For once, I applaud Linus
|
| Author | Content |
|---|---|
| Libervis Sep 26, 2006 8:15 AM EDT |
For once I would applaud Linus for trying to speak in a positive rather than a negative way. When writing directly against GPLv3 the whole discussion tends to put a shadow on all GPL, no matter the version. When you instead try to point out what you consider advantageous about GPLv2 you are approaching the issue from a much more positive standpoint. I think I understand what he is trying to say. The DRM and software patents are the issue of the day, so to speak. The GPLv3 provisions on DRM and software patents in principle don't really ask for anything more than GPLv2 does. The goal is still the same. It merely makes it more explicit. Linus believes that this explicity is not necessary, but unfortunately I am finding it hard to believe that. While the message begind GPLv2 is indeed very simple (tit for tat) and understandable, it is also not only a message, but a copyright license meant to serve a s shield that would preserve the "tit for tat" situation to remain true for software it covers. Unfortunately, DRM has the power to completely ruin that. So how do we address it? How do we prevent "tivoization" while still using only GPLv2? That's the basic question here because after all the criticism of GPLv3 and now praise of GPLv2 they haven't yet offered a very constructive solution for the tivoization and even software patents problem. I am getting really hungry of that now. |
| nalf38 Sep 26, 2006 8:44 AM EDT |
You know, you've actually made me want to read the GPL3 draft a little more closely. Thanks for the honest answer. Sometimes it's hard to separate the messenger from the message, and RMS rubs me the wrong way. |
| dcparris Sep 26, 2006 9:07 AM EDT |
Nalf38:
> Sometimes it's hard to separate the messenger from the message, and RMS rubs me the wrong way. That's exactly what we have to be careful of. A number of people, some of whom I respect, refuse to give RMS an inch simply because he "rubs them the wrong way". The problem is that we risk throwing the baby out with the bath water in doing so. If I responded that way to everyone that rubs me the wrong way, I'd be sitting in a cave in the Himalayas somewhere. I really think that much of the hoopla involves personality differences than real technical issues, despite the public statements made. That said, there might be merit to some of the claims of the kernel developers. It is most definitely a complex issue. |
| tuxchick2 Sep 26, 2006 10:04 AM EDT |
It is complex, and the last thing we want is a GPL fork. That means disaster. If there's anything I've learned from my years in the biz, it's that software licenses bite in all sorts of subtle ways. Here are a few examples: -the Pine mail client does not allow distribution of binaries patched with third-party code. Hey, thanks, that is sooo user-friendly. We all just love to waste time patching and re-compiling sources. -Firefox and Debian are having a name and trademark spat because of third-party modifications. sheesh, don't they have any real work to do? -the XFree86 project persons aced themselves right out of every major Linux distribution with their license change A license fork in the kernel would have huge repercusssions. So it's vital that this be resolved in a way that prevents a fork. |
| dthacker Sep 26, 2006 10:49 AM EDT |
Linus's article appeals to me on an emotional level. However, lawyers rarely work there. TC2, could you please comment on why GPLv2 and GPLv3 could not co-exist? Dave |
| tuxchick2 Sep 26, 2006 11:02 AM EDT |
dt, it all boils down to license compability. If GPL3 adds additional restrictions, as some folks say it does, then it conflicts with GPL2. Which opens the door to having two different kernels- one that is GPL2-only, and one that is mixed. Which presents the possibility of yet more complications and hassle for distributors and end-users, and devs might as well go to law school. I could be wrong, that's just how it looks down here in confused-land. |
| jimf Sep 26, 2006 1:43 PM EDT |
I pretty much agree with you tuxchick. I spent about 4 hours last night reading the last draft of GPLv3 and I'm a bit overloaded at this point, but, the revisions of the portions duplicated by GPLv2 are shaping up nicely. Much, but not all, of the aquwardness in the old licence has been eliminated and that's still being worked on. I suggest that everyone have a look at this : http://gplv3.fsf.org/comments/gplv3-draft-2.html . Plan on spending siginificant time with it. It's not Greek, but close. It would apear that there is even a possibility of the '[13.[8] Geographical Limitations.' section being eliminated. This section creates the real possibility of software censorship, something that should be anathma to GPL. Governments may limit software distribution, but, GPL should not assist them in the process. So removal of the section would be a real positive. The section 3. relating to DRM is of course the sticking point. My take on this is that much, though not all, of the problem is in the language of the draft as it now exists. It would appear that Lawers 'really' have a problem decribing a technical process or specification. As it now exists, I don't see how one would know what they are trying to say. This section really needs to be cleaned up, but again, they appear to be working on it. Let's hope the final is much more understandable, and, palatable to the Developers. Incidently, the mail list archives are here: http://gplv3.fsf.org/pipermail/info-gplv3/ IMO once a month anouncements are not enough communication to the public about this important process. FSF needs to do better. |
Posting in this forum is limited to members of the group: [ForumMods, SITEADMINS, MEMBERS.]
Becoming a member of LXer is easy and free. Join Us!