🚀 go-pugleaf

   ----- Transcript of session follows -----
Connected to sage:
>>> RCPT To:<phil@iseng.compnews.co.uk>
<<< 550 User unknown
550 phil@iseng... User unknown

   ----- Unsent message follows -----
Received: from icdoc.UUCP by hyperion.cps.compnews.co.uk; Sat, 8 Feb 92 06:11:43 GMT
Original-Via:  uk.ac.uknet  ; 8 Feb 92 5:36 GMT
Received: from news.cis.ohio-state.edu by eros.uknet.ac.uk via EUnet 
          with SMTP (PP) id <3045-0@eros.uknet.ac.uk>;
          Sat, 8 Feb 1992 04:45:02 +0000
Received: by news.cis.ohio-state.edu (5.61-kk/5.911008) id AA13560;
          Tue, 4 Feb 92 22:46:36 -0500
Received: from USENET by news.cis.ohio-state.edu with netnews 
          for gnu-misc-discuss-mail@cis.ohio-state.edu (gnu-misc-discuss@cis.ohio-state.edu);
          contact usenet@news.cis.ohio-state.edu if you have questions.
To: gnu-misc-discuss@cis.ohio-state.edu
Date: 5 Feb 92 03:26:29 GMT
Message-Id: <BURLEY.92Feb4222629@geech.gnu.ai.mit.edu>
Organization: Free Software Foundation 545 Tech Square Cambridge, MA 02139
From: Craig Burley <burley@geech.gnu.ai.mit.edu>
Original-Sender: gnu-misc-discuss-request@edu.ohio-state.cis
References: <BURLEY.92Jan29180034@geech.gnu.ai.mit.edu>,
            <Pi1eFB7w164w@mantis.co.uk>s
Subject: Re: Too Good to be True?
Original-Sender: gnu-misc-discuss-request@cis.ohio-state.edu
Sender: gnu-misc-discuss-post@cps.compnews.co.uk

In article <Pi1eFB7w164w@mantis.co.uk> mathew@mantis.co.uk (Lieutenant 030) writes:

   OK.  Suppose I am a user.  Suppose I have just obtained a copy of GNU Emacs.
   What rights does the GPL give me over and above those I would have if GNU
   Emacs were public domain?

The rights to the source code, for a nominal fee, and to share that source
code with others.

And that goes for whether the copy you obtained was from the FSF for machines
it supports, _or_ for, say, the Macintosh, which the FSF currently refuses
to support.  You can't obtain a copy of GNU Emacs for the Mac (if such a thing
exists) without also ending up with the rights to obtain the source code
for that _exact port_ (not just "original EMACS, but we threw in a lot of
special Mac code to make it work, and we don't want you to see that stuff").

Therefore, when you get GNU EMACS vs., say, a PD EMACS, you have the following
additional rights (and you already know about all the "subtractional" rights):

    -  You get the source, or at least access to it for a nominal fee for
       up to three years

    -  You can share the executables and/or source with others (i.e.
       redistribute)

    -  You can modify the source and build derivative versions

    -  As long as you give these rights to others, you can distribute (for
       profit or for free) the modifications and/or derivative versions to
       others

Probably there are others, but to me, these are the important rights.

By the way, I see no reason why someone can't ship "highly proprietary"
(i.e. licensed) source and/or objects separate from GPL-protected source,
but intended for use when linked in with such code.

For example, if you wanted to sell a proprietary page-layout system that
integrated with GNU EMACS, you could license/sell the code as source or
objects with the notice that it only worked when the licensee obtained and
built GNU EMACS, linking in your changed modules.  As long as what you
distribute has no GPL-protected code in it, and as long as the licensee
does not redistribute the rebuilt GNU EMACS (but just uses it internally),
I can't think of any stipulation in the GPL offhand that prevents you from
doing that.

Now, maybe there's something about the GPL I don't know (in fact that is
certain :-).  But it seems to me that since you can license _any_ code
_you_ write under _any_ terms as long as it doesn't include or cohabitate
in a running application with someone else's non-PD code (depending on their
licensing requirements), there's little the GPL can do to then prevent you
from somehow communicating to people that your code is best used (or perhaps
only usable) when linked in with GNU EMACS a certain way.

Is this desirable?  Perhaps it might suggest a "way out" for you.  For GNU
people like myself, it seems no less desirable than typical commercial
licenses, except that it has the advantage of spreading GNU software _and_
the GPL (obviously anybody buying your code will have to understand the GPL
to make sure things are legit and know what they're doing with GNU EMACS).
Your customers might even get to where they try to convince you to put your
code under the GPL or at least PD.  (Replace "you" with "someone who turns
your PD code into proprietary stuff" if necessary to avoid taking offense.)

It also doesn't seem to prevent your customers from acquiring and installing
bug fixes to GNU EMACS and rebuilding it, with your code still linked in.
So it doesn't seem to greatly violate the spirit of the GPL.  Of course, if
your code depends greatly on knowing a lot about the internals of GNU EMACS,
there will be problems, but your customers will interpret them increasingly
as problems with your code rather than with GNU EMACS, and thus keep you
busy by paying for updates (which should make you very happy).

And, if your product is successful in some quarters, maybe some of your
customers will decide to donate money to the FSF (or to me :-) towards getting
a GPL version of your system developed, so they have _all_ the source code
and can make bug fixes in your code without depending solely on you.  That
would be fun.
--

James Craig Burley, Software Craftsperson    burley@gnu.ai.mit.edu
Member of the League for Programming Freedom (LPF)