Re: [GPL] No linking with proprietary programs: where?
On Tue, Mar 14, 2000 at 10:06:11AM -0500, Brian Ristuccia wrote:
> On Tue, Mar 14, 2000 at 03:30:40PM +0100, Stephane Bortzmeyer wrote:
> > On Tuesday 14 March 2000, at 8 h 54, the keyboard of SCOTT FENTON
> > <firstname.lastname@example.org> wrote:
> > > Section 2b is the viral clause and it reads as follows:
> > > You must cause any work that you distribute or publish, that in whole or
> > > in part contains or is derived from the
> > > Program or any part thereof, to be licensed as a whole at no charge to
> > > all third parties under the terms of this License.
> > It is not very clear from Section 2b that linking makes a program GPL-infected, and dynamic linking too (see the rms/Torvalds discussion about binary modules in the Linux kernel) but not popen() or system() (see the discussion about the apt tool by Corel which popens dpkg).
> Let's not be silly here. Whether or not dynamicly linking with Qt causes a
> GPL incompatibility is a nonissue because using the Qt include files causes
> Qt code to be copied verbatim into the body your program's source code and
> thus makes those parts of Qt a part of your program. I personally consider
> dynamic linking use, and as we all know, use is beyond the scope of the GPL.
> But when bits of _executable C code_ in the form of macros from the Qt
> include files are copied into your program it's more than just use - it
> constitutes distribution as well. Those who aren't convinced should check
> out some of the macros in the Qt include files for themselves.
> That said, I think we can all pretty well agree that copying code verbatim
> from or into the body of a program's source code is covered by the GPL and
> that this discussion can safely end here. I'm tired of hearing about it.
If tehy made headers public domain, would it resolve problem ?