Re: Taking a position on anti-patent licenses
On Mon, Jan 31, 2005 at 01:46:47PM -0500, Glenn Maynard wrote: > > 11.1 Term and Termination. The term of this License is perpetual > > unless terminated as provided below. This License and the rights granted > > hereunder will terminate: > > > (d) upon written notice from Licensor if You, at any time during the > > term of this License, commence an action for patent infringement against > > any third party alleging that the Covered Code itself (excluding > > combinations > > with other software or hardware) infringes any patent (including by cross- > > claim or counter claim in a lawsuit). > > I'm undecided about these clauses. One argument against them seems to > be "don't mix patents and copyrights", but I havn't seen much of a case > for that I don't think that's an argument, but rather a warning. Mixing them usually results in trouble. > (I do think Nathanael is misstating the case a bit: despite the long > debates on this topic, I still don't have a very strong opinion myself, > and I don't think there are yet any consenses about patent defense clauses > at all.) My impression has always been of a not-sure-probably-not-free attitude, combined with "but ones like this are clearly not a problem, so if that's good enough for you then please do that instead". -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -><- | signature.asc Description: Digital signature
Re: GPL as a license for documentation: What about derived works?
On Mon, Jan 31, 2005 at 12:09:18PM +0100, Frank K?ster wrote: > But still there's a lot of cruft in it that might be just confusing for > an author who considers GPL for his text, or even add confusion to a > possible lawsuit. Licenses *are* confusing. Not our fault, nor can we do anything about it; there's a limit to the effectiveness of the anti-lawyer spray we use. Deal with it. Any scenario which appears to contain a license which isn't confusing is one which contains a license you don't understand properly, and should be treated with the utmost suspicion, just like a perpetual motion machine. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -><- | signature.asc Description: Digital signature
Re: Firefox/Thunderbird trademarks: a proposal
On Tue, Feb 01, 2005 at 01:21:32PM +, Gervase Markham wrote: > I must admit I'm finding this a bit frustrating. I came to debian-legal, > listened to what people (including, I believe, the Thunderbird package > maintainer) were saying, and drew up a document[0] which I hoped would > meet Debian's requirements, further modifying it based on feedback[1]. > This modified version has been approved of by at least one list > member[2]. However, I am now hearing a completely different viewpoint > from Eric about what sort of things are acceptable and considered > DFSG-free. > > This is not a criticism of Eric - as Firefox package maintainer, his > opinion is clearly important. But is this sort of thing merely something > one has to accept when dealing with Debian, or is there anyone in > authority who can actually give me a consistent story here? Who > eventually decides what sort of licence is acceptable? There's roughly three tests involved. Firstly, we have to determine whether a license permits *us* to do all the stuff we need in order to package it. Then we have to determine whether it's free. Lastly, the maintainer has to determine whether they're willing to work with it. The first two are dealt with here, on -legal. There's no authority per se; this is a discussion forum. But when we (eventually) reach agreement, it's usually accepted by the rest of the project simply because -legal is by definition the set of people who care about this stuff, although sometimes we have to fight off saboteurs who do their best to derail the process. The last one is up to the maintainer. However, if -legal decides something is free but the maintainer doesn't find the license acceptable, the most likely result is a new maintainer (in this case, it would probably take the form of the iceweasel/firefox split that was discussed earlier). > What if the > Firefox and Thunderbird maintainers have totally opposing viewpoints? Generally? We have a big fight. You can pretty much ignore that part, we can fight amongst ourselves effectively enough. > What if we come up with something, and later project-wide discussion on > the general issue of trademarks decides that it's in fact non-free? Well, it's not impossible, but the likelihood of something passing debian-legal and not the rest of the project is small. That's why -legal is the point of first contact. > Eric Dorland wrote: > >Interesting. What about the case of Fedora? They've applied even more > >patches than Debian has to their package (at least it looked that way > >from what I saw). They certainly don't fall under the current > >trademark license. Have you approached them with an agreement? > > There's only one of me, and this isn't my full-time job. Debian > approached us to make sure that they are doing the right thing, and this > is what we are working out here. Or would you rather I did everyone else > first, and left the Debian package in legal limbo? Fedora has the advantage of being able to decide this stuff in a week or two. Debian usually takes rather longer. > >They are certainly practically very difficult, but they need not be > >that exhaustively precise. I certainly believe the Mozilla Foundation > >is acting in good faith. If the Mozilla Foundation puts the general > >things down it wants in the Trademark License and they apply equally > >to everyone, I don't see any reason we need to get too bogged down in > >details and semantics. > > Let's take just one example. The Mozilla Foundation is very keen that > nothing ships as "Firefox" which contains spyware. How would you define > "spyware" in a watertight way for the trademark license document? > Remember, you have to get it perfectly right first time, otherwise the > person exploiting the loophole you left would just say "well, I'm taking > my permissions under version 1.0 of the agreement, not 1.1". As a general rule, it is impossible to pin this stuff down in a legally binding fashion and remain free. We've seen this sort of thing a lot. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -><- | signature.asc Description: Digital signature
Re: handling Mozilla with kid gloves [was: GUADEC report]
On Tue, Feb 01, 2005 at 09:41:33PM -0500, Glenn Maynard wrote: > On Wed, Feb 02, 2005 at 02:24:42AM +, MJ Ray wrote: > > I've found when making my licence notes that there are licences > > with grey areas, licences which could be used for either free > > or non-free software without too much effort. > > I know that any license can be "interpreted" in a non-free way (even > the MIT license), but that's usually the rare exception. Other than > licenses with "options" (which essentially makes them multiple licenses), > and other than questionable "interpretations", when has this actually > happened? The Artistic license would be the classical case. Pine if you want an example of where we got screwed by it. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -><- | signature.asc Description: Digital signature
Re: handling Mozilla with kid gloves [was: GUADEC report]
On Wed, Feb 02, 2005 at 02:24:42AM +, MJ Ray wrote: > I've found when making my licence notes that there are licences > with grey areas, licences which could be used for either free > or non-free software without too much effort. I know that any license can be "interpreted" in a non-free way (even the MIT license), but that's usually the rare exception. Other than licenses with "options" (which essentially makes them multiple licenses), and other than questionable "interpretations", when has this actually happened? (Maybe there are some obvious cases, but I'm just having trouble thinking of any.) -- Glenn Maynard -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: handling Mozilla with kid gloves [was: GUADEC report]
Here's the interesting thing: are the summaries trying to be everything to everyone and that's why they don't work? Francesco Poli wrote: > When I find out some useful or interesting piece of software (i.e. > program or documentation or music or ...), I try to determine its > (DFSG-)freeness. [...] So far, so good. That's the interesting question for most of us. > Thus having a collection of past discussion summaries *is* useful, IMHO. I can see how a good collection of summarised discussions can help. > [...] I try to approach its copyright holders and > persuade them to change license. > In order to be more credible when I point out the issues that makes a > license non-free [...] Here's the flip. All of this talk of a licence being free or non-free doesn't directly answer the question of whether the software is free. Even convincing people isn't the core question. I've found when making my licence notes that there are licences with grey areas, licences which could be used for either free or non-free software without too much effort. > > Do the long licence summaries do much besides fuelling the project > > red-top's debian-legal hate campaign? > I think that the absence of summaries is even worse, because, I suppose, > debian-legal hate campaigners are often not very interested in legal > details: as a consequence, long (and difficult to follow) threads with > no summarized conclusions would seem even more obscure and opaque > to them. Isn't making it more opaque to the hate campaigners a good thing? It would reduce them to empty "I don't comprehend debian-legal" rumblings instead of being able to point at something which suggests to the innocent "look, these people are going to ban Mozilla and the entire Linux kernel from Debian". The problem is how to make it still useful to the rest. Gerv's recent grumble has given me some ideas. -- MJR/slef Maybe My Opinion Only -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Taking a position on anti-patent licenses
On Wed, Feb 02, 2005 at 12:52:58AM +, MJ Ray wrote: > These licences are not normally so considerate as to limit > themselves to swpat claims. Even the RPSL, which seems one of > the less offensive ones, says "any patent". I wouldn't mind so > much if I only lost patent permission that I didn't need anyway. Err. Sorry if I'm being dense, but what type of patent, other than a software patent, might one claim a piece of software infringes? (I'm assuming we're talking about "if you claim this software infringes a patent, you lose something" licenses, and not "if you claim the original author infringes a patent".) > In the latter case, I lose the copyright licence in my home > jurisdiction too. An offensive patenter could sue my US > branch about some patent and the branch uses some patent > (either obtained or a patent not used for software at home) > to help defend it... it's not hard to see possible ways for > interactions to go cross-border, thanks to the Berne Union. :-( I think this is a legitimate problem, even ignoring jurisdictional issues (which I don't really care about--you're the same company, as far as I'm concerned, regardless of which branch you're using to interact with me). That is, I think it *is* a legitimate use of software patents to defend against attack from other software patents. It's not pleasant--no use of swpats are--but it's one of the only effective ways for an entity to protect itself. A company can legitimately want to hold patents as a defense against patent attacks, and at the same time legitimately and honestly have no interest in attacking other companies with them. This is a very difficult situation: no company is going to give a truly Free license to those patents, since they would render them useless as a defense (the attacker would be getting a license, too). Now, a patent license that says "you have a license to our software patents unless you bring litigation against us, in which case you lose it" might deal with that: if you attack me, my defenses become available; if you don't, I can't use them offensively. (However, "if you attack *this software*"-style clauses don't do that, since if you attack my other projects with your patents, my patents on this work don't become available.) -- Glenn Maynard -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Taking a position on anti-patent licenses
Josh Triplett <[EMAIL PROTECTED]> wrote: > MJ Ray wrote: > > My main argument for not mixing them is that most of these terms > > seem extend software patents into places which don't have them > > yet, but do have software copyright. [...] > That seems like a reasonable argument. However, I don't see how these > clauses would cause a problem for people in > non-software-patent-afflicted areas. It seems like the only way these > clauses could affect you would be if you sued someone using the software > because they infringed your software patent. These licences are not normally so considerate as to limit themselves to swpat claims. Even the RPSL, which seems one of the less offensive ones, says "any patent". I wouldn't mind so much if I only lost patent permission that I didn't need anyway. There's also the simpler objection of having to check all these patent-copyright interactions for new problems. We've not done much work on that, mainly because it's so unclear still. Having them split is a lot easier for all. > If you are in a > non-swpat-afflicted area, then either 1) you have no such patents to sue > with, or 2) you would use another legal jurisdiction for the suit, in > which you do hold such patents. In the former case, the clause doesn't > affect you, and in the latter case, we aren't talking about the > swpat-free jurisdiction. In the latter case, I lose the copyright licence in my home jurisdiction too. An offensive patenter could sue my US branch about some patent and the branch uses some patent (either obtained or a patent not used for software at home) to help defend it... it's not hard to see possible ways for interactions to go cross-border, thanks to the Berne Union. :-( -- MJR/slef My Opinion Only: see http://people.debian.org/~mjr/ Subscribed to this list. No need to Cc, thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: GPL - "specifying" the preferred form for modification
On Fri, Jan 28, 2005 at 04:14:15PM -0500, Glenn Maynard wrote: > This has come up several times, so I'm CCing [EMAIL PROTECTED] to get their > take on this. FSF folks: please ignore the documentation aspect above; > I'm interested in the general problem of people "specifying" the preferred > form for modification, which I believe is tantamount to placing an additional > restriction beyond the GPL. Let me know if I'm way off base. licensing@ agreed that this would be an additional restriction. -- Glenn Maynard -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Firefox/Thunderbird trademarks: a proposal
* Gervase Markham ([EMAIL PROTECTED]) wrote: > I must admit I'm finding this a bit frustrating. I came to debian-legal, > listened to what people (including, I believe, the Thunderbird package > maintainer) were saying, and drew up a document[0] which I hoped would > meet Debian's requirements, further modifying it based on feedback[1]. > This modified version has been approved of by at least one list > member[2]. However, I am now hearing a completely different viewpoint > from Eric about what sort of things are acceptable and considered > DFSG-free. Unfortunately we're largely in uncharted waters here. I think in the strictest interpretation of the DFSG we should really just drop the trademark. But the DFSG wasn't really written with trademarks in mind, so there is certainly some flexibility here, and we don't want to be too heavy handed. > This is not a criticism of Eric - as Firefox package maintainer, his > opinion is clearly important. But is this sort of thing merely something > one has to accept when dealing with Debian, or is there anyone in > authority who can actually give me a consistent story here? Who > eventually decides what sort of licence is acceptable? What if the > Firefox and Thunderbird maintainers have totally opposing viewpoints? > What if we come up with something, and later project-wide discussion on > the general issue of trademarks decides that it's in fact non-free? Welcome to Debian :) debian-legal is really just a forum to discuss the legal issues and try to reach some consensus over whether some licenses are problematic or not. It doesn't have any authority in and of itself. I am probably the one who will have to be convinced that trademark policy is alright (along with Mike Hommey, the co-maintainer along with me, who has yet to really make his opinion known.). I can be vetoed by one of the ftp masters (since they can block me from uploading new version), and Project Leader could step in to mediate, but by and large the buck stops at me (and Mike). You could appeal to the rest of the Debian community, but so far everyone I've talked to has supported my position. > [0] http://lists.debian.org/debian-legal/2005/01/msg00503.html > [1] http://lists.debian.org/debian-legal/2005/01/msg00780.html > [2] http://lists.debian.org/debian-legal/2005/01/msg00795.html > > Eric Dorland wrote: > >Interesting. What about the case of Fedora? They've applied even more > >patches than Debian has to their package (at least it looked that way > >from what I saw). They certainly don't fall under the current > >trademark license. Have you approached them with an agreement? > > There's only one of me, and this isn't my full-time job. Debian > approached us to make sure that they are doing the right thing, and this > is what we are working out here. Or would you rather I did everyone else > first, and left the Debian package in legal limbo? Well wouldn't your job be easier if the rules were laid out in the policy document rather than having to make contact with individual distros and reading lengthy threads on debian-devel? I was not making a personal attack, I was just curious how others have handled the issues so we can benefit from any insight they had. But I'm curious what your opinion is of the fedora packages. Would they still qualify to use the trademark, even with their rather large amount of gtk integration patches? What if John Doe did the same thing with a version of Firefox on his website? If the answer to former is yes, and the latter is no, I would certainly find that unacceptable. > >They are certainly practically very difficult, but they need not be > >that exhaustively precise. I certainly believe the Mozilla Foundation > >is acting in good faith. If the Mozilla Foundation puts the general > >things down it wants in the Trademark License and they apply equally > >to everyone, I don't see any reason we need to get too bogged down in > >details and semantics. > > Let's take just one example. The Mozilla Foundation is very keen that > nothing ships as "Firefox" which contains spyware. How would you define > "spyware" in a watertight way for the trademark license document? > Remember, you have to get it perfectly right first time, otherwise the > person exploiting the loophole you left would just say "well, I'm taking > my permissions under version 1.0 of the agreement, not 1.1". Do you feel your current trademark license is watertight? It certainly makes pretty vague references to quality. And that's fine, I believe you're acting in good faith. You can police your trademarks based on your policy, and you can interpret it however you want. If we see you're applying your policy unfairly we can remove our support of it. Well if you screwed up the trademark license for version 1.0, then when 1.1 comes out you fix the license. Sure, it would still be broken for 1.0 but you couldn't use the 1.0 license with 1.1, so what's the problem? > >Well with
Re: a right to privacy is not in the DFSG, therfore you don't have one
On Mon, Jan 31, 2005 at 08:50:42PM +, Steve McIntyre wrote: > *yawn* That's a nice line in rhetoric you have there. The DFSG is the > standard that DDs have agreed should be the basis for deciding on the > Freeness of Software. If you want to extend it, you know what to So you're saying that the DFSG doesn't allow Debian to consider "capture a bear to modify or redistribute this work" non-free, since it's not explicitly said in the DFSG, and that anyone who wants to "keep that software out" is free to "extend the DFSG"? (If that's not what you're saying, please clarify your position.) It's pretty clear to me that the DFSG already disallows this, and needs no "extending" to handle it. > do. Hint: it starts with actually becoming a DD rather than sniping > from the sidelines. I think "you're not a DD, so shut up" is about the weakest argument possible, given the amount people stress that you don't need to be a DD to contribute. Please come back when you have some actual arguments. -- Glenn Maynard -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SableVM/Kaffe pissing contest
Brian Thomas Sniffen wrote: > Michael Poole <[EMAIL PROTECTED]> writes: >>(Incidentally, is not gjc in main? It seems a likely candidate to >>substitute for Kaffe if you wish for another GPL-free way to execute >>Eclipse.) > > I don't think gjc can handle Eclipse. If it can, why not Sable-VM or > some other non-GPL'd JVM? If you mean gcj and gij, then yes, it can; gcj can even be used to natively-compile Eclipse. Also, at least according to the SableVM changelog, SableVM can now run Eclipse as well. - Josh Triplett signature.asc Description: OpenPGP digital signature
Re: SableVM/Kaffe pissing contest
Walter Landry wrote: > Suppose I have a program Foo which uses either GNU readline. I can > compile Foo against GNU readline (but not link it), and distribute the > result. I can also distribute GNU readline separately. But I can not > distribute foo and GNU readline together. How is this different from > your case? Hold on a second. You seem to be arguing against the established interpretation of the GPL here: at least according to the FSF, you may not distribute the GPL-incompatible Foo compiled against GNU readline, linked or not. - Josh Triplett signature.asc Description: OpenPGP digital signature
Re: Taking a position on anti-patent licenses
MJ Ray wrote: > Glenn Maynard <[EMAIL PROTECTED]> wrote: >>I'm undecided about these clauses. One argument against them seems to >>be "don't mix patents and copyrights", but I havn't seen much of a case >>for that--it seems to say "don't try to protect against patents via >>copyright", but copyright is all we have available. [...] > > My main argument for not mixing them is that most of these terms > seem extend software patents into places which don't have them > yet, but do have software copyright. That spread is the opposite > of what some of us seek. I realise other places already have > software patents and need some way to reduce their effect, > but please stop tipping your rubbish into our back yard. If > the two licences for copyright and patents don't interact, > swpat-free residents can ignore one of them and I like giving > people less legal work. That seems like a reasonable argument. However, I don't see how these clauses would cause a problem for people in non-software-patent-afflicted areas. It seems like the only way these clauses could affect you would be if you sued someone using the software because they infringed your software patent. If you are in a non-swpat-afflicted area, then either 1) you have no such patents to sue with, or 2) you would use another legal jurisdiction for the suit, in which you do hold such patents. In the former case, the clause doesn't affect you, and in the latter case, we aren't talking about the swpat-free jurisdiction. - Josh Triplett signature.asc Description: OpenPGP digital signature
Re: Firefox/Thunderbird trademarks: a proposal
Gervase Markham <[EMAIL PROTECTED]> wrote: > This modified version has been approved of by at least one list > member[2]. I don't remember much about Michael K Edwards except he's currently MIA from the New Maintainer queue. http://nm.debian.org/nmstatus.php?email=medwards-debian%40sane.net Then again, I'm not very good at making whoami clear when it would help. I'll try to improve. debian-legal is advisory. The decision-makers are the ftpmasters (ultimately, but they're a bit busy and tend to listen to advice) and the package maintainers. Sorry if that wasn't clear. That's probably the biggest omission from the web site at the minute. I'll see if I can change it soon (needs me to relearn some things). Usually, the maintainers ask debian-legal, but there are some notable exceptions, IMO thanks to various hate campaigns. If two maintainers have totally different views, I guess ftpmasters get to play referee while debian-legal are the linesmen. > What if we come up with something, and later project-wide discussion on > the general issue of trademarks decides that it's in fact non-free? "Ooops."? Looking at it another way, use debian-legal as a vocal, interested focus group for your development. If you can't even convince this bunch of liberal law-abiders, then you really might have a problem. > [...] Debian > approached us to make sure that they are doing the right thing, and this > is what we are working out here. Or would you rather I did everyone else > first, and left the Debian package in legal limbo? This latest round started with your proposal. I'm sorry, but my brain is too addled to remember who approached who and I didn't find a backlink. I'm not sure it matters. I think it's rather unhelpful to suggest that licensing and being left in limbo are the only two options here. At least, there's "acting to avoid" that Eric could do. More to the point, Mozilla Foundation knows about Fedora's infringement and have not done anything. Is this common? Have the trademarks become generic? > Let's take just one example. The Mozilla Foundation is very keen that > nothing ships as "Firefox" which contains spyware. How would you define > "spyware" in a watertight way for the trademark license document? How does MF's trademark scheme stop that happening? No offence, but the Mozilla codebase is damn huge and I can't believe you'll check every line of every package. The debian project is very keen that nothing ships as "debian" which has security flaws. We don't go around stomping on CD distributors who are selling old releases, do we? (Honest question: I might just not have heard about it.) Maybe it's not possible for anyone other than MF to ship Firefox as free software while they have their current aims for the trademarks. It would be a defining case study in free software and trademarks. > > If I did have a trademark on "Eric Dorland", and I didn't like what > > someone was doing with it, I could ask them (legally) to stop. I don't > > need a Trademark License to enforce that. > You would rather Debian was in the situation where the Mozilla > Foundation could ask them to stop using the Firefox trademark > immediately and totally at any time? No, but that power isn't given by current law. . w , )tm. http://people.debian.org/~mjr/ Subscribed to this list. No need to Cc, thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: a right to privacy is not in the DFSG, therfore you don't have one
Glenn Maynard wrote: >On Mon, Jan 31, 2005 at 02:08:30PM -0500, Branden Robinson wrote: > >> Thanks for the props, however. I continue to believe that a DFSG analysis >> is the *beginning* of a process of understanding whether something is free >> software or not, not a substitute for the whole thing. Certain well-known >> people in the project have stridently insisted to me, however, that this >> opinion puts me into an extremely small minority. > >I don't believe you're in the minority at all--if you are, it's probably >time to scrap the DFSG entirely, since the Project must no longer care >about Free Software principles at all. *yawn* That's a nice line in rhetoric you have there. The DFSG is the standard that DDs have agreed should be the basis for deciding on the Freeness of Software. If you want to extend it, you know what to do. Hint: it starts with actually becoming a DD rather than sniping from the sidelines. -- Steve McIntyre, Cambridge, UK.[EMAIL PROTECTED] "...In the UNIX world, people tend to interpret `non-technical user' as meaning someone who's only ever written one device driver." -- Daniel Pead -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [Pkg-alsa-devel] RFS: alsa-tools
Walter Landry <[EMAIL PROTECTED]> wrote: > Unfortunately, that is not the case. All of the source for packages > in main must satisfy the DFSG. For example, if there are some > non-free, but distributable, files in the original tar ball, those > have to be taken out and a new "original" tar ball made. Read the context. The non-free code has been split out into a separate tarball. The issue is that some (but not all) of the code in alsa-tools depends on that non-free code. -- Matthew Garrett | [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Firefox/Thunderbird trademarks: a proposal
Gervase Markham <[EMAIL PROTECTED]> wrote:
> I must admit I'm finding this a bit frustrating. I came to debian-legal,
> listened to what people (including, I believe, the Thunderbird package
> maintainer) were saying, and drew up a document[0] which I hoped would
> meet Debian's requirements, further modifying it based on feedback[1].
> This modified version has been approved of by at least one list
> member[2]. However, I am now hearing a completely different viewpoint
> from Eric about what sort of things are acceptable and considered
> DFSG-free.
There is no way to compel a package maintainer to include code or
functionality under a license that they don't like. As a community of
volunteers, Debian can set a minimum standard ("Your work doesn't get in
unless it's at least this free"), but individual members can hold
different ones.
I believe that the policy you've come up with is likely to be acceptable
to Debian as a whole, but I'm afraid that you do need to convince the
package maintainer as well. There's a limited amount of social pressure
the project can put on someone in this sort of situation.
--
Matthew Garrett | [EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Firefox/Thunderbird trademarks: a proposal
On Tue, Feb 01, 2005 at 01:21:32PM +, Gervase Markham wrote: > This is not a criticism of Eric - as Firefox package maintainer, his > opinion is clearly important. But is this sort of thing merely something > one has to accept when dealing with Debian, or is there anyone in > authority who can actually give me a consistent story here? Well... to some degree it's the way Debian is. As far as authority goes, the package maintainer is the primary authority for issues regarding that package. There are limits and balances on that, but unless we have a really good reason we're not in a position to override the maintainer. -- Raul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Firefox/Thunderbird trademarks: a proposal
I must admit I'm finding this a bit frustrating. I came to debian-legal, listened to what people (including, I believe, the Thunderbird package maintainer) were saying, and drew up a document[0] which I hoped would meet Debian's requirements, further modifying it based on feedback[1]. This modified version has been approved of by at least one list member[2]. However, I am now hearing a completely different viewpoint from Eric about what sort of things are acceptable and considered DFSG-free. This is not a criticism of Eric - as Firefox package maintainer, his opinion is clearly important. But is this sort of thing merely something one has to accept when dealing with Debian, or is there anyone in authority who can actually give me a consistent story here? Who eventually decides what sort of licence is acceptable? What if the Firefox and Thunderbird maintainers have totally opposing viewpoints? What if we come up with something, and later project-wide discussion on the general issue of trademarks decides that it's in fact non-free? [0] http://lists.debian.org/debian-legal/2005/01/msg00503.html [1] http://lists.debian.org/debian-legal/2005/01/msg00780.html [2] http://lists.debian.org/debian-legal/2005/01/msg00795.html Eric Dorland wrote: Interesting. What about the case of Fedora? They've applied even more patches than Debian has to their package (at least it looked that way from what I saw). They certainly don't fall under the current trademark license. Have you approached them with an agreement? There's only one of me, and this isn't my full-time job. Debian approached us to make sure that they are doing the right thing, and this is what we are working out here. Or would you rather I did everyone else first, and left the Debian package in legal limbo? They are certainly practically very difficult, but they need not be that exhaustively precise. I certainly believe the Mozilla Foundation is acting in good faith. If the Mozilla Foundation puts the general things down it wants in the Trademark License and they apply equally to everyone, I don't see any reason we need to get too bogged down in details and semantics. Let's take just one example. The Mozilla Foundation is very keen that nothing ships as "Firefox" which contains spyware. How would you define "spyware" in a watertight way for the trademark license document? Remember, you have to get it perfectly right first time, otherwise the person exploiting the loophole you left would just say "well, I'm taking my permissions under version 1.0 of the agreement, not 1.1". Well with due respect the Community Edition clause is going to be completely useless to any distro. I mean you can't even backport a security fix under it. Indeed. The Community Edition stuff is not designed for Linux distributions. Perhaps you can see where I'm coming from if you think of the most complex piece of software you've written, pretend that when it starts up, it puts up a big "written by Eric Dorland" message, and try and write down all the things you would like people not to do to it and still have your name on the front. I understand what you're saying, and that's why you've sought protection for your mark. In general we don't concern ourselves with trademarks since they're generally easy to circumvent (ie, rename things) if the mark holder objects to our using the name. And we will do our best to make it as easy to circumvent as possible, should anyone wish to. However, you've formalized things with your Trademark Policy. And we will do our best to abide by your wishes in that document. As the document stands we can't use your marks. If you'd like us to, (and I would like that very much), then you need to put the provisions in the trademark license, where anyone can use them. But the trademark should only be used on quality products. And defining "quality" for something as complex as software is an extremely difficult task. Say we wrote a test suite. What happens if Debian's change to fix multi-user stuff broke the suite? Such a thing would easily become a restriction on how you could change the code. The currently suggested mechanism imposes no such restrictions - you can make whatever changes you like. If I did have a trademark on "Eric Dorland", and I didn't like what someone was doing with it, I could ask them (legally) to stop. I don't need a Trademark License to enforce that. You would rather Debian was in the situation where the Mozilla Foundation could ask them to stop using the Firefox trademark immediately and totally at any time? One of the things I thought we'd established earlier in this discussion (which you may not have seen) is that such uncertainty is not acceptable to Debian. Gerv -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Eclipse 3.0 Running ILLEGALY on Kaffe
Walter Landry wrote: Dalibor Topic <[EMAIL PROTECTED]> wrote: You have made a very convincing argument that "required to install" is too broad. My criteria is "required to run". I've showed that your interpretation of 'required to run' is too broad, as you attempt to stretch it in the same direction, by arguing from the 'but they are both installed together in main' angle. Eclipse does not require Kaffe to run, it runs very well on many other VMs. Eclipse in main would require Kaffe to be installed, but wouldn't necessarily require it run, as you still could run it perfectly fine with other VMs. When Debian puts Eclipse into main, Debian is distributing Eclipse to be used with Kaffe. When it is in contrib, Debian is distributing Eclipse to be used by something outside of main. Nope. The GPL does not allow you to say 'you must use this data with that program and that program alone'. That interpretation of the GPL would violate freedom 0. You have no right to limit how I run a program I get from you licensed under the GPL. If you do that, you lose the rights to distribute the program under the GPL at all, as the GPL does not allow you to add restrictions to it. That hold for Kaffe just as well as it holds for, say, your http://www.nongnu.org/arx/ project. You can't restrict me to use ArX only for GPLd projects, in general. ArX is actually pretty facinating. Looking at the ArX stable release, that I downloaded from your page at http://superbeast.ucsd.edu/~landry/ArX/ArX-1.0.20.tar.gz it is under the GPL, but the GPLd tarball actually incorporates some non-GPL-compatible boost libraries according to ArX-1.0.20/src/boost/libs/graph/LICENSE and your post at http://lists.gnu.org/archive/html/arx-users/2004-12/msg00019.html I'm not saying that ArX violates the GPL, as I have no idea what you do with the non-GPL compatible code, and I, contrary to some other people I've talked to, would be very wary of making such accusations to my peers without knowing precisely what I'm talking about, and in the case of ArX, I don't. What strikes me as odd is that under your whole reasoning of how GPL applies to Eclipse and Kaffe you have been trying to argue that debian should show a lot of scruple about putting a GPLd work and a separate GPL-incompatible work on the same medium because you believe that to create a non-distributable combined work, instead of mere aggregation, if the GPLd work is modified, but you seem to show no such scruple to incorporate verbatim a GPL-incomaptible work into your GPLd work, ArX, modify ArX and distribute the result from your web site. Given that ArX is RFP-ed according to http://lists.debian.org/debian-wnpp/2004/12/msg00306.html I'm wondering what the fine line is that you, as the author of ArX draw to explain why you think the collective works clause you seem to cling to applies to Kaffe + Eclipse, but doesn't do so to ArX. To me, it's quite obvious that since they are distributable, independant works, they can be distributed on the same medium, and that's what the GPL says, and the FSF does. [1] The FSF has an explicit exemption from all of the copyright holders (i.e. themselves). Where would that explicit exemption be? I can't find none, because none is necessary, afaict. Do you think the FSF has such a bizarre explicit exemption to the GPL of all the copyright holders on all the code packaged 'to run' [1] in ututo-e which the FSF redistributes from their servers? I prefer to trust the FSF's judgement on the GPL :) cheers, dalibor topic [1] https://e.ututo.org.ar/repository/ [2] ftp://ftp.gnu.org/gnu+linux-distros/ututo-e/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Eclipse 3.0 Running ILLEGALY on Kaffe
On Mon, Jan 31, 2005 at 10:18:56PM -0500, Walter Landry wrote: > You have made a very convincing argument that "required to install" is > too broad. My criteria is "required to run". If you're talking about the scope of copyright law, or the relevance of the license granted by the GPL, you're talking about "required to copy". If you're talking about the DFSG, you're not talking about a legal issue, but a set of guidelines, and the scope is Debian, and adoption into its archives. You might have some personal "require to run" criteria, for some context. But please don't mistake that for a copyright issue. -- Raul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Eclipse 3.0 Running ILLEGALY on Kaffe
Walter Landry wrote: You are correct. It is no longer the case when the work is unmodified. However, Debian does modify Kaffe. Even if all of those modifications were incorporated upstream, Debian still must be able to make security fixes. A security fix would kick Eclipse out of main, which really means that Eclipse should not have been in main from the beginning. The GPL allows you to make modifications to the GPLd work and redistribute them under the GPL. As the GPL of Kaffe does not reach out to Eclipse in the unmodified case, it can't suppendly do so by adding a whitespace to kaffe. cheers, dalibor topic -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Taking a position on anti-patent licenses
Glenn Maynard <[EMAIL PROTECTED]> wrote: > I'm undecided about these clauses. One argument against them seems to > be "don't mix patents and copyrights", but I havn't seen much of a case > for that--it seems to say "don't try to protect against patents via > copyright", but copyright is all we have available. [...] My main argument for not mixing them is that most of these terms seem extend software patents into places which don't have them yet, but do have software copyright. That spread is the opposite of what some of us seek. I realise other places already have software patents and need some way to reduce their effect, but please stop tipping your rubbish into our back yard. If the two licences for copyright and patents don't interact, swpat-free residents can ignore one of them and I like giving people less legal work. -- MJR/slef My Opinion Only: see http://people.debian.org/~mjr/ Subscribed to this list. No need to Cc, thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
