[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Corel's apt frontend



Raul Miller <moth@debian.org> writes:

> I'll still take the stance that because this hasn't gone to court
> that we don't know what the courts will decide.

If the courts agree with your reasoning I think we'll be in big
trouble. As far as I can see it would imply:

1) Linus' exception for system calls is important: it is the only
   thing that prevents any Linux binary from being a deriviate of
   the kernel.

2) Hence, Linux's distribution terms GPL+exception are effectively
   different from GPL alone.

3) GCC and the kernel taken togeter is, by your definition, a
   "program" for compiling C programs. Therefore GCC+Linux is
   a deriviate of GCC.

4) Hence, by the distribution terms for GCC, Debian must distribute
   the GCC+Linux combo under straight GPL, which, by (2) is not
   what is done presently.

5) Which means that Debian cannot simultaneously distribute GCC and
   the Linux kernel, unless we drop Linus' exception (which I think
   is legally possible but I'm not totally sure).

6) Which makes the aggregation of *any* binary in Debian and the
   kernel a "program" to which GPL 2b can be applied.

7) Which means that Debian cannot distribute *any* binary that is
   not GPL.

8) Which, happily, means that we don't have to discuss whether
   anything meets the DFSG - the only relevant question become:
   are the distribution terms the GPL?

9) Repeat steps (1) through (7) for any Linux distribution which
   includes any GNU software at all. We might have to learn to
   love BSD...

.. but of course, only if Raul's reasoning turns out to be valid in
court. *And* if RMS does not release version 3 of the GPL clarifying
the aggregation exception quite soon after the court has sponen. In
which case the "or (at your option) any later version" in the
copyright for dpkg is going to apply.

-- 
Henning Makholm                         "This imposes the restriction on any
                                  procedure statement that the kind and type
                             of each actual parameter be compatible with the
                       kind and type of the corresponding formal parameter."


Reply to: