Merijn de Weerd <[EMAIL PROTECTED]> writes: > On 2006-10-16, David Kastrup <[EMAIL PROTECTED]> wrote: >> Merijn de Weerd <[EMAIL PROTECTED]> writes: >>> The more correct terminology is that the OP can only distribute >>> the resulting (i.e. linked) work as a whole under the GPL. >>> If he cannot do that, "then as a consequence [the OP] may not >>> distribute the Program at all." No infection, just a legal >>> choice: either release under GPL, or don't release at all. >> >> The unlinked work may be affected, too, if its purpose can't be met >> without linking, and thus the act of linking from the enduser becomes >> a formality instead of an available technical option. > > Then you're introducing another reason why the unlinked work may > be a derivative of the GPL work. Sure. It could also have > copy&pasted GPL code inside. > > Let's focus on things that can "reasonably [be] considered > independent and separate works". > >> However, if >> there are practical uses without linking to the GPLed library (for >> example, if an API-compatible different library exists that could be >> employed equally well), then the case might become shaky where the >> distribution of the unlinked executable or the source is concerned. > > Shaky? Not according to the GPL: > "this License, and its terms, do not apply to those [independent and > separate works] when you distribute them as separate works."
Well, the question is when a work can be considered independent and separate. Where there are several different options for linking, this criterion is likely met. -- David Kastrup, Kriemhildstr. 15, 44793 Bochum _______________________________________________ gnu-misc-discuss mailing list gnu-misc-discuss@gnu.org http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss