Unpublished GPL Software

Luca Andreucci andrew at andrew.org
Thu Jul 11 09:20:27 UTC 2002


Jan Wildeboer wrote:
 > Rodger Etz-Brown wrote:
 >> The issue now is that neither company A nor B have published the
 >> software. But there is interest from outside those companies to see
 >> and use the software.
 > All parties that have received the code (in binary or in source) have
 > the freedom to distribute. Gratis or for a fee of any amount.
 > Sounds good.
 > But it also means that they have the freedom to not distribute. The
 > only thing is that when company B decides to publish company A cant
 > do anything against it.

#include <IANAL.h>

We seem to agree that the software IS licensed under the GPL whether 
it's been distributed or not, and, as the GPL states:

   Activities other than copying, distribution and modification are not
   covered by this License; they are outside its scope.

IMHO it will all come down to the definition of "software distribution".
  Specifically, what has to be made clear is whether the "transfering" 
of the program from B to A, in source or binary form, constitutes 
"distribution".  If it does, GPL would apply.

I have read the GPL without finding answers or clues.

Good sense suggests me that distribution can't be implied.  Let's take 
this concept to the extreme.  A programmer writes a _brand new_ (as 
opposed to derivative) work, with the intention of licensing it under 
the GPL, so he puts GPL header into the source, references into the 
docs, LICENSE file and all; but defers the final decision to the future. 
  Now let's say another user gets his hands on the source code without 
the original author's knowledge and/or permit, and starts distributing it.

Considering the distribution as legally valid would mean that the mere 
adding of some files and notes to the source implies that the program is 
bound to be distributed as GPL.  Which makes sense in some way, but 
raises some question:

- If the GPL status is granted by the files, who grants that the second 
user (the felon) hasn't modified the source, adding GPL hooks?
- If it is the other way around, who grants that the programmer won't 
repudiate the GPL nature at some future point?

This all reminds me of an old Peanuts strip, where Peppermint Patty 
comes to the conclusion that since she realized there are more questions 
than answers in life, you'd better be the one who makes questions :-)))

 > Personally I think that is not what the spirit of the free software
 > is about, but that's my personla opinion.

Seems to me that the freedom NOT to redistribute is sort of implied. 
How could you possibly enforce the opposite?   But I think we would all 
agree that it goes against the spirit of FS by design.

--andrew




More information about the Discussion mailing list