[Prev][Next][Index][Thread]
GPL Discussion Digest
For people uninterested in the GPL discussion, please hit your delete
key now. This is a digest of the messages received (and not passed
on) which concerned the GPL.
All followup discussion should go to oskit-chat (or -flames. :-)
@flux.cs.utah.edu.
Included are messages from:
* shap@eros.cis.upenn.edu
* rms@gnu.org (Richard Stallman)
* gback@cs.utah.edu (Godmar Back)
-Dave
>From owner-oskit-users@fast.cs.utah.edu Fri Jan 22 18:32:04 1999
From: shap@eros.cis.upenn.edu
Message-Id: <199901230223.VAA02358@snocrash.cis.upenn.edu>
To: oskit-users@cs.utah.edu
Subject: GPL, Necessity, and FOIA
In-reply-to: Your message of "Tue, 19 Jan 1999 11:00:35 +0100."
<199901191000.LAA10715@trillian.softwell.se>
Date: Fri, 22 Jan 1999 21:23:59 -0500
Sender: shap@eros.cis.upenn.edu
Some portions of the OsKit were derived from GPL'd code, and as such
Utah has no choice but to distribute them under GPL. In the past,
only this subset of OsKit has been GPL'd.
The issue of who owns what at the end of DARPA contracts is complex,
to say the least. This isn't the first time that large, DARPA-funded
works have been published under GPL. If you are really concerned
about it, don't hassle the Utah folks. Go find your congressman and
bitch about the way DARPA is mis-spending your tax dollars and
software paid for in public trust is thereby being made unusable by
those segments of the public who wish to use it in proprietary form.
shap
>From owner-oskit-users@fast.cs.utah.edu Fri Jan 22 19:32:04 1999
Message-Id: <199901230226.VAA02368@snocrash.cis.upenn.edu>
To: oskit-users@cs.utah.edu
Subject: Re: GPL Liscensing on New Release: What Gives?
In-reply-to: Your message of "Tue, 19 Jan 1999 11:28:45 +0100."
<19990119112845.A48946@bitbox.follo.net>
Date: Fri, 22 Jan 1999 21:26:55 -0500
Sender: shap@eros.cis.upenn.edu
> This is actually not correct. The FSF allow the copyright holder to
> disclaim or transfer copyright; the clue is that the FSF want to have
> a copyright on that piece of code.
>
> However, I think you may be able to disclaim copyright on the code,
> and then re-claim it, possibly with some modifications that force the
> right to be held. It seems it would involve a lot of hassle to get it
> right.
This is incorrect.
You may license your code to someone else but retain the copyright.
In this case you retain the option of future licensing under any terms
you wish.
You may *transfer* the rights to your code; this is what FSF requests.
You may place your code in the public domain, which disclaims all
copyright. Anyone can subsequently assert a copyright on that code,
but they cannot enforce it against people who work from the public
domain version.
shap
>From owner-oskit-users Sat Jan 23 01:40:13 1999
Date: Sat, 23 Jan 1999 01:40:01 -0700 (MST)
Message-Id: <199901230840.BAA15439@wijiji.santafe.edu>
From: Richard Stallman <rms@gnu.org>
to: Godmar Back <gback@cs.utah.edu>
To: Bengt Kleberg <bengt@softwell.se>
Cc: gback@fast.cs.utah.edu, neil@causality.com, oskit-users@cs.utah.edu
cc: lepreau@cs.utah.edu
In-reply-to: <Pine.LNX.3.96.990119235101.26482G-100000@geeky1.ebtech.net>
(message from Paul Anderson on Tue, 19 Jan 1999 23:52:38 -0500 (EST))
Subject: Re: GPL Liscensing on New Release: What Gives? (fwd)
Reply-to: rms@gnu.org
References: <Pine.LNX.3.96.990119235101.26482G-100000@geeky1.ebtech.net>
I support the decision that the OSK developers have made, to release
under the GPL and sell alternative licenses. It is not the best
possible approach, but it is an acceptable one.
It does have drawbacks. One is that it does not treat all users
equally. The FSF never sells alternative licenses, because our policy
is that we should always treat all users alike; whatever license we
offer to one, we offer to all. No one gets special privilege, not at
any price.
Another drawback is that it allows a way for the library to be used in
proprietary software. We can support the free software community more
by telling proprietary software developers, "If you'd like to use this
library in your program, make your program free software!"
So the approach of using the GPL and selling alternative licenses is
not the best. But it does make the library available as free software
for free software projects, and that is the most important thing. It
is much, much better than releasing the library under a less-than-free
license. And it can be better than releasing the library under a lax
non-copyleft license. I've suggested this approach from time to time,
to developers who reject the alternative of using the GPL and only the
GPL.
But while I support this decision, I object to your use of the word
"contamination" to describe the GPL. You haven't stated a
justification for this accusation, so there is nothing I can refute,
but surely you don't want to engage in mere name-calling. If you want
to say such harsh things about someone's work, you ought to present
justification for them.
In the past, I've found that these harsh views of the GPL were often
based on a misunderstanding of its consequences. If you tell me a
specific scenario in which you think that the GPL does something
wrong, we may find that the GPL doesn't actually do what you thought.
>From owner-oskit-users Sat Jan 23 13:38:03 1999
From: Godmar Back <gback@cs.utah.edu>
Message-Id: <199901232039.NAA15231@lal.cs.utah.edu>
Subject: Re: GPL Liscensing on New Release: What Gives? (fwd)
To: rms@gnu.org
Date: Sat, 23 Jan 1999 13:39:07 -0700 (MST)
Cc: gback@cs.utah.edu, bengt@softwell.se, gback@fast.cs.utah.edu,
neil@causality.com, oskit-users@cs.utah.edu, lepreau@cs.utah.edu
In-Reply-To: <199901230840.BAA15439@wijiji.santafe.edu> from "Richard Stallman" at Jan 23, 99 01:40:01 am
X-Mailer: ELM [version 2.4 PL25]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Richard,
>
> But while I support this decision, I object to your use of the word
> "contamination" to describe the GPL. You haven't stated a
> justification for this accusation, so there is nothing I can refute,
> but surely you don't want to engage in mere name-calling. If you want
> to say such harsh things about someone's work, you ought to present
> justification for them.
>
> In the past, I've found that these harsh views of the GPL were often
> based on a misunderstanding of its consequences. If you tell me a
> specific scenario in which you think that the GPL does something
> wrong, we may find that the GPL doesn't actually do what you thought.
>
Whether or not the use of the word "contamination" is an accusation very
much depends on your point of view. I am not a native speaker, but my
webster defines "contamination" as:
con.tam.i.na.tion \k*n-.tam-*-'na--sh*n\ n : a contaminating or state of
being contaminated; also : something that contaminates
while webster defines "contaminate" as:
con.tam.i.nate \k*n-'tam-*-.na-t\ \-.na-t-iv\ \-.na-t-*r\ vt [L
contaminatus, pp. of contaminare; akin to L contagio conta]gion 1: to soil,
stain, or infect by contact or association 2: to make unfit for use by
introduction of unwholesome or undesirable elementsmake impure or unclean.
CONTAMINATE implies intrusion of or contact with an outside source as the
cause; TAINT implies that corruption and decay have begun to take effect;
POLLUTE stresses the loss of purity and cleanness through contamination;
DEFILE implies befouling of what ought to be clean and pure and suggests
violation or desecration - con.tam.i.na.tive aj SYN syn CONTAMINATE, TAINT,
POLLUTE, DEFILE mean to
I am referring to the meaning "to make unfit for use by introduction of
unwholesome or undesirable elements."
It seems obvious to me that the use of GPL'ed code makes a product unfit
for a specific use, namely the use in proprietary products. It does so
by contact, as webster demands, namely the contact between proprietary
and GPL'ed code that is introduced by the act of linking them together.
While you may disagree and argue that preventing free code from being included
in proprietary software is a desirable goal that does not make it unfit for
general use, and that proprietary software in general is undesirable [1],
I feel very much justified in my using the word "contamination". You can
argue positions, but you cannot claim words.
- Godmar
[1] http://www.fsf.org/philosophy/shouldbefree.html
>From owner-oskit-users Sat Jan 23 23:34:06 1999
Date: Sun, 24 Jan 1999 01:39:03 -0500
Message-Id: <199901240639.BAA00292@psilocin.gnu.org>
From: Richard Stallman <rms@gnu.org>
To: gback@cs.utah.edu
Cc: gback@cs.utah.edu, bengt@softwell.se, gback@fast.cs.utah.edu,
neil@causality.com, oskit-users@cs.utah.edu, lepreau@cs.utah.edu
In-reply-to: <199901232039.NAA15231@lal.cs.utah.edu> (message from Godmar Back
on Sat, 23 Jan 1999 13:39:07 -0700 (MST))
Subject: Re: GPL Liscensing on New Release: What Gives? (fwd)
Reply-to: rms@gnu.org
References: <199901232039.NAA15231@lal.cs.utah.edu>
Whether or not the use of the word "contamination" is an accusation very
much depends on your point of view.
How about if you ask a few native speakers, "If your work were
described as a kind of `contamination', how would you feel?" I
suspect that almost everyone will tell you it would feel like an
insult.
Of course, the answers might not be what I expect; if so, I would
definitely benefit from learning I have misjudged how most people see
this word.
I am referring to the meaning "to make unfit for use by introduction of
unwholesome or undesirable elements."
That's a legitimate definition, but applying it to the GPL runs into
two problems: a questionable assumption about values, and a factual
misunderstanding.
It seems obvious to me that the use of GPL'ed code makes a product unfit
for a specific use, namely the use in proprietary products.
The GPL cannot make a program unfit for any sort of use, but it
thwarts proprietary software developers from doing something that they
want to do. This, however, is a good thing, because what proprietary
software developers want to do is deny freedom to other people, the
users. The GPL disappoints them by protecting the user's freedom.
I think that protecting freedom for all users is something wholesome
and desirable. If you are against this, and you are speaking in the
context of a discussion about free software, I suggest that you state
clearly where you are coming from.
It does so
by contact, as webster demands, namely the contact between proprietary
and GPL'ed code that is introduced by the act of linking them together.
This is the factual misunderstanding. The GPL does not work this way.
It does not *do* anything to proprietary code that you link with
free code. It just says you may not link them.
If you look more carefully at a specific fully-fleshed-out scenario,
and correctly understand the consequences of the GPL, I think you will
see that this description does not really fit.
=================================
To unsubscribe from this list, send a message with a body of "unsubscribe"
to oskit-users-request@flux.cs.utah.edu