Hi, On Thu, Aug 23, 2007 at 08:07:48AM +0200, Gerfried Fuchs wrote: > But that link might point to v3, v4, vwhatever - and if we point only > there (like it is done now) we are not pointing to the license text we > were distributing it back then. Can you please make up your mind? I'm saying we should list all currently available options, meaning both GPL2 and GPL3. I never said we should point only to GPL3. (I did say we could, not that we should.) I'm asking what is your problem with that. > > As I wrote, I agree we should _also_ point to version 2. But there's no > > reason to _not_ point to version 3 (which at the moment is what the link > > should be). > > I am just saying that there is no _need_ to point to version 3. And > there is a reason: We would need to update the copyright file every > time a new subsequent version of the GPL is released to list all the > choices the user has. Or you are losing your own reasoning of listing > all the choices. Behold, the last change was 15 years ago. If this trend continues (and it likely will), we will need to reupload a large part of the archive once every 15 years! That's unacceptable, especially because most of the programs get an upload for a new version within an acceptable time anyway. Just to be sure it's clear: I really don't see any problem in this. > > After all, the copyright holder did offer that license to the users > > (implicitly by allowing the FSF to make that choice for them). > > The copyright holder doesn't have to ship all subsequent versions of > the license, so why do you think we would need to do so? Because we want to. Because our ftp master will reject our packages if we don't. We don't have a legal requirement to do this, just providing a link would be enough. But we want to provide the complete license, as a service to our users. > Especially like you said, "long after it was written" - i.e. written > before v3 release (like, uhm, a tiny bit of software in the pool?) and > read nowadays? I think the problem that at this moment not all licenses are listed for GPL v2 or later programs is something which should be solved, but it isn't urgent. We're not saying wrong things, we're not omitting legally required information, we're just not providing the level of service to users that we would want to provide. > > Now that it will (hopefully) soon be pointing to GPL3, it does make > > sense to add an extra link to GPL2. But there's no reason at all for > > omitting the link to the latest version. > > See above, with every subsequent release of the GPL you are losing one > version if you want to list a complete set. For a short time. As I said before, I'm suggesting to add a link to version 2 as well as keep the link to the latest version. > And there is no need to do so. But there is a need to point to the v2 > because that's the version we are distributing it _now_. Not at all. If I, as a user, now download a GPL2-or-later-program, it is offered to me under two licenses, GPL2 and GPL3. Which means we are already distributing it using GPL3 if our users desire so. > > I get the feeling you have something against GPL3, but of course > > that's just my impression. > > I have a bit more than an impression that you are hating the v2 when > reading that lines. I never said I have anything against the v3, No, you didn't, but you gave me the feeling. That's what I wrote, too. ;-) The reason I'm writing my impressions about this is so you can correct them. Because if they're wrong, it would make the discussion unproductive. About me hating v2, I don't. I'm certainly not suggesting we should remove that choice for our users. I do think v3 is a better license, and will probably start licensing my own work as "version 3 or later", but that's a different matter. > but you just said that you would work against both distribution method > upstream chose and limiting the choices our users have. Did I? Please explain, I don't think I said anything of the sort. Or are you still responding to my claim that we could (but shouldn't) relicense the programs GPLv3 only? > > Because according to you that is what would be the effect of not > > pointing to that link, following your reasoning that "it's version 2 > > or later, see GPL" would make it version "lastest" only. > > No. Please read again. By pointing to v2 you say that it is > _corrently_ distributed as v2, not that one isn't allowed to use a later > version. Sorry, I should have been more clear. Be "see GPL" I meant pointing to the "latest version" link. As I understand you, you claim: If we tell people the license is GPL version 2 or later, and point only to the full text of version 3 of the license, users are no longer allowed to choose to use version 2. I disagree with this. I am saying: If we tell people the license is GPL version 2 or later, then it is, no matter what full text we point to (even if we point to an irrelevant license, such as a BSD license). The link is only a service to our users, and doesn't change anything about the actual licensing. > > Assume that this link now points to version 3. The only things that are > > wrong then are: > > - Not all license options which are given to the user are listed (or > > linked to) as full text. Note that this will always be the case, > > because a new version may have been released from the time of writing > > the copyright file. > > It's not about listing all options, because we can't do that: Actually, we can, if we assume that the naming scheme doesn't change, with the formulation that I provided in the previous e-mail. Just don't mention any license, only say "there's a list over there". > a.) the v3 isn't in the pool yet, It will soon be. We're not discussing this short period directly after a new version is released. (I hope. At least I'm not.) > b.) a hypothetical v4 is neither ... No. But when it is, we can update the copyright file. It's not like it's a huge burden or something. It happens less than once a decade. If we really care, we can use the formulation I provided and even have it forwards-compatible. > It never can be complete, because its not a dynamical list. Please > read the GPL again, you don't have to ship the "or later" versions of > the license. The GPL only recommends to ship the license, it doesn't require it. If I write a program and write on it "This is licensed under the GPL, see http://www.gnu.org";, I'm pretty confident that a judge will consider that sufficiently clear intent, and thus enforcable. (Although it falls in the category "no version mentioned, so you may choose any version", as the GPL says.) We provide full texts of licenses because we think it's a good thing, not because it's required. If the license is version 2 or later, we don't have to ship version 2 either. We don't have to ship anything. But we want to. We want to be complete. So we should just do what we consider the most complete. Which is IMO to point to all acceptable versions. > > However, it is totally clear that there is a choice for the user. They > > can choose to use version 2, even if we don't link to that full text. > > Which is a policy violation and the core of my mail. I'll stop responding to this. I'm just saying it is possible to not link to version 2, I'm not suggesting we actually do it. By the way, not linking to version 3 is also a policy violation, because then we haven't provided the full license terms. > > If we want to be really complete, we could use something like: > > See above, that wouldn't be complete. It might be soon, and propably > for a very long line, but bring us to the same situation once a v4 is > released. For a very short time, until the version 4 text is added to the archive. After that, it is automatically complete again. Really, this short time is nothing to worry about. > > If we want to say "the latest version of the GPL", then we certainly > > should point to it. > > And why would we want to do that? What would be the benefit to anyone? > What requires that? As I wrote, some people will consider the newer version better, otherwise it would never have been released. If the user can choose from several licenses, and wants to read as little license text as possible, it is likely that she starts at what some people consider "the best version". And that's always the latest. Sure, there're always also people who consider other versions better. But that's not really relevant, because it doesn't help the user in deciding what version to read. The license writers are the only entity with some authority about this, so it is reasonable to expect that a large part of the users will follow their choice. > > It seems you want to limit the user's freedom to choose GPL3 as a > > license. I should have been more clear here. I'm not trying to put words in your mouth, I'm just asking for a confirmation if this is indeed what you want to do. If not, then I'm misunderstanding you, and I would like to solve that. :-) > "It seems you want to limit the user's freedom to choose GPL2 as a > license." Does that sound any better? Does that get us anywhere? > Please drop accusions that are all but wrong and doesn't help the > situation. Sorry for not being clear about my intentions then. I wasn't trying to accuse you of anything, I was just trying to find out what exactly you are trying to say. For completeness, I don't want to limit the user's freedom to choose GPL2 as a license. > > > That way you tell them already that it's v3 licenced, if you point them > > > to v3. > > > > But it _is_. Since version 3 was released, the programs are licensed v2 > > _and_ v3. That's the whole meaning of "or (at your option) any later > > version". > > Sorry, s/licensed/distributed/, my bad with unaccurate wording. Please apply that rule to my response as well then. :-) Thanks, Bas -- I encourage people to send encrypted e-mail (see http://www.gnupg.org). If you have problems reading my e-mail, use a better reader. Please send the central message of e-mails as plain text in the message body, not as HTML and definitely not as MS Word. Please do not use the MS Word format for attachments either. For more information, see http://pcbcn10.phys.rug.nl/e-mail.html
Attachment:
signature.asc
Description: Digital signature