Raul Miller wrote: >> On Thu, May 06, 2004 at 09:57:41AM -0400, Nathanael Nerode wrote: >> > Now, again, some restrictions on creating derived works are generally >> > considered acceptable. But required inclusion of arbitrary lumps of >> > text in a particular manner certainly isn't one of them (even with the >> > oft-ignored GFDL restriction that they must be 'off topic'). > > On Sun, May 09, 2004 at 12:41:52PM -0400, Anthony DeRobertis wrote: >> The oft-ignored restriction that invarient sections must be off-topic >> probably just makes the DFSG 3 problems worse: It also limits derived >> works to not covering certain topics (or, at least makes their status >> *very* unclear if they do cover those topics). > > Huh? This would be true if the rules about secondary sections applied > to the document as a whole. > > But they don't. Um, are you listening?
>> I can't use part of the GNU Emacs manual in my essay on software >> freedom, because I'd have to include an invarient section about that >> subject (but I can't because invarient sections have to be off-topic...) > > They have to be about the relationship between the authors and the > document. They're clunky. But that's not the same thing as forbidding > them. Perhaps you missed the point? He would have to include the GNU Manifesto in order to use part of the GNU Emacs manual (beyond what's allowed by fair use). He plans to license his essay under the GFDL, same as the GNU Emacs manual. But in his essay on software freedom, the GNU Manifesto would not be a Secondary Section. So he couldn't use part of the GNU Emacs manual for this purpose at all (beyond what's allowed by fair use). Or at least we haven't figured out a way to do so. > You'd be much worse off trying to include example code from some "no > changes, only patches" source code license. (1) Is a selection actually a change? (2) He could simply license his whole manual (source code) under that license, and supply it in the bizarre form of a patch to the original source code; to be free, the license must permit the creation and redistribution of the derivative object code, so the manual viewed by the enduser (object code) wouldn't have any of this oddity left. > You'd havr even worse problems if you tried to include examples from > two sources which had incompatible licenses. Well, there's that -- there really is specific language in the DFSG which indicates that we don't care about that, however, as others have pointed out. > In other words, this is ugly, but it's not even close to the most serious > issue with stuff we do distribute. Au contraire, it is very close. -- There are none so blind as those who will not see.
