Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site mit-eddie.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!harvard!think!mit-eddie!
martillo%mit-herac...@mit-athena.ARPA
From: martillo%mit-herac...@mit-athena.ARPA
Newsgroups: net.emacs
Subject: Using Gnu Emacs (Clarification)
Message-ID: <4440@mit-eddie.UUCP>
Date: Wed, 12-Jun-85 08:06:50 EDT
Article-I.D.: mit-eddi.4440
Posted: Wed Jun 12 08:06:50 1985
Date-Received: Sun, 16-Jun-85 07:27:41 EDT
Sender: dae...@mit-eddi.UUCP
Organization: MIT, Cambridge, MA
Lines: 17

From: martillo%mit-herac...@mit-athena.ARPA (Joaquim Martillo)
I want to emphasize I am  not telling people  to stop using  gnu emacs
because of rumored legal hassles.  I mean the sources  of these rumors
should either come   up with something  concrete  to  back  up   their
statements or they should shut up.  Until then we will continue to use
gnu emacs.

I should also like  to point out that  having  looked at the internals
of Montgomery Emacs  (from ATT),  CCA Emacs, Unipress (Gosling)  Emacs
and Gnu Emacs.  CCA Emacs is a lot more like ATT  Emacs than Gnu Emacs
is like  Unipress Emacs.  An example of  this  both Montgomery and CCA
emacs handle glob (*) via a backgroup shell process  to which emacs is
writing and from  which emacs is  reading.  Actually, since  Gnu Emacs
contains  a full implementation of   Lisp  (unlike  Gosling Emacs),  I
would say that conceptually Gnu Emacs is more similar to Prime Emacs.  

Yakim Martillo

Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site mit-eddie.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!shane@mit-grape-nehi
From: shane@mit-grape-nehi
Newsgroups: net.emacs
Subject: Using Gnu Emacs (Clarification)
Message-ID: <4441@mit-eddie.UUCP>
Date: Wed, 12-Jun-85 08:36:47 EDT
Article-I.D.: mit-eddi.4441
Posted: Wed Jun 12 08:36:47 1985
Date-Received: Thu, 13-Jun-85 01:59:24 EDT
Sender: dae...@mit-eddi.UUCP
Organization: MIT, Cambridge, MA
Lines: 15

From: shane@mit-grape-nehi (K. Shane Hartman)

Not to mention the fact that there are only 6 or 7 files that have
Gosling's name in them (for which RMS has permission to distribute)
and if you run diff on Gnu and Unipress (or Gosling) sources the
output fills reams of screens.  I did this last night out of curiosity
after your messages; the differences are significantly more than cosmetic.

Anyone who had to ever go to another terminal to kill a Unipress emacs
because the mucklisp ran rampant would know why `the sources of these
rumors' are muttering about the best software system I have ever seen
under Unix (a trademark of some giganto conglomerate whose name I have
forgotten).

-[Shane]->

Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site masscomp.UUCP
Path: utzoo!watmath!clyde!bonnie!masscomp!z
From: z...@masscomp.UUCP (Steve Zimmerman)
Newsgroups: net.emacs
Subject: Re: Using Gnu Emacs (Clarification)
Message-ID: <721@masscomp.UUCP>
Date: Mon, 17-Jun-85 12:20:57 EDT
Article-I.D.: masscomp.721
Posted: Mon Jun 17 12:20:57 1985
Date-Received: Tue, 18-Jun-85 04:39:13 EDT
References: <4440@mit-eddie.UUCP>
Organization: Masscomp - Westford, MA
Lines: 72

I expected some flames in reply to my mesage, and I must say, I'm not
disappointed.

I think it's time for me to clear up some misinformation about CCA EMACS
being spread on the net, largely by Yakim Martillo.  Since its release
as a product, CCA EMACS has contained no code from Montgomery's EMACS.
Early, free versions of CCA EMACS contained some of Montgomery's code
(which was being freely distributed by other means and which had no
indication in the sources that it was in any way proprietary).  When CCA
decided to sell EMACS as a product, it made inquiries at Bell Labs to
make sure that they did not consider it proprietary.  Bell Labs took
several months studying the issue, during which time they discovered the
widespread distribution of Montgomery's EMACS, and the fact that they
had apparently not taken the necessary care to protect this code as
demanded by trade secret law.  Nevertheless, they informed CCA that they
still considered the code to be proprietary, and CCA offered to remove
Montgomery's code.  Bell Labs accepted the offer, and the issue has been
closed since then.  There is also no evidence that Bell Labs was
"unhappy" with either me or CCA; their correspondence with us always
mentioned our productive past relationship, thanked us for cooperating
fully with them and for our understanding of the matter, etc.  In
general, they seemed to be grateful that we took the time to check this
situation out with them, and that we did not contest their claim of the
proprietariness of Montgomery's code.

> An example of  this  both Montgomery and CCA
> emacs handle glob (*) via a backgroup shell process  to which emacs is
> writing and from  which emacs is  reading.  

This statement is patently false.  There was no background shell
process, no "glob" routine, nor anything else performing the same or
similar functions in the version of Montgomery's EMACS (4.0) that I once
used, which always ran as a single process and which is the only version
I ever saw.  These features were all completely original with me.
Similarly, the statement about all the file names being the same is also
false.  For one thing, Montgomery's EMACS 4.0 had around twelve source
files, while mine has had over sixty for several years.  Now, it is
known that my original hybrid of my EMACS and Montgomery's made its way
to many Bell Labs sites.  It appears that Mr. Martillo saw some of this
code and thought it was Montgomery's, not realizing that much of it was
mine.  As it is, Mr. Martillo notes that he never saw a pure version of
Montgomery's Emacs.

On another note, RMS says that he got permission from Fen Labalme to
distribute Gosling's Emacs (or portions thereof).  When I was at
Usenix, I talked to Unipress, and this is what RMS told them as well.
However, according to Unipress, Labalme has no right to distribute
Gosling's Emacs, and when RMS was confronted with this, he was unable to
produce any written permission from either Gosling or Unipress.

When I talked with Unipress, they felt that Gosling was clearly in
violation of the law.  (Note that copyright law covers "derivative
works" as well as original source.)  However, with GNU Emacs already
distributed all over the place, they seemed to feel that trying to step
in at this late date would probably be counterproductive.  So, I gather
that they will not press this issue.  However, I would hope that RMS is
more careful when it comes to the writing of GNU itself.

As for "browbeating" people, what does RMS think I am suggesting?  That
everyone throw away their copy of GNU Emacs and forget that it ever
existed?  Not at all.  It's simply that I'm sure that the creator of the
original EMACS is intelligent enough to write an entire UNIX Emacs
himself; under the circumstances, I hope he would do the honorable thing
and rewrite from scratch those portions of GNU Emacs derived from
Gosling's code.  If he does not, I think a cloud will hang over the GNU
project for some time to come.

	Steve Zimmerman
	MASSCOMP

"The opinions expressed above are not necessarily those of MASSCOMP,
Uniworks, or anybody else."