On Nov 22, 2007 3:46 AM, Herbert Snorrason <[EMAIL PROTECTED]> wrote:
> I believe the 'viral' clause of the GPL to be a good idea, and that it > is in more or less everyone's interest to create the largest _single_ pool of free software commons possible; in that case, adoption of the prevalent licence is for the best. The problem with the viral clause of the GPL, is that it has the effect of painting the FSF and others into a corner. I agree that it's a good idea to try to create a large commons, and something like the viral clause is a good idea. However, I feel that it should be more flexible, allowing it to be combined with any OSI approved license. This is a difficult issue precisely because standards are a good thing, and flexibility is also a good thing. So what is the right balance? That is what makes this issue so difficult. In the case of a software licenses, do we want one standard license for every piece of software on the planet? Or can we agree on a few basic points, and allow flexibility from there? In this case, my preference is for the latter -- but reasonable people may conclude that the former is the better choice in this case. > I believe the need for any of these licences to exist to be a major flaw of copyright law, and that said law should be reconsidered from the ground > up. > This is a sentiment that I can wholeheartedly agree with. Intellectual property has become intellectual monopoly, because of rent seeking activities. > The CDDL is shorter, but in my view far harder to understand. The GPL > was written by a layman, while the CDDL is the product of lawyers, > which does much to explain it. I just went back and reread the licenses, and I found the CDDL to be easier to understand and cleaner. The fact that the CDDL was written by lawyers rather than a layman is a big positive. Professionals should write licenses, while the layman should simply tell the professionals what their objectives are. So this is a mark in favor of the CDDL. > IMO there's been a lot of FUD around GPLv3. Its patent clauses, for > instance, are no harsher than the CDDL's. The problem, of course, is that people are assuming that GPLv3 will become widespread, and manage to actually cause problems for the MAD balance of the patent arms race. Other issues with the GPLv3 would mostly be the DRM stuff - which is really only > directed against the more insane parts of the DMCA, not DRM as such. I don't see where the FUD would be in reference to GPLv3 patent protection. I don't think the people using the GPL, or the CDDL for that matter, would be particularly warm to the idea of software patents in any case. As to the DRM stuff, I have no objection to the clauses in the GPLv3. However, reading the GPL gives one a sense that one is watching a latter day Don Quixote tilting at windmills. It strikes me as highly unlikely that one will be able to save the world from itself by use of a software license. Nonetheless, Don "RMS" Quixote's brand of insanity is a lovable one, and I can understand the attraction. > The whole idea of the GPL is to use the legal structure to enforce the creation of an expanding commons. It's a clever hack, which unfortunately can't elegantly get around the legal system's insistence on specific > licences. The combination of copyleft and virality thus creates the undesirable phenomenon of exclusivity. This is a bug, but its root cause lies in the > legal system... I agree that the phenomenon of what exclusivity is a bug. I disagree that it lies in the legal system. As I stated above, this could have been solved by stating that the license is compatible with OSI approved licenses, or other similar means. I believe that the reason that this was not pursued, is that the key players such as RMS, do not want to play well with others. It's worth noting that the GPLv3 contains a clause which allows certain extensions to it. This kind of flexibility is a step in the right direction but it does not go far enough. And for that last part: IANAL, but if you have a component that's > clearly enough demarcated, you should be able to have it LGPL and > legally combine it with the OpenSolaris internals. I'm just not sure > there's a point. > I believe there is a point. Take ZFS as an example. Linux could benefit by adding support for ZFS -- as Linux has no decent filesystem at this point. They could add support for ZFS by using an LGPL bridge -- but it's an option they won't even consider because of GPL only snobbery. Of course I'm not being entirely fair here -- part of the problem is that ZFS comes from the Solaris community, and they don't want something that they did not invent. I've been reading a lot of history lately. Thomas Sowell's Cultural Trilogy is utterly fascinating. One of the points that this trilogy really drives home is that no one civilization invented every good idea -- the civilizations that have been wildly successful are the ones that have been open to the adoption of many ideas from other places. Thus, I believe it is in the best interests of the Solaris and Linux communities to play well with each other and try to learn as much as possible from each other. To the extent that the GPL gets in the way of that, the GPL is a hindrance and not a help to free software. -- Cheers, ~Thomas "I believe that only scientists can understand the universe. It is not so much that I have confidence in scientists being right, but that I have so much in nonscientists being wrong." - Isaac Asimov
_______________________________________________ gnusol-users mailing list [email protected] http://lists.sonic.net/mailman/listinfo/gnusol-users
