Subject: Re: Who holds the copyright?
From: Ian Lance Taylor <ian@airs.com>
Date: 21 Nov 1999 22:36:04 -0500

   Date: Sun, 21 Nov 1999 16:00:31 -0500
   From: Brian Bartholomew <bb@wv.com>

   Cygnus/gcc is a better example than qmail.  There is a commercial
   version of gcc that behaves proprietarily*, and Cygnus seeks to
   distance themselves from the libre version to maintain sales.  The
   free community has not forked the libre gcc in righteous anger against
   Cygnus.  This is an example contrary to Bruce's contention, that the
   last libre version of a newly proprietary product will be picked up by
   volunteers against the vendor.  If I understand qmail correctly, it's
   an example of an libre version orphaned by the vendor, with no
   corresponding proprietary version.  So it's different than the
   situation I was talking about.

   * Acknowledged point of contention.  The example would be better if
     the commercial gcc wasn't GPL.  I'm eager to see if Red Hat will put
     all of the embedded cpu gcc variants up for ftp.

For those who are new to this mailing list, Brian has gone through
this sort of thing before.  I've never seen him change his tune.  As
far as I can tell nobody else agrees with him.

Given that Cygnus puts all their changes into the publically available
version of gcc, given that Cygnus devotes significant resources to
maintaining the publically available version of gcc, given that all
Cygnus distributions of gcc include source, are under the GPL, and are
not proprietary under any possible definition of the word, it is
beyond ludicrous to expect the free community to fork the libre gcc in
righteous anger.

To my knowledge, no version of qmail was libre.

You might have better luck using Sendmail vs. Sendmail Inc., or Tcl/Tk
vs. Scriptics.  In both cases, though, I believe the libre version
remains libre.

Another possibility would be *BSD vs. BSDI; I believe that the BSDI
code is proprietary.  In that case, the free software community did
arguably fork, but I admit the situation can not be described in such
simple terms.

Ian