David Kastrup wrote: > >> >> That does not help in itself. Creating an artificial API does > >> >> not create an independent work abstraction as long as the > >> >> library remains the only actual implementation of that API. > >> >> > >> > I tried to point it out during OpenGl example, probably wrong way. > >> > There is an API (well documented) for library and there are two or > >> > more libraries using the API. One is GPL licensed and the rest > >> > not. The product doesnt have to be a GPL licensed. > >> > > >> > Personally I think this falls under "If identifiable sections of that > >> > work are not derived from the Program,.." text from section 2, but > >> > anyway: > >> > True/False? > >> > >> Uh what? > >> > > I am giving parallel example to glibc working on different kernels with > > different licenses. > > Program can use one of different OpenGl implementation with different > > licenses. > > I am afraid that both your case as well as the question appear so > confused that it is not even possible guess what you mean here. > You cant say that only kernel and glibc can interact and that kernel can be GPL and glibc can be LGPL. What condition does software A and B fullfill in order to be used same way as kernel and glibc.
Honza _______________________________________________ gnu-misc-discuss mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss
