[Top][All Lists]

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

Re: [Gcl-devel] Re: [gnu.org #48656] Re: GCL compliance with GNU GPL

From: Camm Maguire
Subject: Re: [Gcl-devel] Re: [gnu.org #48656] Re: GCL compliance with GNU GPL
Date: 30 Jun 2003 21:47:15 -0400
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

Greetings, and thank you for your helpful input.

I just thought I'd take a quick moment to note down my recollection of
some of the GCL history in this regard.  This is just to my knowledge,
and may be incorrect in several places!  Dr. William Schelter worked
on GCL for many years before it had any sort of official FSF status --
at this time it was known as KCL and then AKCL.  It is also my
understanding that he knew and/or worked with RMS back in the
'original lisp' days at MIT.  At some point (1992?) he succeeded in
releasing the system under a new name, GCL, with a new license (LGPL),
and having it adopted as the official Common Lisp of the GNU system.
On his web pages, he emphasized the LGPL license as a strength,
protecting the code to the system itself while allowing proprietary
products to be developed with it.  I assume he and RMS agreed on this
at that point.  I believe it is RMS' position that the choice between
LGPL and GPL should be one of 'strategy' -- i.e. based on an
assessment of the comparative uniqueness of the functionality offered
-- and surmise that they decided that as the lisp world had gone
notoriously proprietary, LGPL was preferable.

In any case, Dr. Schelter helped a GCL user develop the readline hook
toward the end of his life.  The code was posted as a patch on one of
the mailing lists.  These bits and a few of the other last works of
Dr. Schelter were gathered together after his passing and folded into
the main source tree.  As no mention of any license implications
accompanied this patch (at least that I've seen), we accepted this
work on the strength of Dr. Schelter's relationship to the project.
Indeed, the first several months of GCL development were spent trying
to retrace his last steps, as some of the code is quite involved.  It
is possible however that the license implications of the readline
patch were never adequately considered due to Dr. Schelter's untimely

It appears we have several options:

1) remove or replace readline in GCL and keep the LGPL
2) make use of the clause you cite below for a multiple license
        strategy depending on whether readline is linked in.
3) make GCL GPL and add a note in the COPYING file allowing
        proprietary images built with proprietary standard common lisp

My first question is concerning the difference in practice between the
LGPL and the GPL with the clause as in 3).  Aren't they completely
equivalent?  It would not appear that option 3) would pose a
significant obstacle to anyone.

That having been said, I feel that, as long as the right to compile
proprietary programs is safeguarded, the decision ought to be a
strategic one made primarily by the FSF.  The code has copyright
comments throughout listing primarily Dr. Schelter as the copyright
holder, through a few other individuals are also named.  As the
primary author is deceased, it would make sense to take some steps
toward a transfer of the copyright to the FSF if this is at all
possible.  I don't know if anyone else can be guaranteed to be around
long enough to fulfill this role otherwise.

Anyway, these are just thoughts.  I'm no lawyer, and will be happy
with anything that keeps everyone else happy :-).

Take care,

David Turner <address@hidden> writes:

> On Mon, 2003-06-30 at 19:36, address@hidden via RT wrote:
> > 
> > The following message is a courtesy copy of an article
> > that has been posted to gmane.lisp.gcl.devel as well.
> > 
> > Hi Camm, thanks for your reply.
> > 
> > > * In message <address@hidden>
> > > * On the subject of "Re: GCL compliance with GNU GPL"
> > > * Sent on 30 Jun 2003 17:57:09 -0400
> > > * Honorable Camm Maguire <address@hidden> writes:
> > >
> > > Readline is a compile time gcl option.  I do agree that linking it in
> > > changes the license.  It does seem that a note in the COPYING file to
> > > this effect might be the most appropriate -- i.e. if you compile with
> > > readline, the product is GPL.  This gives users the opportunity to
> > > compile an lgpl version without readline should they desire.
> I agree that the licensing ought to be clarified.
> > Did you read the e-mail exchange between RMS and Bruno Haible about
> > readline and CLISP?  The link is here:
> > <http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/clisp/clisp/doc/Why-CLISP-is-under-GPL?rev=HEAD>.
> > 
> > for the terminally lazy, the upshot is that the mere possibility of
> > linking with readline means GPL for the whole product.
> > CLISP does come with a "noreadline" library and it can also be linked
> > without readline; CLISP does not use readline on win32 and amiga.
> > Nevertheless it is covered by GNU GPL because it _can_ be linked with
> > readline.  _This is the intent of the GNU GPL_.
> Right, but it could *also* be covered by some other license *when it
> doesn't link with or otherwise is not derived from readline*:
> Here's the quote from the GPL:
> These requirements apply to the modified work as a whole.  If
> identifiable sections of that work are not derived from the Program,
> and can be reasonably considered independent and separate works in
> themselves, then this License, and its terms, do not apply to those
> sections when you distribute them as separate works.  But when you
> distribute the same sections as part of a whole which is a work based
> on the Program, the distribution of the whole must be on the terms of
> this License, whose permissions for other licensees extend to the
> entire whole, and thus to each and every part regardless of who wrote
> it.
> I reiterate that the problem with GCL in 1992 *must not have been* that
> it was LGPL, but that it was under some incompatible license.
> -- 
> -Dave Turner
> GPL Compliance Engineer
> Support my work: http://svcs.affero.net/rm.php?r=novalis&p=FSF
> _______________________________________________
> Gcl-devel mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gcl-devel

Camm Maguire                                            address@hidden
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah

reply via email to

[Prev in Thread] Current Thread [Next in Thread]