On Tue, May 11, 2004 at 03:50:15PM -0700, Josh Triplett wrote: > The DFSG requires that it be possible to make and distribute _all_ > derived works based on the original, as long as such works can be > distributed under the terms of the original license (ignoring the patch > clause DFSG4 for the moment).
You've emphasized the word "all" here. When I read the DFSG, I see "all" used in the context of people, but not in the context of programs. I'd love it if you could show me where I'm wrong. [There's no particular need to assert that I'm wrong -- that's already been done a number of times today.] I'll grant that the "spirit of the DFSG" makes some kind of analogous assertion, but for reasons I've already pointed out I don't think that this exact assertion is completely accurate. > With GPLed works, for example, we have > the right to make any derived work we want, as long as it is under the > GPL, so the GPL satisfies DFSG3. "as long as ..." Sure, it's "all programs except ___" or "not everything". That's pretty much my point. > > I'd want to examine it to determine the kernel interfaces it relies on > > and write something else which uses those same interfaces. > > > > The key issue, in this example, is the quality of information provided > > for this task. > But again, that is a technical concern, not a legal one. Legally, you > should have the right to make such a derived work of the documentation. I believe that relates back to an issue with the GFDL which has a technical solution. -- Raul
