Permission
rms@mit-prep
Fen Labalme got permission from Gosling in a message several years ago.
I'm told that this is legally binding on Gosling.
This was before Unipress was involved in distributing Gosling Emacs
and therefore Unipress has no direct knowledge about the question.
Megatest uses that old Gosling Emacs (what Fen gave me originally) in
its IC test systems. Fen says that Megatest's lawyers tell him they
can defend his permission in court if it were necessary.
However, we don't expect it to be necessary. We don't think Unipress
wants to sue anyone. It would be a shame if people refrain from using
GNU Emacs because of a hypothetical suit that would not win and that
won't happen.
It's a shame that it is so hard to PROVE that someone won't
sue someone else. However, we are working on it.
James Gosling
out of hand for me to remain silent. In no way did I ever give Fen
Lebalm (and by dubious transitivity, Richard Stallman) permission to
distribute any part of the Emacs that I wrote. Version 85 was not in
the public domain - I was very careful to get everyone who recieved a
copy to agree to not redistribute it. Unfortunatly, two moves have
left my records in a shambles.
I am not against public domain software: I am very much in favor of
it. There is an awful lot of public domain software floating around on
the net that was written by me. Emacs is not a part of it.
I would applaud the GNU effort (Emacs has needed a real lisp in it for
a long time) if it hadn't started off with what is at best a dreadful
misunderstanding.
Properly maintaining a piece of software like Emacs over a long period
is hard to do on a volunteer basis. I did it with my version for three
years and nearly destroyed my chances to graduate. I spent a long time
trying to find a new maintainer (either commercial or public domain).
I found no one willing to maintain it in the public domain and provide
the quality and the commitment that I thought was necessary. I talked
with quite a few companies, and by far the best was Unipress. They may
prostitute their souls for a dollar, but they do a pretty good job of
looking after Emacs.
Juha I. Heinanen
software brought to my mind the days when I filled applications for
educational licenses for Unix Versions 6 and 7. I still clearly
remember that in order to get such a license I had to state in the
application that the machine will be used for teaching and research
purposes only and that the results of the research (including the
code) had to be made publicly available. The impression I got was
that development of commercial software and an educational license
don't mix and that if we were ever going to sell any of our code, the
very first line of it had to be written on a purely commercial
machine.
Have the things changed since those days? Is it now ok to charge
more than a handling fee or to sumbit the software to a commercial
distributor such as Unipress in order to add value to it? If not, I
would like to know how many purely commercial 4.x Unix systems CMU
had when James wrote his Emacs and which one he used; or maybe he
really did all the development in a company down the street? What
about Amsterdam Compiler Kit, another product of Unipress whose early
versions were distributed for a nominal cost by EUUG?
I have nothing againsts James or the folks at Amsterdam and I very
well understand their motives in submitting the whole thing to a
company that frees them from the hassle and maybe even improves the
product. But IF the software was developed on a machine with an
educational Unix license, can the company be nothing else than a
distributor of public domain software? (Henry, are you listening
this group?)
--
Juha Heinanen
USL, P.O. Box 44330, Lafayette, LA 70504-433, tel. (318)231-5345
UUCP: {ut-sally, akgua}!usl!jih ARPA: usl!jih@ut-sally
Juha I. Heinanen
educational Unix license for commercial purposes. Below are some
relevant quotes from the terms for System V:
1.01 (a) Uses for "academic and educational purposes" means uses
directly related to teaching ... and uses in noncommercial research
by students and faculty, ..., provided that (i) neither the results
of such research nor any enhancement or modification so developed is
intended primarily for the benefit of a third party, (ii) such
results, enhancements and modifications (...) are made available to
anyone (...) without restriction on use, copying, or further
distribution, notwithstanding any proprietary right (such as a
copyright or patent right) that could be asserted by LICENSEE, its
students or faculty members and (iii) any copy of such result,
enhancement or modification furnished by LICENSEE is furnished for no
more than the cost of reproduction and shipping.
(b) Commercial use by LICENSEE of the LICENSED SOFTWARE or of any
such result, enhancement or modification is not permitted under this
agreement. Such commercial use is permissible only pursuant to the
terms of an appropriate commercial software agreement between AT&T
and LICENSEE. Any such result, enhancement or modification may be
developed further by LICENSEE for commercial use (...) only on CPUs
covered by such commercial software agreement and only provided that
the result, enhancement or modification, in the form in which it
exists when such commercial software agreement is executed by
LICENSEE, remains available to anyone as specified in Section 1.01
(a).
So, the question remains, did the machine James (I am using him here
just as an example) developed his Emacs have a commercial Unix
license. If it didn't, James' copyright notice doesn't prevent
anybody (including Richard) (re)distributing James' code in the
form it was when it left the educational machine. The same, of
course, applies to any copyrighted code floating around in various
universities. This is provided that the earlier licenses said
roughly the same thing.
Steve Zimmerman
> just as an example) developed his Emacs have a commercial Unix
> license. If it didn't, James' copyright notice doesn't prevent
> anybody (including Richard) (re)distributing James' code in the
> form it was when it left the educational machine. The same, of
> course, applies to any copyrighted code floating around in various
> universities. This is provided that the earlier licenses said
> roughly the same thing.
> --
> Juha Heinanen
Not true. There are several different issues here. Let's assume for
the moment that Gosling did develop his Emacs on a machine with an
educational license. (James, is this true?) First of all, according to
the copyright law, he has an automatic copyright on anything he
develops, assuming it wasn't done as work for hire, which it apparently
wasn't. As part of that copyright, he has the right to put copyright
notices on all his code, which he promptly did. Neither of these steps
are in violation of the System V license; Berkeley did essentially the
same thing when it put all of its "Copyright 198? Regents of the
University of California" legends on much of its software.
Now, if Gosling developed his software on a machine licensed under the
terms you quoted, the University would be required to share Gosling's
code free of charge with others. Whether or not it could compel Gosling
to do so would depend on what agreements Gosling had made with the
University. In any case, Gosling's refusal to give unlimited
distribution to his code would be at most a violation of a civil
contract, and would therefore be a matter between him, the University,
and AT&T. It would not affect the copyright protection granted to him
by federal law. That is, other people might try to convince AT&T to
force CMU and/or Gosling to distribute his code, but if AT&T didn't want
to do that, Gosling's copyright protection would stand. For example,
AT&T might decide that since the educational license terms had been
violated, it would force CMU to upgrade the license to a commercial one,
and not bother Gosling. Regardless of what course of action AT&T took,
someone who copied Gosling's code without his permission would be in
violation of federal copyright law, even if Gosling and/or CMU were
found to be in violation of the AT&T license.
This is all from a legal standpoint, of course; I have not addressed the
moral question here. Unfortunately, moral arguments usually don't get
you very far in court if you get sued.
My background in all of this is that during the development of my EMACS,
I had occasion to study the entire copyright law quite carefully, and I
also at one point retained an attorney whose specialty was copyright and
trade secret protection, and who had represented clients such as Data
General. However, I am not a lawyer myself, and therefore cannot
present any of this as definitive legal advice. For that, people must
go to their own lawyers.
On a related note, I recently received some mail from someone who did go
to the trouble of checking with his company's lawyer about GNU Emacs. I
thought that people reading this newsgroup would be interested in what the
lawyer said. The person sending me the message indicated that he didn't
want to get in the middle of this whole thing, so I've changed the names
to protect the innocent. Here's the relevant part of his message:
I actually did talk to our lawyer in charge of licensing and
associated software stuff about GNU, described the situation and
asked him what his opinion was. That was that we probably didn't
want to touch it with a 10 foot pole. Unless RMS can come up
with something that is legally binding (and electronic mail,
which is easily forgeable, doesn't count) he won't let us add
GNU to our [line of software] unless we can get a release from
Gosling allowing us to. He doesn't buy the third hand
redistribution, especially in light of [Gosling's] statements,
and doesn't want us to expose ourselves to liability by
distributing further. He more or less thinks that RMS is all
wet.
Phil Ngai
> Here's the relevant part of his message:
>
> I actually did talk to our lawyer in charge of licensing and
> associated software stuff about GNU, described the situation and
> asked him what his opinion was. That was that we probably didn't
> want to touch it with a 10 foot pole. Unless RMS can come up
> with something that is legally binding (and electronic mail,
> which is easily forgeable, doesn't count)
I got Gosling's permission to use his emacs but it was in the form
of electronic mail. I suspect many users are in the same situation.
I hope this doesn't mean I could get in trouble for using it on my
site because I don't have something legally binding.
--
Phil Ngai +1 408 749 5720
UUCP: {ucbvax,decwrl,ihnp4,allegra}!amdcad!phil
ARPA: amdcad!ph...@decwrl.ARPA
Chuq Von Rospach
>So, the question remains, did the machine James (I am using him here
>just as an example) developed his Emacs have a commercial Unix
>license. If it didn't, James' copyright notice doesn't prevent
>anybody (including Richard) (re)distributing James' code in the
>form it was when it left the educational machine. The same, of
>course, applies to any copyrighted code floating around in various
>universities.
Wrong. The copyright notices are correct regardless of whether or not Jim
broke AT&T's licensing. Copyright law is a different field from contract
law (or trade secret, for that matter). IF Gosling broke the license, it is
up to AT&T to do something about it; nobody else can because the
transgression was against them. In any case, the copyright would stand.
From what I can see, AT&T would have the following three options:
o cancel CMU's licenses.
o force CMU to upgrade the license to commercial and pay whatever fees
are required to do so.
o require Gosling to assign the copyright to them (he still has the
copyright, but they control the distribution of the code). This would be
similar to the cases in publishing where an author sells a story to a
magazine and gives up specific rights.
In no case does the copyright become invalid. Don't make assumptions that
it will, or you could get yourself in trouble.
chuq
--
:From the misfiring synapses of: Chuq Von Rospach
{cbosgd,fortune,hplabs,ihnp4,seismo}!nsc!chuqui nsc!chu...@decwrl.ARPA
The offices were very nice, and the clients were only raping the land, and
then, of course, there was the money...