On Thu, 17 Aug 2006 19:11:19 +0100, Alex Bligh wrote: > > Well (e.g.) from the link you provided: > Finally, for non-software works the "copyright" line included at > the start of the "source code" of the work is modified in language > slightly: > > The whole thing seems to presume something called "source code" exists > for these designs, as a concept different from "binary form". The GPL (I'll uses "GPL" here to refer to GPLv2 always) does talk about "source code" but carefully defines that term as "the preferred form of the work for making modifications" to the work. I think that definition is a stroke of genius in the GPL, particularly because it allows straightforward application to so many different types of work. And I don't think the GPL presumes two distinct forms at all. Only one section of the GPL talks about two separate forms. That is section 3 which grants an additional permission to distribute an "object code or executable form" of the work under certain conditions. If, for any given work, no such form can be said to exist, then all effective rights granted by the GPL are still intact. One could not rationally complain about the inability to distribute something that does not exist. Another way to look at this is that section 3 only comes into play if someone desired to re-distribute by relying on the additional permissions of section 3. Such a person would necessarily have to have a notion of separate "source" and "object" forms of the work, or else they could simply rely on sections 1 and 2. So, anyone that relies on any GPL "assumptions" of two forms must necessarily satisfy that assumption before it would make sense to want to rely on it. Similarly, one could point to something like section 2.c which contains a requirement to "print or display an announcement". But such a condition also causes no problem since it applies only "if the modified program normally reads commands interactively when run". So, sure there are some pieces of the GPL that do not apply to some forms of work. But I don't see any that would prevent a recipient from exercising the standard, reciprocal rights of modification and redistribution that the GPL is designed to grant. Do you? > It isn't clear (at least to me) that it makes sense, and indeed > what the potential licensor's obligations are. Could you be more specific about a situation in which the licensor's obligations are not clear to you? > I suppose one > way to resolve this is to dual-license it under the GPL and the > CC licenses. Sure. That could work I think. > Sadly, the fact people X put Y under license Z does not mean license > Z is effective for the purpose people X thought it would be. Certainly not. You had requested a reciprocal license suitable for non-software that would be accepted as "free" by Debian. I was merely looking for empirical evidence to suggest that the GPL satisfies those constraints. -Carl
Attachment:
pgpvrE0dAzZpQ.pgp
Description: PGP signature