-only, it is also non-free. If by discriminates against you mean prevents distribution by, then you are correct. If by discriminates against you mean something else, you may or may not be correct, depending on the specifics [if any] of what you mean. -- Raul
Re: Open Software License v2.1
Glenn Maynard [EMAIL PROTECTED] writes: On Wed, Sep 15, 2004 at 12:01:15AM -0400, Brian Thomas Sniffen wrote: Consider a copyright-only case: Alice and Bob each release some software under a copyleft, with a clause mentioning that any lawsuit claiming copyright infringement on the work or any derivative forfeits all right to the original work and any derivative. Alice and Bob each use each other's software extensively, though they don't actually like each other much at all. Now Charlie comes along and derives a new work from Alice's and Bob's software. He violates the copyleft. They'd each like to sue him for copyright infringement, but if either one sues to defend his property rights, he loses his rights to the other's software. Is this free? This isn't claiming that the works of Alice or Bob are infringing copyright; it's claiming that Charlie is infringing copyright. Neither Alice nor Bob face license termination for each other's work for suing Charlie over Charlie's use of those works; they'd only lose the license to Charlie's derivative work. Or at least they shouldn't, if this type of license is implemented properly. But that's where patents differ from copyright -- they have no concept of derivative works, only of protected methods. So if you sue claiming that the implementation in Charlie's is bad, you're also claiming the implementation in Alice's is bad. This is suing the Licensor or any licensee over that implementation. The suit is *motivated* by failure to comply with the license, but it's over patent infringement in Alice or Bob's code. Take a simpler case. Alice writes a program. Bill contributes somewhat to it--enough to have a copyright claim. John takes the result, and violates the license. Bill sues John for violating his part of the copyright. Does Bill lose his license to Alice's work? No; he's not saying that Alice's work is in violation, he's saying that John is in violation (through his act of distributing without eg. offering source). But you can't sue for license violation, not of a free license -- all you can do is sue for patent infringement. So he does, in the patent case, have to claim infringement of his patent on that method. I'm a little mixed up, though: I'd expect that an action, not a work, violates a copyright or patent license--by copying in a way that isn't allowed, etc.--whereas the work itself infringes on a patent if there's no license at all. I'm not sure, though, and I'm getting confused just thinking about it ... I think you're right about that. But I have to think about it a lot more. -Brian -- Brian Sniffen [EMAIL PROTECTED]
Re: most liberal license
* Even worse, you are required to include the permission notice, thus it is half way towards copyleft. (I.e. it doesn't affect other software, but still you can't sell it in a proprietary way.) You can take MIT-licensed software and sell it to people without providing source, and you don't have to place your modifications under the same license; you can place them under a heavily restrictive EULA. If that's not selling in a proprietary way, could you please explain what you mean by that? It says you have to include the permission notice in any substantial portions of the Software no matter if source or binary only. I think this make merging the Software into some proprietary product quite difficult. But even worse is the issue with your statement below: (You can never take someone else's work, place restrictions on it and sell it. Nobody but the copyright holder has the ability to do that; if a work is in the public domain, nobody can. You can only place restrictions on your modifications, which the MIT license allows you to do.) Perhaps you can't claim copyright of a copy of something you are not the copyright holder, because simply copying is no intellectual work at all. But proprietary software is often not only restricted by copyright but by an EULA which actually is a contract. By such a contract you can restrict copying of something you actually are not the copyright holder. But I don't see how you could do this while still including the permission notice. The MIT license is in no way a copyleft. Half way is no way, isn't it? ;) Harald
Re: Open Software License v2.1
Brian Thomas Sniffen writes: Glenn Maynard [EMAIL PROTECTED] writes: This isn't claiming that the works of Alice or Bob are infringing copyright; it's claiming that Charlie is infringing copyright. Neither Alice nor Bob face license termination for each other's work for suing Charlie over Charlie's use of those works; they'd only lose the license to Charlie's derivative work. Or at least they shouldn't, if this type of license is implemented properly. But that's where patents differ from copyright -- they have no concept of derivative works, only of protected methods. So if you sue claiming that the implementation in Charlie's is bad, you're also claiming the implementation in Alice's is bad. This is suing the Licensor or any licensee over that implementation. The suit is *motivated* by failure to comply with the license, but it's over patent infringement in Alice or Bob's code. You're mixing your examples in a way unclear to me, especially by adding a vague you to the three-party example you started with. A) If you are Alice and sue Charlie for patent infringement, and he has complied with your open patent license, he can use that license as a defense. B) If you are Charlie and sue Alice for patent infringement in her program, you would lose the rights Alice granted to you in that program. This is the primary case that we are interested in, although it no longer resembles the case where Alice or Bob sues Charlie for copyright infringement. C) If you are some third party, any licenses between Alice and Charlie are irrelevant to your claim that one or both infringe your patent. Take a simpler case. Alice writes a program. Bill contributes somewhat to it--enough to have a copyright claim. John takes the result, and violates the license. Bill sues John for violating his part of the copyright. Does Bill lose his license to Alice's work? No; he's not saying that Alice's work is in violation, he's saying that John is in violation (through his act of distributing without eg. offering source). But you can't sue for license violation, not of a free license -- all you can do is sue for patent infringement. So he does, in the patent case, have to claim infringement of his patent on that method. Perhaps you should inform IBM that they cannot sue SCO for GPL violations, as they are currently doing -- or clarify what you mean by you can't sue for license violation, not of a free license. See also the Netfilter team's recent copyright lawsuit in Germany. Michael Poole
Re: Open Software License v2.1
On Wed, 15 Sep 2004, Matthew Garrett wrote: An elementary point of Free Software is to protect the rights of the users, not excluding bad ones. (Or will GPL3 have a section termination the licence if you breach any FSF copyright?) forfeits the right to distribute the code at all, which implies that the GPL doesn't protect the rights of people who have violated it in the past. I know that something like this is in the GPL, but I wonder how he reconciles it with this: # Each time you redistribute the Program (or any work based on the # Program), the recipient automatically receives a license from the # original licensor to copy, distribute or modify the Program subject to # these terms and conditions. That suggests that although violating the GPL causes the license on the copy you have to be revoked, when someone distributes another copy to you, you gain another license for the new copy.
Re: GPL Compatibility of IFRIT License
Last month, Mark Hymers wrote: I'm currently working on packaging IFRIT (a piece of data visualization software which uses VTK and QT). I've been auditing all of the source files to check their licenses before I finish off the packaging and apart from a couple of small issues (small files not having licenses) which I'm going to contact upstream about, it all looks reasonably good. I just want to check one thing. The license for the majority of the IFRIT code is this: [modified BSD license text ellided] I saw the announcement of this package's inclusion in Debian Weekly News and took a look at upstream's site since I have an on-again, off-again interest in 3D visualization software. From what I see, I wonder if you'll have trouble with this due to upstream's apparent confusion about licensing? After all, you quote license text that seems to be straightforward modified BSD, yet in the link for downloading the software: http://casa.colorado.edu/~gnedin/IFRIT/vtkdownload.html upstream claims to be releasing it under the GPL or the QPL, but with a blatantly GPL-incompatible proviso against generating any income with the software: IFRIT is distributed under the terms of GNU Public License (GPL) or Qt Public License (QPL). You are given unrestricted permission to use, copy, distribute, and modify IFRIT. However, you do not have permission to sell IFRIT or to use IFRIT in any way that results in you obtaining income from such a use. By downloading IFRIT you are agreeing to abide by the terms of QPL and GPL. -- D. Joe Anderson http://www.etrumeus.com/~deejoe
Re: Open Software License v2.1
On Wed, Sep 15, 2004 at 01:12:06PM -0400, Brian Thomas Sniffen wrote: But that's where patents differ from copyright -- they have no concept of derivative works, only of protected methods. So if you sue claiming that the implementation in Charlie's is bad, you're also claiming the implementation in Alice's is bad. This is suing the Licensor or any licensee over that implementation. The suit is *motivated* by failure to comply with the license, but it's over patent infringement in Alice or Bob's code. In the patent case, this seems something like reasonable. If A creates a work (that, unknown to him, violates a patent); B creates a minor derivative work (that also violates the same patent), and C sues B for violating it, he's attempting to prevent both A and B from being distributed, and it seems close to reasonable that he lose his license to both A and B. If this wasn't the case, these clauses would be very weak. Suppose that B was in much wider use than A, but A contains most of the work (for example, B is a Debian package with a couple security fixes). C would be free to sue the users of B, without losing his license to A. It does lead to more serious situations. For example, X writes a program and reuses several distinct blocks of code that have been made publically available under this license (say, MD5.c and RSA.c). A finds out that he has a patent he can claim applies to MD5, and sues X. Which licenses does he lose? Obviously, he loses the license to X's program. Losing his license to MD5.c in all programs that use it would make sense; he's trying to stifle all use of it in every free program. However, if he loses his license to MD5.c, he'd presumably also globally lose his license to RSA.c--and that's not reasonable at all. Take a simpler case. Alice writes a program. Bill contributes somewhat to it--enough to have a copyright claim. John takes the result, and violates the license. Bill sues John for violating his part of the copyright. Does Bill lose his license to Alice's work? No; he's not saying that Alice's work is in violation, he's saying that John is in violation (through his act of distributing without eg. offering source). But you can't sue for license violation, not of a free license -- all you can do is sue for patent infringement. So he does, in the patent case, have to claim infringement of his patent on that method. Huh? Of course you can sue for license violation of a free license; for example, for refusing to comply with the source requirements of the GPL. (The example was in the context of copyrights and copyright-reciprocity only.) -- Glenn Maynard
Re: GPL Compatibility of IFRIT License
On Wed, 15, Sep, 2004 at 03:23:39PM -0500, [EMAIL PROTECTED] spoke thus.. upstream claims to be releasing it under the GPL or the QPL, but with a blatantly GPL-incompatible proviso against generating any income with the software: IFRIT is distributed under the terms of GNU Public License (GPL) or Qt Public License (QPL). You are given unrestricted permission to use, copy, distribute, and modify IFRIT. However, you do not have permission to sell IFRIT or to use IFRIT in any way that results in you obtaining income from such a use. By downloading IFRIT you are agreeing to abide by the terms of QPL and GPL. I'll take this up with upstream immediately and report back to debian-legal. Thanks for spotting this. Mark -- Mark Hymers, University of Newcastle Medical School Intercalating Medical Student (MBBS / PhD)
Re: GPL Compatibility of IFRIT License
On Wed, 15, Sep, 2004 at 09:54:24PM +0100, Mark Hymers spoke thus.. I'll take this up with upstream immediately and report back to debian-legal. Thanks for spotting this. Upstream has (very promptly, I must say) removed the clause from the website. Hope that solves all the problems with the IFRIT licenses. Mark -- Mark Hymers, University of Newcastle Medical School Intercalating Medical Student (MBBS / PhD) signature.asc Description: Digital signature
Re: Open Software License v2.1
On Wed, Sep 15, 2004 at 09:00:39AM -0400, Michael Poole wrote: Andrew Suffield writes: Long-standing conclusions, summarised: Terminating licenses (copyright, patent, trademark, dog-humping, or whatever else might interfere with distribution/modification/use) for any reason other than non-compliance is a bit of legal insanity to get contract-like provisions into a license. These provisions have to be considered like any other restriction (invert the sense of the conditional to get the restriction). Anything that requires a contract-like construct, rather than a simple license, is probably non-free. DFSG-free licenses give things to the licensee, not to the copyright holder. They are not a trade (although the grant of permissions does not have to be the most generous possible), even if their social behaviour resembles one. (Corollary of these two: terminating a license for any reason other than non-compliance is probably non-free) Other corollary: Claiming something is a contract-like provision is a useful wedge to make something like the GPL a non-free license. That's a summary of an old discussion which apparently you didn't read. Redefining it arbitrary to something else will obviously generate an arbitrary result. A restriction saying You may not sue me for patent issues is non-free. If any licenses said that, it might be relevant. Congratulations, you missed the point. Patent licenses are ignored unless there are actively enforced patents. In almost every case where we come across these patent clauses, there are no actively enforced patents, so we simply ignore them - but sometimes people write clauses like this one, which remain non-free in the absence of patents. If the patent licenses are ignored unless actively enforced, do you have a problem with the Apache License 2.0, in which only the patent license terminates in the event of patent litigation? Not in the absence of interesting patents. That was what we said way back when the proposed apache license 2.0 had the very same bug in it. This sort of clause is acceptable because when you don't have patent issues, it's a no-op - and that's by far the most common case. If a work were to appear for which somebody was running around litigating patents, it'd probably be non-free even with this patent license. We usually try to nail the clauses down to the point where this is unlikely to happen. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: Open Software License v2.1
Andrew Suffield writes: On Wed, Sep 15, 2004 at 09:00:39AM -0400, Michael Poole wrote: Andrew Suffield writes: Long-standing conclusions, summarised: Terminating licenses (copyright, patent, trademark, dog-humping, or whatever else might interfere with distribution/modification/use) for any reason other than non-compliance is a bit of legal insanity to get contract-like provisions into a license. These provisions have to be considered like any other restriction (invert the sense of the conditional to get the restriction). Anything that requires a contract-like construct, rather than a simple license, is probably non-free. DFSG-free licenses give things to the licensee, not to the copyright holder. They are not a trade (although the grant of permissions does not have to be the most generous possible), even if their social behaviour resembles one. (Corollary of these two: terminating a license for any reason other than non-compliance is probably non-free) Other corollary: Claiming something is a contract-like provision is a useful wedge to make something like the GPL a non-free license. That's a summary of an old discussion which apparently you didn't read. Redefining it arbitrary to something else will obviously generate an arbitrary result. Then please stop arbitrarily defining contract-like provision to what is convenient for you. Others will have their own arbitrary definitions that are useful to them. I think you are extending the conclusion of that discussion beyond the point where it is supportable; there are fairly clear differences between You must do X to get these rights and You lose these rights if you do Y, especially when Y prevents others from exercising those same rights. By way of example, you have no right to distribute a GPLed work if you attempt to charge users for patent licenses related to the work. A restriction saying You may not sue me for patent issues is non-free. If any licenses said that, it might be relevant. Congratulations, you missed the point. I rather think you were the one who misses the point, but how is it productive to make an unsupported insult like that? Michael Poole
Re: Patent clauses in licenses
On 2004-09-14 23:38:26 +0100 Michael Poole [EMAIL PROTECTED] wrote: MJ Ray writes: The OSI lists no licences as free. While pedantically true, I claim this is irrelevant on the basis of the similarity between the Open Source Definition and the DFSG. The only significant difference is that different groups interpret them. I think the significant difference is that OSD is a checklist, while DFSG are a set of guidelines. Further, DFSG are interpreted in public, while OSD appears to be checked mostly in private with the case in favour of approval made by a hired lawyer. Whose patent would make a difference to the FSF? I believe they regard the software as not free when it is subject to any encumberment harmful to free software. I've written before about the problems getting [EMAIL PROTECTED] to answer. In January, I was told [EMAIL PROTECTED] is seeking clarification from their licensing committee, but I've not heard back yet. Loss of patent license means the user cannot use the software. Loss of copyright license (at least in the USA) only removes the license of a user to modify or copy the software further. [...] Maybe it's not narrower for the USA then. Your law is not my law. I've elaborated on this in the OSL thread this morning. -- MJR/slefMy Opinion Only and not of any group I know Creative copyleft computing - http://www.ttllp.co.uk/ http://www.thewalks.co.uk stand 13,Lynn Carnival,12 Sep
Re: most liberal license
Harald Geyer wrote: there are some things I dislike about the MIT-License: * It is an enumerate style license, which means that - you might forget something - it is water on the mills of those who write wired legal text saying you might do everything, but afterwards try to define what everything is. - it is based upon US copyright law and the rights enumerated therein, but there might exist other juristdictions with additional/other rights. Are you sure? I thought the text to deal in the Software without restriction, including *without limitation* the rights to... (my emphasis) meant that it explicitly granted the rights to do anything with the software, and that the terms following it (use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so,) were just examples of what could be done with the right to deal in the Software without restriction? -- Lewis Jardine IANAL IANADD
Re: Debian Hardened project (question about use of the Debian trademark)
Hi, El mié, 15-09-2004 a las 09:35, Sven Luther escribió: On Tue, Sep 14, 2004 at 04:40:53PM +0100, Martin Michlmayr wrote: * Lorenzo Hernandez Garcia-Hierro [EMAIL PROTECTED] [2004-09-08 16:26]: I want to know if i can use the trademark Debian on the name of a project that i've started , Debian Hardened which i want to see as an official Debian sup-project. I personally feel that this name has the same problems that Trusted Debian has - it suggests that normal Debian is not secure. In any case, I think you should post your question to debian-project rather than -legal since -project is more appropriate and might get more feedback. Notice that unlike the Trusted Debian case, Lorenzo seem to be willing for it to be an officially recognized sub project, like the custom debians are. yeah, That's it! Adamantix (old Trusted Debian) is a different distro (and also i have good relationship with many of its developers, Peter for example). Debian Hardened is like Debian Junior, and the rest of subprojects. *We* must provide the best (and the easiest) way to harden Debian for advanced users, sysadmins or just people that want a really *more* secure environment than the common one, that does not need to be insecure but it will be more unsafe if you compare it with the same system but hardened.As a good example...you can forget to update your Bind9 named daemon 'cos somebody announced a new BOF in its code, but if you a hardened binary (+SSP/ProPOlice and a library to trace the BOF conditions) in a hardened environment (hardened kernel and RBAC/RSBAC policies) it will be not dangerous as having a simple Debian! That's the difference. We can start asking ourselves about Why not making Debian hardened directly?, we need to respect the freedom of choice and also, a normal user wouldn't want to use RBAC...or not? If somebody has read the TRNG tasks, on SF.net, it's related with some enhancements in the LEP (Linux Entropy Pool) using a TRNG device, that would make 99.9% (just for be paranoid, at the momment no body has demonstrated that the atomic decay is not unpredictable) unpredictable random numbers to be used within the LEP (/dev/random) making an user-space daemon and modifying random.c to take care of the TRNG (some of this work is not completed, but i've started doing some dirty hacks to random.c). Thanks in advance, Cheers. -- Lorenzo Hernandez Garcia-Hierro [EMAIL PROTECTED] PS: If i want to colaborate to the Debian project that's because i think i must give something back to the community that developed this fantastic distro! signature.asc Description: Esta parte del mensaje está firmada digitalmente
Re: Open Software License v2.1
* Michael Poole [EMAIL PROTECTED] [040915 15:00]: On the other hand, I always thought free software was about protecting users, not patent litigants who are supposed to already have working forms of the patented invention. cynism What about people selling non-free software, builders of nuclear bombs, terrorists, criminals, US-goverment agents, abortion medics, religious fundamentals, communists, lawyers, ... /cynism SCNR, Bernhard R. Link
Re: Open Software License v2.1
* Michael Poole [EMAIL PROTECTED] [040915 17:12]: (Your choice whether that describes you or an item you omitted. If you want to make a serious attempt to apply DFSG #5 or #6 to patent termination, I'm listening, but please craft your argument so that it does not classify the GPL as non-free because it discriminates against people who want to sell binary-only versions of free software.) If a software discriminates against people wanting to sell or even only those selling free software as binary-only, it is also non-free. (As this is quite equivalent to discrimation against people selling propietary code, as BSD licensed is also free) GPL does not do so, it only states things like you have to give (or offer) source with the binary of this program under GPL. An elementary point of Free Software is to protect the rights of the users, not excluding bad ones. (Or will GPL3 have a section termination the licence if you breach any FSF copyright?) Hochachtungsvoll, Bernhard R. Link
Re: Open Software License v2.1
Bernhard R. Link writes: * Michael Poole [EMAIL PROTECTED] [040915 17:12]: (Your choice whether that describes you or an item you omitted. If you want to make a serious attempt to apply DFSG #5 or #6 to patent termination, I'm listening, but please craft your argument so that it does not classify the GPL as non-free because it discriminates against people who want to sell binary-only versions of free software.) If a software discriminates against people wanting to sell or even only those selling free software as binary-only, it is also non-free. (As this is quite equivalent to discrimation against people selling propietary code, as BSD licensed is also free) GPL does not do so, it only states things like you have to give (or offer) source with the binary of this program under GPL. That's very nice, but it has approximately as much to do with patent termination clauses as do unicorns. An elementary point of Free Software is to protect the rights of the users, not excluding bad ones. (Or will GPL3 have a section termination the licence if you breach any FSF copyright?) One of the recurrent suggestions I have seen for GPL3 is some parallel to the Open Software License's External Deployment clause -- perhaps on the basis that end users of the software are users too, rather than the person who operates the machine executing the software being the user. GPL3 may not include such a clause, or may not ever happen, so I do not think it is productive to argue about what imaginary license clauses mean. Michael Poole
Re: Open Software License v2.1
Bernhard R. Link [EMAIL PROTECTED] wrote: An elementary point of Free Software is to protect the rights of the users, not excluding bad ones. (Or will GPL3 have a section termination the licence if you breach any FSF copyright?) RMS is quoted as saying Misusing a GPL-covered program permanently forfeits the right to distribute the code at all, which implies that the GPL doesn't protect the rights of people who have violated it in the past. -- Matthew Garrett | [EMAIL PROTECTED]
Re: most liberal license
Harald Geyer [EMAIL PROTECTED] writes: * Even worse, you are required to include the permission notice, thus it is half way towards copyleft. (I.e. it doesn't affect other software, but still you can't sell it in a proprietary way.) You can take MIT-licensed software and sell it to people without providing source, and you don't have to place your modifications under the same license; you can place them under a heavily restrictive EULA. If that's not selling in a proprietary way, could you please explain what you mean by that? It says you have to include the permission notice in any substantial portions of the Software no matter if source or binary only. I think this make merging the Software into some proprietary product quite difficult. Software under that license is distributed as part of Windows XP. Clearly it's not insurmountable. That permission notice applies only to the code covered by the licensor's copyright, not to other works merged with it. But even worse is the issue with your statement below: (You can never take someone else's work, place restrictions on it and sell it. Nobody but the copyright holder has the ability to do that; if a work is in the public domain, nobody can. You can only place restrictions on your modifications, which the MIT license allows you to do.) Perhaps you can't claim copyright of a copy of something you are not the copyright holder, because simply copying is no intellectual work at all. But proprietary software is often not only restricted by copyright but by an EULA which actually is a contract. By such a contract you can restrict copying of something you actually are not the copyright holder. But I don't see how you could do this while still including the permission notice. Sure you can. The permission notice is included. But I'll pet this cat here for you if you agree not to exercise it. The MIT license is in no way a copyleft. Half way is no way, isn't it? ;) No. It really is just a public license. You're objecting to the parts which make it such -- the fact that the license to the MIT licensed code is extended to anyone who receives it. But that's not a copyleft, just your inability to mess with the license granted by MIT. -Brian -- Brian Sniffen [EMAIL PROTECTED]
Re: most liberal license
On Wed, Sep 15, 2004 at 09:32:27AM +0200, Harald Geyer wrote: [ Please keep me on cc as I'm not subscribed ] Please set your Mail-Followup-To mail header. * Even worse, you are required to include the permission notice, thus it is half way towards copyleft. (I.e. it doesn't affect other software, but still you can't sell it in a proprietary way.) You can take MIT-licensed software and sell it to people without providing source, and you don't have to place your modifications under the same license; you can place them under a heavily restrictive EULA. If that's not selling in a proprietary way, could you please explain what you mean by that? (You can never take someone else's work, place restrictions on it and sell it. Nobody but the copyright holder has the ability to do that; if a work is in the public domain, nobody can. You can only place restrictions on your modifications, which the MIT license allows you to do.) The MIT license is in no way a copyleft. -- Glenn Maynard
Re: Open Software License v2.1
On Tue, Sep 14, 2004 at 06:59:29PM -0400, Glenn Maynard wrote: On Tue, Sep 14, 2004 at 10:53:55PM +0100, Andrew Suffield wrote: This whole consensus nonsense is just an excuse to discard any argument without responding to it. Note how it is only ever advanced by people who want to discard valid arguments; it is never used by people who want to introduce one. Freeness questions are not black and white. Rational people can disagree on whether a given restriction is free, due to differences in personal judgement and values, and they can agree on whether it's free but arrive at the conclusion for different reasons. When there are multiple rational answers, consensus helps choose among them. If there was no room for rational disagreement, this list would have a much easier job. That doesn't mean no consensus is an excuse to discard them all. If it means anything, it means that thing has to stay the hell away from the archive. Furthermore, it's entirely irrelevant when there's only one rational answer. Further, the case of terminating copyright licenses based on patent action specifically against the work just hasn't been discussed properly. I dispute that. We've seen these before and there was little to discuss. Long-standing conclusions, summarised: Terminating licenses (copyright, patent, trademark, dog-humping, or whatever else might interfere with distribution/modification/use) for any reason other than non-compliance is a bit of legal insanity to get contract-like provisions into a license. These provisions have to be considered like any other restriction (invert the sense of the conditional to get the restriction). Anything that requires a contract-like construct, rather than a simple license, is probably non-free. DFSG-free licenses give things to the licensee, not to the copyright holder. They are not a trade (although the grant of permissions does not have to be the most generous possible), even if their social behaviour resembles one. (Corollary of these two: terminating a license for any reason other than non-compliance is probably non-free) A restriction saying You may not sue me for patent issues is non-free. Patent licenses are ignored unless there are actively enforced patents. In almost every case where we come across these patent clauses, there are no actively enforced patents, so we simply ignore them - but sometimes people write clauses like this one, which remain non-free in the absence of patents. I do not believe we have ever encountered a scenario where there were actively enforced patents and we were offered something resembling a free software license for the patents. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: Open Software License v2.1
On Tue, Sep 14, 2004 at 08:14:40PM -0700, Josh Triplett wrote: Furthermore, if you *sue claiming that the work infringes your patent*, I see absolutely no reason why you should have any rights to the work, since you are trying to eliminate the rights of others to the work. I can understand the objection to terminating the license over unrelated lawsuits, but not the objection to termination when you actually sue over the software in question. The alternative would be that *no one except you* would have rights to the software, which means you have now essentially made it your own proprietary software. That's a false dilemma; the alternative which will normally occur is that the software gets the offending feature removed. Patent lawsuits do not have to be maximally destructive, and your argument only holds if they are. This sort of clause says that you can ignore anybody else's patents and implement anything you wish, and ignore the patents because invoking them will wreck everybody. I can't see how this could possibly be free, even if it might be desireable; for it to be free, it would have to still be free after s/patent/copyright/g (it would still be desireable). I can see a company like SCO adopting a scorched-earth strategy to eliminate some project that their sponsors don't like. So it's failed on both counts: non-free *and* still vulnerable to patent assaults. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: Open Software License v2.1
On Tue, Sep 14, 2004 at 08:20:02PM -0700, Josh Triplett wrote: I'm not sure that this clause necessarily passes the DFSG, but it's clear that the OSI has made a good and, in my opinion, successful effort to clean it up. It's neither fair nor correct to say that nothing has changed. It's still non-free for the same reasons, so nothing relevant has changed. You mean that you still believe it's non-free because nothing relevant to your reasons has changed, and you're pretending that other perspectives don't exist. Irrelevant by the law of limiting factors. But I haven't seen anybody seriously advance any other positions. Alright, now you have: I think that terminating a license because of unrelated lawsuits is unacceptable, but that terminating a license because you sue claim the work infringes one of your patents is perfectly acceptable. By suing, you are trying to take away other people's rights over the software, so you should lose yours as well. That's not a different position, it's an irrelevant compatible one. It might as well say I like bees. All of these are *discarded* by the law of limiting factors. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: Open Software License v2.1
Brian Thomas Sniffen writes: No, you did that when you invented it and filed for a patent. It's *already* your own proprietary software, and you're going to the courts to get that enforced. Consider a copyright-only case: Alice and Bob each release some software under a copyleft, with a clause mentioning that any lawsuit claiming copyright infringement on the work or any derivative forfeits all right to the original work and any derivative. Alice and Bob each use each other's software extensively, though they don't actually like each other much at all. Which license actually works like this? The Academic Free License 2.1, Apache License 2.0, IBM Public License 1.0, and others only address the issue of the program itself is the subject of the lawsuit, and exclude combinations with other works. I think we all agree that If you sue the Original Author for any patent violation, you lose rights granted by this license (e.g. RPSL 1.0) is not free, but that's different from either of the above. Michael Poole
