GNU A Possible Ally?

Mike McDonald mikemac@titian.engr.sgi.com
Tue, 06 May 1997 20:00:37 -0700


>From: "cosc19z5@bayou.uh.edu" <cosc19z5@Bayou.UH.EDU>
>Subject: GNU A Possible Ally?
>To: lispos@math.gatech.edu
>Date: Tue, 6 May 1997 20:28:52 -0500 (CDT)
>
>Hey all.
>
>If this has been mentioned before, then please forgive me.  There
>is a lot of volume here and very often I'll skim the email for
>want of time.
>
>GNU seems to have a vested interest in propagating Lisp as well.
>As a matter of fact, right off the bat I find the following:
>
>  1) One of the goals of GNU is to develop "a Lisp-based window 
>       system through which several Lisp programs and ordinary
>       Unix programs can share a screen", in addition to having
>       Lisp (along with C) as a system programming language (is
>       Guile the culmination of that effort?).

  At one time, many, many moons ago, FSF was talking about writing an
X type of windowing system in lisp. Nothing every came of it as far as
I know. This might be refering to it. I don't think Guile is this
project.

>  2) Emacs (originally written by Stallman, the author of the
>       GNU manifesto which states the above goal), is largely
>       written in Lisp and uses Lisp as its extension language.

  We should steal as much of this code as makes sense. We'll have some
problems similar to what the Guile people encountered. (Dynamic
scoping be one.) We may be able to steal ideas from them too.

>  3) A portable Common Lisp implementation (already done right? GCL?)

  GCL is Lisp for C hackers! It sucks big time! (AKCL wasn't bad in
it's day.) This is a classic example of where the insistance on the
GNU license virus caused the FSF to shot themselves in the head. CMUCL
would have been a better basis but it got a doing anything you want
but don't blame us license on it. GCL is not a viable option for this
project.

>[There may be more]

>All opinions are welcome.

  I'm in favor of stealing code that meets our needs from where ever
we can. Better yet is stealing ideas. But if something doesn't match,
let's not tie ourselves down.

  Mike McDonald
  mikemac@engr.sgi.com