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!rms@mit-prep
From: rms@mit-prep
Newsgroups: net.emacs
Subject: honor
Message-ID: <4489@mit-eddie.UUCP>
Date: Tue, 18-Jun-85 02:57:42 EDT
Article-I.D.: mit-eddi.4489
Posted: Tue Jun 18 02:57:42 1985
Date-Received: Wed, 19-Jun-85 03:32:28 EDT
Sender: dae...@mit-eddi.UUCP
Organization: MIT, Cambridge, MA
Lines: 49
From: Richard M. Stallman
I can't understand how honor could require me to replace any
of the code I am distributing in GNU Emacs.
Using the modified Gosling display code in GNU Emacs is
not immoral. Redistributing any code is always morally good.
It's not disobeying the spirit of the law. (We know
we really do have permission.)
It's not disobeying the letter of the law. (That is, we expect
we could win if we were sued. That's the most anyone can ever say.)
It's not breaking any promise Fen or I made.
Where is the dishonor?
Here is why I don't want to rewrite the display code:
Any display code would have a considerable resemblance to that display
code, just by virtue of doing the same job.
Without any clear idea of exactly how much difference there would have
to be to reassure you users, I cannot tell whether the rewrite would
accomplish that. The law is not any guidance here. Just because something
is legal doesn't mean you won't be scared--if you are scared now.
Writing display code that is significantly different is not easy.
I don't want to do that much work which is not really necessary,
just to reassure someone, when I can't be sure it will even do that.
The real problem is not a legal one; it is that some of you might be
intimidated. You are the only ones who stand to lose anything.
Perhaps some of you would like to write a replacement for the display
code? You are the best one to know what it takes to reassure you.
There is another point in favor of this. It is that the people who
think a certain piece of work is necessary are the ones who ought to
get it done. Remember, I have written GNU Emacs as a volunteer,
trying to serve the public. Many people like it, but some don't. If
you aren't satisfied with it, you can't blame me for that. I did a lot
for you just taking it as far as I have. If you want something that
I have not done, you should take some responsibility and do it.
Zimmerman, are you interested?
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: honor
Message-ID: <729@masscomp.UUCP>
Date: Mon, 24-Jun-85 10:14:51 EDT
Article-I.D.: masscomp.729
Posted: Mon Jun 24 10:14:51 1985
Date-Received: Tue, 25-Jun-85 03:02:09 EDT
References: <4489@mit-eddie.UUCP>
Organization: Masscomp - Westford, MA
Lines: 134
> From: Richard M. Stallman
> I can't understand how honor could require me to replace any
> of the code I am distributing in GNU Emacs.
>
> Using the modified Gosling display code in GNU Emacs is
> not immoral. Redistributing any code is always morally good.
Clearly, you and I have differing conceptions of "honor" and what is
"morally good". I seriously doubt that either of us is going to convince
the other, though, so let's leave it at that and not burn up any more
phone bills over the issue.
> It's not disobeying the spirit of the law. (We know
> we really do have permission.)
>
> It's not disobeying the letter of the law. (That is, we expect
> we could win if we were sued. That's the most anyone can ever say.)
I can only say that it appears that you have not received any competent
legal advice. First of all, your record is pretty poor, since Mark
Linton (the author of dbx) has publicly denied statements you attributed
to him, and James Gosling has likewise publicly denied that he took the
actions that you say he did. However, let's even assume for the moment
that your account is correct, and that Gosling sent some message to Fen
Labalme that he later forgot, saying something like the following.
You have my permission to make copies of Emacs #85 for whomever
you want, whenever you want, without limitation. Furthermore,
this permission can never be revoked.
Even such a statement from Gosling to Labalme would not help your case
at all. For although it would give Labalme permission to make copies,
it does not transfer that permission to anyone else. The presence of
the copyright notice at the top of each module reserves the right of
copy to the author and to those to whom he has explicitly granted
permission. The fact the a publisher may have permission to make copies
of a book does not mean that people who buy those copies or receive them
for free have any right to copy them. This would be the same type of
situation. In order for you to have the right to distribute GNU Emacs
in the manner in which you have, you would have to have had something
like the following attached to the previous statement:
Furthermore, you may grant permission to copy these files to
anyone else, and they may in turn grant such permission to
anyone else, and so on, ad infinitum.
This would be tantamount to removing the copyright notices (a recourse
available only to the author), which would be a much simpler procedure.
Such a statemtent would be completely contrary to all of Gosling's
public statements, as well as his public actions; he has taken great
care throughout the years to keep his Emacs from falling into the public
domain.
But you keep mentioning this message to Labalme without producing it;
once again, why don't you post this message to the net so that we can
all see what permission was or was not granted to Labalme? (Of course,
verifying the authenticity of the message would be another interesting
task, and should not be taken for granted.
> It's not breaking any promise Fen or I made.
This is irrelevant. Copyright law has nothing to do with people making
promises.
> Here is why I don't want to rewrite the display code:
>
> Any display code would have a considerable resemblance to that display
> code, just by virtue of doing the same job.
>
> Without any clear idea of exactly how much difference there would have
> to be to reassure you users, I cannot tell whether the rewrite would
> accomplish that. The law is not any guidance here. Just because something
> is legal doesn't mean you won't be scared--if you are scared now.
>
> Writing display code that is significantly different is not easy.
> I don't want to do that much work which is not really necessary,
> just to reassure someone, when I can't be sure it will even do that.
This is all nonsense, and I speak as someone who has had a fair amount
of experience in this area. When CCA decided to rewrite Montgomery's
display code in CCA EMACS, I was working under the restriction that none
of the same algorithms could even be used. Montgomery's code was quite
reasonable, but there were some fundamental limitations in his
algorithms. I designed a new display manager from the ground up that
bore no resemblance to his, and that had none of the limitations that
his did. (I understand that Montgomery later did a major revision of
his display code himself, so his current version went off in yet another
direction.) My display manager now looks no more like Montgomery's than
EMACS looks like vi. (Both edit text, remember.) The whole project
took me three weeks, and that was while I also had responsibility for
maintaining CCA's two VAXes.
You, on the other hand, would be working under less stringent
restrictions. Since copyright law doesn't protect algorithms, you could
even use Gosling's algorithms if you really wanted, as long as the new
code wasn't derived from the old. Of course, the difference would
probably be more clearcut if you used different algorithms. The display
code in Emacs #85 is certainly not the ultimate in a display manager;
one of its features was that certain changes in the display caused the
CPU time required to update the display to grow quadratically with the
number of lines affected. I'm sure you could do better than that with
different algorithms; several other people have.
> The real problem is not a legal one; it is that some of you might be
> intimidated.
There is a very real legal problem here. Unless both James Gosling and
Mark Linton are lying, and unless you can produce and authenticate a
message to Labalme from Gosling effecting renouncing Gosling's
copyright, then you have violated the law with respect to both Emacs and
dbx and have encouraged others to do the same.
> You are the only ones who stand to lose anything.
This is quite true. It would be hard for anyone to win much of a
judgement against you personally, since I assume you are not rich.
However, a real liability is created for others on this net and for
their companies. Lotus and others have established by precedent that
companies are liable for illegal duplication of software by their
employees even if the companies were not directly aware of their
employees' activities. Now, Unipress has indicated that it is not
inclined to sue at this point. However, If AT&T sees parts of UNIX
gradually being put into the public domain, it may feel forced to move
its legions of attorneys into action. To dismiss this possibility out
of hand would be foolish.
> Perhaps some of you would like to write a replacement for the display
> code? Zimmerman, are you interested?
No, thanks. I'm quite happy with my own EMACS. GNU Emacs is your
project.
Steve Zimmerman
Masscomp