Raul Miller wrote:
>On 5/12/05, Humberto Massa <[EMAIL PROTECTED]> wrote: > >>Suppose the libc runtime is given in some system by a work named >>gpld_libc. Is hello_world.c a derivative work of gpld_libc ? I >>don't think so. >> >>#include <stdio.h> >> >>int main(int, char**) >>{ >> puts("Hi"); >> return 0; >>} >> >> >>What is a dynamically compiled file hello_world? An intangible >>medium containing: my hello_world.c work, translated automatically >>+ (possibly) some (non-protectable by virtue of defining an >>interface) bits grinded by the compiler/linker, extracted from >>gpld_libc (eg, compiler macros, etc). Can I distribute it under >>any license I see fit? Yes, I think so. > >That depends on what went into the binary. > Usually, in a dynamically-linked binary (as is the case you responded to), the only links extraneous to the original work are those that are not protectable.
> >For example, if hello_world now provides fourth order Runga-Kutta >solutions, I'd probably need to know whether the code which >provides this functionality is licensed such that I can >redistribute it. > I object, because this paragraph is irrelevant. I already stated what is hello_world.c, and it's unreasonable to think that its dynamically linking would provide or even contain Runga-Kutta-whatever.
>If hello_world does nothing more than print a string, this might >not be a big issue. Or it might be. For example, if "print a >string" requires a full copy of some proprietary firmware because >the compiler target was an emulator for some proprietary hardware. > I will do my "repeated assertion" act: It's a dynamically linked executable, for the love of $DEITY!
> >I"ll avoid presenting any GPL examples, since that seems overly >contentious. > Oh, but those are the juicy ones.
-- HTH Massa
-- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]