This will hopefully help diminish certain myths about the code licensing.
C.
---------- Forwarded message ----------
From: Theo de Raadt <[email protected]>
Date: 31-Aug-2007 21:40
Subject: That whole "Linux stealing our code" thing
To: [email protected]
[bcc'd to Eben Moglen so that people don't flood him]
I stopped making public statements in the recent controversy because
Eben Moglen started working behind the scenes to 'improve' what Linux
people are doing wrong with licensing, and he asked me to give him
pause, so his team could work. Honestly, I was greatly troubled by
the situation, because even people like Alan Cox were giving other
Linux developers advice to ... break the law. And furthermore, there
are even greater potential risks for how the various communities
interact.
For the record -- I was right and the Linux developers cannot change
the licenses in any of those ways proposed in those diffs, or that
conversation (http://lkml.org/lkml/2007/8/28/157).
It is illegal to modify a license unless you are the owner/author,
because it is a legal document. If there are multiple owners/authors,
they must all agree. A person who receives the file under two
licenses can use the file in either way.... but if they distribute
the file (modified or unmodified!), they must distribute it with the
existing license intact, because the licenses we all use have
statements which say that the license may not be removed.
It may seem that the licenses let one _distribute_ it under either
license, but this interpretation of the license is false -- it is
still illegal to break up, cut up, or modify someone else's legal
document, and, it cannot be replaced by another license because it may
not be removed. Hence, a dual licensed file always remains dual
licensed, every time it is distributed.
Now I've been nice enough to give Eben and his team a few days time to
communicate inside the Linux community, to convince them that what
they have proposed/discussed is wrong at a legal level. I think that
Eben also agrees with me that there are grave concerns about how this
leads to problems at the ethical and community levels (at some level,
a ethos is needed for Linux developers to work with *BSD developers).
And there are possibilities that similar issues could loom in the
larger open source communities who are writing applications.
Eben has thus far chosen not to make a public statement, but since
time is running out on people's memory, I am making one. Also, I feel
that a lot of Linux "relicencing" meme-talkin' trolls basically have
attacked me very unfairly again, so I am not going to wait for Eben to
say something public about this.
In http://lkml.org/lkml/2007/8/29/183, Alan Cox managed to summarize
what Jiri Slaby and Luis Rodriguez were trying to do by proposing a
modification of a Dual Licenced file without the consent of all the
authors. Alan asks "So whats the problem ?". Well, Alan, I must
caution you -- your post is advising people to break the law.
I will attempt to describe in simple terms, based on what I have been
taught, how one must handle such licenses:
- If you receive dual licensed code, you may not delete the license
you don't like and then distribute it. It has to stay, because you
may not edit someone's else's license -- which is a three-part legal
document (For instance: Copyright notice, BSD, followed by GPL).
- If you receive ISC or BSD licensed code, you may not delete the
license. Same principle, since the notice says so. It's the law.
Really.
- If you add "large pieces of originality" to the code which are valid
for copyright protection on their own, you may choose to put a different
and seperate (must be non-conflicting...) license at the top of the file
above the existing license.
(Warning: things become less clear as to what the combination of
licenses mean, though -- there are ethical traps, too).
- If you wish for everyone to remain friends, you should give code back.
That means (at some ethical or friendliness level) you probably do
not want to put a GPL at the top of a BSD or ISC file, because you
would be telling the people who wrote the BSD or ISC file:
"Thanks for what you wrote, but this is a one-way street, you give
us code, and we take it, we give you you nothing back. screw off."
In either case, I think a valuable lessons has been taught us here in
the BSD world -- there are many many GPL loving people who are going
to try to find any way to not give back and share (I will mention one
name: Luis Rodriguez has been a fanatic pushing us for dual licensed,
and I feel he is to blame for this particular problem). Many of those
same people have been saying for years that BSD code can be stolen,
and that is why people should GPL their code.
Well, the lesson they have really taught us is that they consider the
GPL their best tool to take from us!
GPL fans said the great problem we would face is that companies would
take our BSD code, modify it, and not give back. Nope -- the great
problem we face is that people would wrap the GPL around our code, and
lock us out in the same way that these supposed companies would lock
us out. Just like the Linux community, we have many companies giving
us code back, all the time. But once the code is GPL'd, we cannot get
it back.
Ironic.
I hope some people in the GPL community will give that some thought.
Your license may benefit you, but you could lose friends you need.
The GPL users have an opportunity to 'develop community', to keep an
ethic of sharing alive.
If the Linux developers wrap GPL's around things we worked very hard
on, it will definately not be viewed as community development.
Thank you for thinking about this.
[I ask that one person make sure that one copy of this ends up on the
linux kernel mailing list]
Constantine A. Murenin wrote:
> This will hopefully help diminish certain myths about the code licensing.
What myth? The myth that Theo understands dual licensing?
Jeff
On 01/09/07, Jeff Garzik <[email protected]> wrote:
> Constantine A. Murenin wrote:
> > This will hopefully help diminish certain myths about the code licensing.
>
> What myth? The myth that Theo understands dual licensing?
Reyk's code was never dual licensed, so it's not like it even matters
to the original dispute.
That said, I don't see what exact wording you consider inaccurate.
C.
> It is illegal to modify a license unless you are the owner/author,
> because it is a legal document. If there are multiple owners/authors,
Oh dear - Theo, go talk to a lawyer, or do a course on licencing.
The owner generally starts with the rights to control who performs acts
covered by copyright law. They pass some of those rights on to others by
contract, licence or statutory means. It is quite normal for the owner to
pass on the right to relicence or modify the licencing of a work. In many
cases the owner actually hands on all such rights to a third party
(eg an evil music company).
[Owner and author often differ as many legal systems start from the basis
that an employee produces a work for the employer rather than it being
transferred solely by contract]
The ath5k C file in question (not the headers) seems to give recipients
permission to further convey the work under a choice of two licences. It
doesn't say they must redistribute under both. So I appear to have a
right to convey the work under the GPL to a third party, who from me
receives no right to use it except under the GPL.
The Ath5K C code is very clear about the intention of the licencing:
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
The choice appears to be delegated to the recipient very clearly and
very specifically by the licencing on the file. It does not say that I
must convey the work under both licences. It quite specifically says I may
convey the work under whichever of the two I prefer (and probably both if
I wish). Clearly if that had not been the intent it would not have
included the clause giving the choice.
This is quite different to the case Theo tries to discuss.
> In http://lkml.org/lkml/2007/8/29/183, Alan Cox managed to summarize
> what Jiri Slaby and Luis Rodriguez were trying to do by proposing a
> modification of a Dual Licenced file without the consent of all the
> authors. Alan asks "So whats the problem ?". Well, Alan, I must
> caution you -- your post is advising people to break the law.
Re-read my email and then apologize. I do question the .h files where
they are BSD licence and no changes were made to the work. I also point
out that the dual licence on that code appears to give permission to
distribute under one of those licences by choice.
> - If you receive dual licensed code, you may not delete the license
> you don't like and then distribute it. It has to stay, because you
> may not edit someone's else's license -- which is a three-part legal
> document (For instance: Copyright notice, BSD, followed by GPL).
If you got BSD licenced code that doesn't give you a choice of licence
then of course the original work remains BSD and you can't go around
removing that information, the copyright holder's name and other things
protected variously by different legal systems. I would submit the ath5k
header files fit this (if they are even copyrightable works at all)
> - If you wish for everyone to remain friends, you should give code back.
That's about the first thing I would agree on - its somewhat rude and
not something I personally woul usually choose todo. However to many
there are problems as the BSD licence doesn't mean giving it back to the
community it means giving a copy to everyone who wants rip it off for
private proprietary use.
> "Thanks for what you wrote, but this is a one-way street, you give
> us code, and we take it, we give you you nothing back. screw off."
The BSD licence allows this. If you in the BSD world don't want that to
happen you need to look hard at your licencing. Linux takes very little
from the BSD world this way, the big one way takers are all proprietary
software companies. Perhaps Theo should write to that nice Bill guy
instead.
> GPL fans said the great problem we would face is that companies would
> take our BSD code, modify it, and not give back. Nope -- the great
> problem we face is that people would wrap the GPL around our code, and
You just don't realise who takes your code and what they do with it. The
proprietary people don't tell you, but the free ones you can see.
> If the Linux developers wrap GPL's around things we worked very hard
> on, it will definately not be viewed as community development.
So you'd prefer that the Linux developers worked on it and then Microsoft
took the results of all our work and didn't give anything back. At least
if the Linux work is GPL licenced its protected from further abuse. See
the viewpoint the free software people come from - you may not agree with
it but it has a logic.
If OpenBSD wants a world where code must be returned, but you can mix it
with free code in a product in some fashion and do binary only releases
then OpenBSD needs to fix its licencing. Not to GPL which is clearly not
the BSD intention but to something which does what BSD wants rather than
an academic research licence developed thirty odd years ago for the
purpose of showing that US research funds were properly spent. Perhaps
its time for BSD2 licencing ?
Alan
On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > Constantine A. Murenin wrote:
> > > This will hopefully help diminish certain myths about the code licensing.
> >
> > What myth? The myth that Theo understands dual licensing?
>
> Reyk's code was never dual licensed, so it's not like it even matters
> to the original dispute.
It's no longer dual licenced in the FreeBSD tree because the FreeBSD
people removed the GPL choice of the dual licenced code 3 months ago.
So all of Theo's accusations of people breaking the law by making this
dual licenced code GPL-only apply as well to the FreeBSD people...
> That said, I don't see what exact wording you consider inaccurate.
Both the FreeBSD and Linux people draw the logical conclusion that this
"Alternatively" means everyone can always choose to remove one of the
two choices alternatively offered.
According to Theo, that is "breaking the law"...
> C.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Sat, Sep 01, 2007 at 09:30:52PM +0100, Alan Cox wrote:
> If OpenBSD wants a world where code must be returned
OpenBSD does not want this.
OpenBSD wants a world where people do things because they are the
right thing to do.
OpenBSD lets you decide; it doesn't dictate.
someone poo-poos your decision, well, it was your decision.
someone poo-poos you because you do exactly the thing that you are so
afraid of having happen to you, well, it was your decision.
--
[email protected]
SDF Public Access UNIX System - http://sdf.lonestar.org
On Sat, Sep 01, 2007 at 10:54:57PM +0200, Adrian Bunk wrote:
> On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > Constantine A. Murenin wrote:
> > > > This will hopefully help diminish certain myths about the code licensing.
> > >
> > > What myth? The myth that Theo understands dual licensing?
> >
> > Reyk's code was never dual licensed, so it's not like it even matters
> > to the original dispute.
Oh, and if you look at the OpenBSD CVS you see versions 4 months old
with dozens of contributions by Reyk and with:
/* $OpenBSD: ath.c,v 1.63 2007/05/09 16:41:14 reyk Exp $ */
/* $NetBSD: ath.c,v 1.37 2004/08/18 21:59:39 dyoung Exp $ */
/*-
* Copyright (c) 2002-2004 Sam Leffler, Errno Consulting
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer,
* without modification.
* 2. Redistributions in binary form must reproduce at minimum a disclaimer
* similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any
* redistribution must be conditioned upon including a substantially
* similar Disclaimer requirement for further binary redistribution.
* 3. Neither the names of the above-listed copyright holders nor the names
* of any contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* NO WARRANTY
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY
* AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY,
* OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
* IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
* THE POSSIBILITY OF SUCH DAMAGES.
*/
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On 01/09/07, Adrian Bunk <[email protected]> wrote:
> On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > Constantine A. Murenin wrote:
> > > > This will hopefully help diminish certain myths about the code licensing.
> > >
> > > What myth? The myth that Theo understands dual licensing?
> >
> > Reyk's code was never dual licensed, so it's not like it even matters
> > to the original dispute.
>
> It's no longer dual licenced in the FreeBSD tree because the FreeBSD
> people removed the GPL choice of the dual licenced code 3 months ago.
FreeBSD doesn't have Reyk's ath(4) HAL, which OpenHAL is based on.
FreeBSD has a driver written by Sam, and a binary-only HAL, also written by Sam.
> So all of Theo's accusations of people breaking the law by making this
> dual licenced code GPL-only apply as well to the FreeBSD people...
How? FreeBSD doesn't have Reyk's ath(4) HAL from OpenBSD, so there are
no possible licensing accusations and violations.
> > That said, I don't see what exact wording you consider inaccurate.
>
> Both the FreeBSD and Linux people draw the logical conclusion that this
> "Alternatively" means everyone can always choose to remove one of the
> two choices alternatively offered.
>
> According to Theo, that is "breaking the law"...
FreeBSD's ath(4) code, both the driver and the HAL, is entirely
written by Sam Leffler, who can licence it in whichever way he seems
reasonable. The driver part of Sam's code is also present in OpenBSD,
but the HALs in OpenBSD and FreeBSD are entirely different.
C.
On 01/09/07, Adrian Bunk <[email protected]> wrote:
> On Sat, Sep 01, 2007 at 10:54:57PM +0200, Adrian Bunk wrote:
> > On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > > Constantine A. Murenin wrote:
> > > > > This will hopefully help diminish certain myths about the code licensing.
> > > >
> > > > What myth? The myth that Theo understands dual licensing?
> > >
> > > Reyk's code was never dual licensed, so it's not like it even matters
> > > to the original dispute.
>
> Oh, and if you look at the OpenBSD CVS you see versions 4 months old
> with dozens of contributions by Reyk and with:
>
> /* $OpenBSD: ath.c,v 1.63 2007/05/09 16:41:14 reyk Exp $ */
> /* $NetBSD: ath.c,v 1.37 2004/08/18 21:59:39 dyoung Exp $ */
>
> /*-
> * Copyright (c) 2002-2004 Sam Leffler, Errno Consulting
> * All rights reserved.
> *
> * Redistribution and use in source and binary forms, with or without
> * modification, are permitted provided that the following conditions
> * are met:
> * 1. Redistributions of source code must retain the above copyright
> * notice, this list of conditions and the following disclaimer,
> * without modification.
> * 2. Redistributions in binary form must reproduce at minimum a disclaimer
> * similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any
> * redistribution must be conditioned upon including a substantially
> * similar Disclaimer requirement for further binary redistribution.
> * 3. Neither the names of the above-listed copyright holders nor the names
> * of any contributors may be used to endorse or promote products derived
> * from this software without specific prior written permission.
> *
> * Alternatively, this software may be distributed under the terms of the
> * GNU General Public License ("GPL") version 2 as published by the Free
> * Software Foundation.
> *
> * NO WARRANTY
> * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
> * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
> * LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY
> * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
> * THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY,
> * OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
> * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
> * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
> * IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
> * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
> * THE POSSIBILITY OF SUCH DAMAGES.
> */
Where exactly do you see Reyk's copyright in the above quote?
http://www.openbsd.org/cgi-bin/man.cgi?query=ath&sektion=4#AUTHORS
C.
On Sat, Sep 01, 2007 at 05:27:03PM -0400, Constantine A. Murenin wrote:
> On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > > Constantine A. Murenin wrote:
> > > > > This will hopefully help diminish certain myths about the code licensing.
> > > >
> > > > What myth? The myth that Theo understands dual licensing?
> > >
> > > Reyk's code was never dual licensed, so it's not like it even matters
> > > to the original dispute.
> >
> > It's no longer dual licenced in the FreeBSD tree because the FreeBSD
> > people removed the GPL choice of the dual licenced code 3 months ago.
>
> FreeBSD doesn't have Reyk's ath(4) HAL, which OpenHAL is based on.
>
> FreeBSD has a driver written by Sam, and a binary-only HAL, also written by Sam.
>
> > So all of Theo's accusations of people breaking the law by making this
> > dual licenced code GPL-only apply as well to the FreeBSD people...
>
> How? FreeBSD doesn't have Reyk's ath(4) HAL from OpenBSD, so there are
> no possible licensing accusations and violations.
OK, I begin to understand this, there seem to be three different types
of files changed by Jiri's patch:
1. dual licenced files planned to make GPL-only
2. previously dual licenced files with a too recent version used planned
to make GPL-only
3. never dual licenced files planned to make GPL-only
For files under 1. and 2. Reyk did contribute to dual licenced code
without touching the licence, but I missed that there's also code unter 3.
So there is a problem, but not with the code under 1. (unless you plan
to change the semantics of the word "alternatively"), the problem is
with some headers under 2. plus the code under 3.
It's funny how Theo missed the part of Jiri's patch that actually is a
copyright violation and instead complains about the part that is OK...
> > > That said, I don't see what exact wording you consider inaccurate.
> >
> > Both the FreeBSD and Linux people draw the logical conclusion that this
> > "Alternatively" means everyone can always choose to remove one of the
> > two choices alternatively offered.
> >
> > According to Theo, that is "breaking the law"...
>
> FreeBSD's ath(4) code, both the driver and the HAL, is entirely
> written by Sam Leffler, who can licence it in whichever way he seems
> reasonable. The driver part of Sam's code is also present in OpenBSD,
> but the HALs in OpenBSD and FreeBSD are entirely different.
Sam also changed the licence of a file additionally containing an
Copyright (c) 2004 Video54 Technologies, Inc.
> C.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
Adrian Bunk wrote:
> On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
>
>> On 01/09/07, Jeff Garzik <[email protected]> wrote:
>>
>>> Constantine A. Murenin wrote:
>>>
>>>> This will hopefully help diminish certain myths about the code licensing.
>>>>
>>> What myth? The myth that Theo understands dual licensing?
>>>
>> Reyk's code was never dual licensed, so it's not like it even matters
>> to the original dispute.
>>
>
> It's no longer dual licenced in the FreeBSD tree because the FreeBSD
> people removed the GPL choice of the dual licenced code 3 months ago.
>
> So all of Theo's accusations of people breaking the law by making this
> dual licenced code GPL-only apply as well to the FreeBSD people...
>
Sigh, why actually check the facts when you can make them up. The code
in question is my code. It has my copyright (modulo bits shared with
onoe-san who was consulted on the switch from dual-bsd/gpl to bsd only
in freebsd). Of course what was amusing was how after I changed the
license on the current code in freebsd certain folks retroactively
applied the license changes to code that was 3 years old.
But is there a point to all this nonsense? I dual-licensed the code so
folks could adopt and use it however they saw fit. As I've said before
I don't care what people do with the work I give away so long as they
don't claim it's their own.
>
>> That said, I don't see what exact wording you consider inaccurate.
>>
>
> Both the FreeBSD and Linux people draw the logical conclusion that this
> "Alternatively" means everyone can always choose to remove one of the
> two choices alternatively offered.
>
> According to Theo, that is "breaking the law"...
>
>
I've yet to see "FreeBSD people" speak up so again you're just spouting
jibberish. I am speaking up as the author of the code that set the dual
license in place. I have the definitive say and I have said that any of
my code that is dual-licensed can be made gpl only.
Sam
On Sat, Sep 01, 2007 at 05:51:49PM -0400, Constantine A. Murenin wrote:
> On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > On Sat, Sep 01, 2007 at 10:54:57PM +0200, Adrian Bunk wrote:
> > > On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > > > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > > > Constantine A. Murenin wrote:
> > > > > > This will hopefully help diminish certain myths about the code licensing.
> > > > >
> > > > > What myth? The myth that Theo understands dual licensing?
> > > >
> > > > Reyk's code was never dual licensed, so it's not like it even matters
> > > > to the original dispute.
> >
> > Oh, and if you look at the OpenBSD CVS you see versions 4 months old
> > with dozens of contributions by Reyk and with:
> >
> > /* $OpenBSD: ath.c,v 1.63 2007/05/09 16:41:14 reyk Exp $ */
> > /* $NetBSD: ath.c,v 1.37 2004/08/18 21:59:39 dyoung Exp $ */
> >
> > /*-
> > * Copyright (c) 2002-2004 Sam Leffler, Errno Consulting
> > * All rights reserved.
> > *
> > * Redistribution and use in source and binary forms, with or without
> > * modification, are permitted provided that the following conditions
> > * are met:
> > * 1. Redistributions of source code must retain the above copyright
> > * notice, this list of conditions and the following disclaimer,
> > * without modification.
> > * 2. Redistributions in binary form must reproduce at minimum a disclaimer
> > * similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any
> > * redistribution must be conditioned upon including a substantially
> > * similar Disclaimer requirement for further binary redistribution.
> > * 3. Neither the names of the above-listed copyright holders nor the names
> > * of any contributors may be used to endorse or promote products derived
> > * from this software without specific prior written permission.
> > *
> > * Alternatively, this software may be distributed under the terms of the
> > * GNU General Public License ("GPL") version 2 as published by the Free
> > * Software Foundation.
> > *
> > * NO WARRANTY
> > * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
> > * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
> > * LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY
> > * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
> > * THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY,
> > * OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
> > * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
> > * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
> > * IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
> > * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
> > * THE POSSIBILITY OF SUCH DAMAGES.
> > */
>
> Where exactly do you see Reyk's copyright in the above quote?
>...
He has automatically a copyright on his contributions if they are
non-trivial.
> C.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Sat, Sep 01, 2007 at 03:03:36PM -0700, Sam Leffler wrote:
> Adrian Bunk wrote:
>> On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
>>
>>> On 01/09/07, Jeff Garzik <[email protected]> wrote:
>>>
>>>> Constantine A. Murenin wrote:
>>>>
>>>>> This will hopefully help diminish certain myths about the code
>>>>> licensing.
>>>>>
>>>> What myth? The myth that Theo understands dual licensing?
>>>>
>>> Reyk's code was never dual licensed, so it's not like it even matters
>>> to the original dispute.
>>>
>>
>> It's no longer dual licenced in the FreeBSD tree because the FreeBSD
>> people removed the GPL choice of the dual licenced code 3 months ago.
>>
>> So all of Theo's accusations of people breaking the law by making this
>> dual licenced code GPL-only apply as well to the FreeBSD people...
>>
>
> Sigh, why actually check the facts when you can make them up. The code in
> question is my code. It has my copyright (modulo bits shared with onoe-san
> who was consulted on the switch from dual-bsd/gpl to bsd only in freebsd).
The latter is the code by Video54 Technologies?
> Of course what was amusing was how after I changed the license on the
> current code in freebsd certain folks retroactively applied the license
> changes to code that was 3 years old.
>
> But is there a point to all this nonsense? I dual-licensed the code so
> folks could adopt and use it however they saw fit. As I've said before I
> don't care what people do with the work I give away so long as they don't
> claim it's their own.
Fully agreed. :-)
>>> That said, I don't see what exact wording you consider inaccurate.
>>>
>>
>> Both the FreeBSD and Linux people draw the logical conclusion that this
>> "Alternatively" means everyone can always choose to remove one of the two
>> choices alternatively offered.
>>
>> According to Theo, that is "breaking the law"...
>
> I've yet to see "FreeBSD people" speak up so again you're just spouting
> jibberish. I am speaking up as the author of the code that set the dual
> license in place. I have the definitive say and I have said that any of my
> code that is dual-licensed can be made gpl only.
Sorry, this has been a thinko on my side:
If noone except you and onoe-san made any contributions to this code
that were non-trivial enough for automatically giving its author a
copyright on his contributions (whatever this means in various
jurisdictions...) it was indeed an author-only change.
> Sam
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On 01/09/07, Adrian Bunk <[email protected]> wrote:
> On Sat, Sep 01, 2007 at 05:27:03PM -0400, Constantine A. Murenin wrote:
> > On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > > On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > > > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > > > Constantine A. Murenin wrote:
> > > > > > This will hopefully help diminish certain myths about the code licensing.
> > > > >
> > > > > What myth? The myth that Theo understands dual licensing?
> > > >
> > > > Reyk's code was never dual licensed, so it's not like it even matters
> > > > to the original dispute.
> > >
> > > It's no longer dual licenced in the FreeBSD tree because the FreeBSD
> > > people removed the GPL choice of the dual licenced code 3 months ago.
> >
> > FreeBSD doesn't have Reyk's ath(4) HAL, which OpenHAL is based on.
> >
> > FreeBSD has a driver written by Sam, and a binary-only HAL, also written by Sam.
> >
> > > So all of Theo's accusations of people breaking the law by making this
> > > dual licenced code GPL-only apply as well to the FreeBSD people...
> >
> > How? FreeBSD doesn't have Reyk's ath(4) HAL from OpenBSD, so there are
> > no possible licensing accusations and violations.
>
> OK, I begin to understand this, there seem to be three different types
> of files changed by Jiri's patch:
> 1. dual licenced files planned to make GPL-only
> 2. previously dual licenced files with a too recent version used planned
> to make GPL-only
> 3. never dual licenced files planned to make GPL-only
>
> For files under 1. and 2. Reyk did contribute to dual licenced code
> without touching the licence, but I missed that there's also code unter 3.
>
> So there is a problem, but not with the code under 1. (unless you plan
> to change the semantics of the word "alternatively"), the problem is
> with some headers under 2. plus the code under 3.
>
> It's funny how Theo missed the part of Jiri's patch that actually is a
> copyright violation and instead complains about the part that is OK...
I'm not sure how you conclude that Theo missed the relevant parts --
there were many messages posted to [email protected] mailing list and
to The OpenBSD Journal in the last few days, and to me it appears as
all of the problems were discussed ad nauseam.
After the obvious copyright violations were addressed, I think the
problem started being an ethical one.
As a free software user and developer, the question I have is how come
the Linux community feels that they can take the BSD code that was
reverse-engineered at OpenBSD, and put a more restrictive licence onto
it, such that there will be no possibility of the changes going back
to OpenBSD, given that the main work on the code has happened at
OpenBSD? (Obviously, such a scenario it is permitted by the licence,
but my question is an ethical one -- after all, most components of
OpenHAL were specifically based on the OpenBSD's ath(4) HAL code.)
You can see that Christoph Hellwig agrees with this ethical problem,
as in the message below.
C.
On 28/08/07, Christoph Hellwig <[email protected]> wrote:
> On Tue, Aug 28, 2007 at 12:00:50PM -0400, Jiri Slaby wrote:
> > ath5k, license is GPLv2
> >
> > The files are available only under GPLv2 since now.
>
> Is this really a good idea? Most of the reverse-engineering was
> done by the OpenBSD folks, and it would certainly be helpful to
> work together with them on new hardware revisions, etc..
( from http://lkml.org/lkml/2007/8/28/178 )
I urge developers to not bait into this and just leave this alone.
Those involved know what they are doing and have a strong team of
attorneys watching their backs. Any *necessary* discussions are be
done privately.
Luis
>As a free software user and developer, the question I have is how come
>the Linux community feels that they can take the BSD code that was
>reverse-engineered at OpenBSD, and put a more restrictive licence onto
>it, such that there will be no possibility of the changes going back
>to OpenBSD, given that the main work on the code has happened at
>OpenBSD? (Obviously, such a scenario it is permitted by the licence,
>but my question is an ethical one -- after all, most components of
>OpenHAL were specifically based on the OpenBSD's ath(4) HAL code.)
>
>You can see that Christoph Hellwig agrees with this ethical problem,
>as in the message below.
>
>C.
>
>
>>On 28/08/07, Christoph Hellwig <[email protected]> wrote:
>> On Tue, Aug 28, 2007 at 12:00:50PM -0400, Jiri Slaby wrote:
>> > ath5k, license is GPLv2
>> >
>> > The files are available only under GPLv2 since now.
>>
>> Is this really a good idea? Most of the reverse-engineering was
>> done by the OpenBSD folks, and it would certainly be helpful to
>> work together with them on new hardware revisions, etc..
I couldn't agree more. The point is, while we BSD license fans know and
expect people from private industry to take our stuff and use it, at
least private industry does not come to the table with "hey, let's
cooperate" - we know who the corporate whores are, and we act accordingly.
However, when a linux developer comes to us and say "hey lets cooperate"
usually there is a thought of "this is a kindred spirit who understands
what our mutual goals are and won't stab us in the back". My concern
is that this situation will change if this is not rectified.
I think the community needs to decide, should cooperation be based on
morals and trust, or does the Linux community need to be regarded with
less trust than a Corporation, something to be avoided, as while
corporations can be counted on to act without morals, the knife is up
front and visible. They do not come to you with one hand of
cooperation extended and a knife kept behind their back.
-Bob
On Sat, Sep 01, 2007 at 07:29:39PM -0400, Constantine A. Murenin wrote:
> On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > On Sat, Sep 01, 2007 at 05:27:03PM -0400, Constantine A. Murenin wrote:
> > > On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > > > On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > > > > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > > > > Constantine A. Murenin wrote:
> > > > > > > This will hopefully help diminish certain myths about the code licensing.
> > > > > >
> > > > > > What myth? The myth that Theo understands dual licensing?
> > > > >
> > > > > Reyk's code was never dual licensed, so it's not like it even matters
> > > > > to the original dispute.
> > > >
> > > > It's no longer dual licenced in the FreeBSD tree because the FreeBSD
> > > > people removed the GPL choice of the dual licenced code 3 months ago.
> > >
> > > FreeBSD doesn't have Reyk's ath(4) HAL, which OpenHAL is based on.
> > >
> > > FreeBSD has a driver written by Sam, and a binary-only HAL, also written by Sam.
> > >
> > > > So all of Theo's accusations of people breaking the law by making this
> > > > dual licenced code GPL-only apply as well to the FreeBSD people...
> > >
> > > How? FreeBSD doesn't have Reyk's ath(4) HAL from OpenBSD, so there are
> > > no possible licensing accusations and violations.
> >
> > OK, I begin to understand this, there seem to be three different types
> > of files changed by Jiri's patch:
> > 1. dual licenced files planned to make GPL-only
> > 2. previously dual licenced files with a too recent version used planned
> > to make GPL-only
> > 3. never dual licenced files planned to make GPL-only
> >
> > For files under 1. and 2. Reyk did contribute to dual licenced code
> > without touching the licence, but I missed that there's also code unter 3.
> >
> > So there is a problem, but not with the code under 1. (unless you plan
> > to change the semantics of the word "alternatively"), the problem is
> > with some headers under 2. plus the code under 3.
> >
> > It's funny how Theo missed the part of Jiri's patch that actually is a
> > copyright violation and instead complains about the part that is OK...
>
> I'm not sure how you conclude that Theo missed the relevant parts --
> there were many messages posted to [email protected] mailing list and
> to The OpenBSD Journal in the last few days, and to me it appears as
> all of the problems were discussed ad nauseam.
>...
Then it's your fault that you forwarded the wrong email - in the email
you forwarded the only action for which Theo accused the Linux
developers of breaking the law was for choosing one licence when using
dual licenced code.
> After the obvious copyright violations were addressed, I think the
> problem started being an ethical one.
>
> As a free software user and developer, the question I have is how come
> the Linux community feels that they can take the BSD code that was
> reverse-engineered at OpenBSD, and put a more restrictive licence onto
> it, such that there will be no possibility of the changes going back
> to OpenBSD, given that the main work on the code has happened at
> OpenBSD? (Obviously, such a scenario it is permitted by the licence,
> but my question is an ethical one -- after all, most components of
> OpenHAL were specifically based on the OpenBSD's ath(4) HAL code.)
>
> You can see that Christoph Hellwig agrees with this ethical problem,
> as in the message below.
Is it a legal problem or is it "only" an ethical problem?
If choosing one licence when using dual licenced code is not a legal
problem then Theo repeatedly talking about it would "break the law" in
the email you forwarded was very unethical and the worst he could do
for his cause.
> C.
>...
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On 01/09/07, Luis R. Rodriguez <[email protected]> wrote:
> I urge developers to not bait into this and just leave this alone.
> Those involved know what they are doing and have a strong team of
> attorneys watching their backs. Any *necessary* discussions are be
> done privately.
Err...
I don't understand why you need a lawyer to interpret this document:
/* $OpenBSD: ar5210.c,v 1.39 2007/04/10 17:47:55 miod Exp $ */
/*
* Copyright (c) 2004, 2005, 2006, 2007 Reyk Floeter <[email protected]>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
If you want to add more code to it, and your contribution is
significant, simply add you name next to Reyk's. Where's the problem?
I don't know how licensing could be any simpler than this. Please,
notice, that there are no additional documents (other than the
copyright law) to read here -- _this is the complete licence_! (And
you have to read the copyright law even if you use the GNU GPL.)
C.
When companies have taken our wireless device drivers, many many of
them have given changes and fixes back. Some maybe didn't, but that
is OK.
When Linux took our changes back, they immediately locked the door
against changes moving back, by putting a GPL license on guard.
Why does our brother Linux take a file that is 90% BSD licensed,
and refuse to let us see the 10% he adds?
On 01/09/07, Adrian Bunk <[email protected]> wrote:
> On Sat, Sep 01, 2007 at 07:29:39PM -0400, Constantine A. Murenin wrote:
> > On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > > On Sat, Sep 01, 2007 at 05:27:03PM -0400, Constantine A. Murenin wrote:
> > > > On 01/09/07, Adrian Bunk <[email protected]> wrote:
> > > > > On Sat, Sep 01, 2007 at 01:37:18PM -0400, Constantine A. Murenin wrote:
> > > > > > On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > > > > > > Constantine A. Murenin wrote:
> > > > > > > > This will hopefully help diminish certain myths about the code licensing.
> > > > > > >
> > > > > > > What myth? The myth that Theo understands dual licensing?
> > > > > >
> > > > > > Reyk's code was never dual licensed, so it's not like it even matters
> > > > > > to the original dispute.
> > > > >
> > > > > It's no longer dual licenced in the FreeBSD tree because the FreeBSD
> > > > > people removed the GPL choice of the dual licenced code 3 months ago.
> > > >
> > > > FreeBSD doesn't have Reyk's ath(4) HAL, which OpenHAL is based on.
> > > >
> > > > FreeBSD has a driver written by Sam, and a binary-only HAL, also written by Sam.
> > > >
> > > > > So all of Theo's accusations of people breaking the law by making this
> > > > > dual licenced code GPL-only apply as well to the FreeBSD people...
> > > >
> > > > How? FreeBSD doesn't have Reyk's ath(4) HAL from OpenBSD, so there are
> > > > no possible licensing accusations and violations.
> > >
> > > OK, I begin to understand this, there seem to be three different types
> > > of files changed by Jiri's patch:
> > > 1. dual licenced files planned to make GPL-only
> > > 2. previously dual licenced files with a too recent version used planned
> > > to make GPL-only
> > > 3. never dual licenced files planned to make GPL-only
> > >
> > > For files under 1. and 2. Reyk did contribute to dual licenced code
> > > without touching the licence, but I missed that there's also code unter 3.
> > >
> > > So there is a problem, but not with the code under 1. (unless you plan
> > > to change the semantics of the word "alternatively"), the problem is
> > > with some headers under 2. plus the code under 3.
> > >
> > > It's funny how Theo missed the part of Jiri's patch that actually is a
> > > copyright violation and instead complains about the part that is OK...
> >
> > I'm not sure how you conclude that Theo missed the relevant parts --
> > there were many messages posted to [email protected] mailing list and
> > to The OpenBSD Journal in the last few days, and to me it appears as
> > all of the problems were discussed ad nauseam.
> >...
>
> Then it's your fault that you forwarded the wrong email - in the email
> you forwarded the only action for which Theo accused the Linux
> developers of breaking the law was for choosing one licence when using
> dual licenced code.
For the sake of the discussion, at the time I forwarded the message,
the obvious licensing problems (e.g. the original infamous patch by
Jiri) were already addressed.
Personally, these problems were so obvious -- entirely changing the
licence under Reyk's Copyright notice -- that I didn't even take them
for real when they first came across.
BTW, I've now once again re-read the original message that I've
forwarded, and it does contain Theo's reiteration of his response that
the original re-licensing patch had clear violations. E.g. re-read
this part of his message:
- If you receive ISC or BSD licensed code, you may not delete the
license. Same principle, since the notice says so. It's the law.
Really.
> > After the obvious copyright violations were addressed, I think the
> > problem started being an ethical one.
> >
> > As a free software user and developer, the question I have is how come
> > the Linux community feels that they can take the BSD code that was
> > reverse-engineered at OpenBSD, and put a more restrictive licence onto
> > it, such that there will be no possibility of the changes going back
> > to OpenBSD, given that the main work on the code has happened at
> > OpenBSD? (Obviously, such a scenario it is permitted by the licence,
> > but my question is an ethical one -- after all, most components of
> > OpenHAL were specifically based on the OpenBSD's ath(4) HAL code.)
> >
> > You can see that Christoph Hellwig agrees with this ethical problem,
> > as in the message below.
>
> Is it a legal problem or is it "only" an ethical problem?
I don't particularly like to repeat myself -- after the obvious
licensing issues were addressed, it has indeed become an ethical
problem: why do you think that you as the Linux community has to act
ruder to the *BSD community than the supposed corporations that we
always hear about in the BSD/GPL licensing arguments?
I really like the response that Bob Beck gave on this question:
http://lkml.org/lkml/2007/9/1/197
> If choosing one licence when using dual licenced code is not a legal
> problem then Theo repeatedly talking about it would "break the law" in
> the email you forwarded was very unethical and the worst he could do
> for his cause.
My understanding is that with dual-licensed code, you choose to comply
with all of the terms of either licence. However, you cannot simply
remove either of these licences from the code, unless you specifically
receive such right from the copyright holder (remember, with the
copyright law, unless the rights are specifically given, they are
retained). This is what Theo was trying to educate the community on. I
don't see anything unethical in explaining the legal issues.
Constantine.
On Sep 1, 2007, at 5:52 PM, Adrian Bunk wrote:
> OK, I begin to understand this, there seem to be three different types
> of files changed by Jiri's patch:
> 1. dual licenced files planned to make GPL-only
> 2. previously dual licenced files with a too recent version used
> planned
> to make GPL-only
> 3. never dual licenced files planned to make GPL-only
>
> For files under 1. and 2. Reyk did contribute to dual licenced code
> without touching the licence, but I missed that there's also code
> unter 3.
>
> So there is a problem, but not with the code under 1. (unless you plan
> to change the semantics of the word "alternatively"), the problem is
> with some headers under 2. plus the code under 3.
The BSD license plainly states:
"Permission to use, copy, modify, and/or distribute this software for
any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies."
Once the grantor (Reyk) releases his code under that license, it must
remain. You are free to derive work and redistribute under your
license, but the original copyright and license permission remains
intact. Many other entities (Microsoft, Apple, Sun, etc) have used
BSD code and have no problem understanding this. Why is this so
difficult for the Linux brain share to absorb?
As a former Linux advocate and current OpenBSD user/developer, I'm
appalled that fellow open-source developers would see fit to
cavalierly disregard the rights of the original copyright holder.
You wield the GPL when it suits you, and trample the courtesies of
non-GPL developers just because you [think you] can. As bad as
Jiri's offense was, it pales to the impudence displayed by Alan Cox,
one of the so-called defenders of free software.
Shame on you all.
---
Jason Dixon
DixonGroup Consulting
http://www.dixongroup.net
On Sat, Sep 01, 2007 at 06:36:36PM -0600, Theo de Raadt wrote:
> When companies have taken our wireless device drivers, many many of
> them have given changes and fixes back. Some maybe didn't, but that
> is OK.
>
> When Linux took our changes back, they immediately locked the door
> against changes moving back, by putting a GPL license on guard.
>
> Why does our brother Linux take a file that is 90% BSD licensed,
> and refuse to let us see the 10% he adds?
Theo, the primary claim you made in your email that was forwarded to
linux-kernel was:
<-- snip -->
In http://lkml.org/lkml/2007/8/29/183, Alan Cox managed to summarize
what Jiri Slaby and Luis Rodriguez were trying to do by proposing a
modification of a Dual Licenced file without the consent of all the
authors. Alan asks "So whats the problem ?". Well, Alan, I must
caution you -- your post is advising people to break the law.
<-- snip -->
It is a quite heavy accusation against Alan that saying it was OK to
change dual licenced code to one of the offered licences would advise
to break the law.
There's nothing about goodwill or other ethical questions in your
statement, this statement you made can be verified or falsified by
lawyers.
If it is true, all ethical questions about this are anyway moot because
it was illegal as you claim.
If you wrongly accused Alan, you owe Alan an apology.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Sat, Sep 01, 2007 at 06:02:26PM -0600, Bob Beck wrote:
> >As a free software user and developer, the question I have is how come
> >the Linux community feels that they can take the BSD code that was
> >reverse-engineered at OpenBSD, and put a more restrictive licence onto
> >it, such that there will be no possibility of the changes going back
> >to OpenBSD, given that the main work on the code has happened at
> >OpenBSD? (Obviously, such a scenario it is permitted by the licence,
> >but my question is an ethical one -- after all, most components of
> >OpenHAL were specifically based on the OpenBSD's ath(4) HAL code.)
> >
> >You can see that Christoph Hellwig agrees with this ethical problem,
> >as in the message below.
> >
> >C.
> >
> >
> >>On 28/08/07, Christoph Hellwig <[email protected]> wrote:
> >> On Tue, Aug 28, 2007 at 12:00:50PM -0400, Jiri Slaby wrote:
> >> > ath5k, license is GPLv2
> >> >
> >> > The files are available only under GPLv2 since now.
> >>
> >> Is this really a good idea? Most of the reverse-engineering was
> >> done by the OpenBSD folks, and it would certainly be helpful to
> >> work together with them on new hardware revisions, etc..
>
> I couldn't agree more. The point is, while we BSD license fans know and
> expect people from private industry to take our stuff and use it, at
> least private industry does not come to the table with "hey, let's
> cooperate" - we know who the corporate whores are, and we act accordingly.
>
> However, when a linux developer comes to us and say "hey lets cooperate"
> usually there is a thought of "this is a kindred spirit who understands
> what our mutual goals are and won't stab us in the back". My concern
> is that this situation will change if this is not rectified.
>
> I think the community needs to decide, should cooperation be based on
> morals and trust, or does the Linux community need to be regarded with
> less trust than a Corporation, something to be avoided, as while
> corporations can be counted on to act without morals, the knife is up
> front and visible. They do not come to you with one hand of
> cooperation extended and a knife kept behind their back.
Theo explicitely accused Alan that telling people that it was OK to
choose one licence for dual licenced code was "advising people to break
the law".
I hope you agree when talking about "cooperation [...] based on morals
and trust" that such accusations should either be proven correct or the
moral position of the person who made such accusations becomes quiet
weak.
> -Bob
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
Jason Dixon wrote:
> Once the grantor (Reyk) releases his code under that license, it must
> remain. You are free to derive work and redistribute under your
> license, but the original copyright and license permission remains
> intact. Many other entities (Microsoft, Apple, Sun, etc) have used BSD
> code and have no problem understanding this. Why is this so difficult
> for the Linux brain share to absorb?
Why is it so difficult to understand dual licensing?
Jeff
On 01/09/07, Jeff Garzik <[email protected]> wrote:
> Jason Dixon wrote:
> > Once the grantor (Reyk) releases his code under that license, it must
> > remain. You are free to derive work and redistribute under your
> > license, but the original copyright and license permission remains
> > intact. Many other entities (Microsoft, Apple, Sun, etc) have used BSD
> > code and have no problem understanding this. Why is this so difficult
> > for the Linux brain share to absorb?
>
> Why is it so difficult to understand dual licensing?
Maybe because Reyk's code was never dual-licensed?
C.
On 9/1/07, Constantine A. Murenin <[email protected]> wrote:
> On 01/09/07, Jeff Garzik <[email protected]> wrote:
> > Jason Dixon wrote:
> > > Once the grantor (Reyk) releases his code under that license, it must
> > > remain. You are free to derive work and redistribute under your
> > > license, but the original copyright and license permission remains
> > > intact. Many other entities (Microsoft, Apple, Sun, etc) have used BSD
> > > code and have no problem understanding this. Why is this so difficult
> > > for the Linux brain share to absorb?
> >
> > Why is it so difficult to understand dual licensing?
>
> Maybe because Reyk's code was never dual-licensed?
We asked SFLC to work with us to make sure that everyone's copyrights
were respected in the right places, and that the licenses various developers
wanted for their copyrights were implemented correctly. The patch I sent
implements SFLC's suggestions in that regard.
Luis
Constantine A. Murenin wrote:
> On 01/09/07, Jeff Garzik <[email protected]> wrote:
>> Jason Dixon wrote:
>>> Once the grantor (Reyk) releases his code under that license, it must
>>> remain. You are free to derive work and redistribute under your
>>> license, but the original copyright and license permission remains
>>> intact. Many other entities (Microsoft, Apple, Sun, etc) have used BSD
>>> code and have no problem understanding this. Why is this so difficult
>>> for the Linux brain share to absorb?
>> Why is it so difficult to understand dual licensing?
>
> Maybe because Reyk's code was never dual-licensed?
And yet a good portion of Theo's response, in particular his accusations
of Alan Cox exhorting people to break the law, were directly related to
dual licensing.
Jeff
On 01/09/07, Theo de Raadt <[email protected]> wrote:
> When companies have taken our wireless device drivers, many many of
> them have given changes and fixes back. Some maybe didn't, but that
> is OK.
>
> When Linux took our changes back, they immediately locked the door
> against changes moving back, by putting a GPL license on guard.
>
> Why does our brother Linux take a file that is 90% BSD licensed,
> and refuse to let us see the 10% he adds?
Indeed, it's upsetting that people like Luis Rodriguez push for the
lawyers to be involved to (fight?) an open source project. Why, may I
ask?
Why Luis puts the phrase "legal hell" next to entirely free software?
[0] Why is he trying to go against the BSD community, which gave him
the entire HAL framework for the driver in question?
Best regards,
Constantine.
[0] http://marc.info/?l=linux-wireless&m=118857712529898&w=2
On Sat, Sep 01, 2007 at 09:42:54PM -0400, Luis R. Rodriguez wrote:
> We asked SFLC to work with us to make sure that everyone's copyrights
> were respected in the right places, and that the licenses various developers
> wanted for their copyrights were implemented correctly. The patch I sent
> implements SFLC's suggestions in that regard.
You know, I'm rapidly losing any respect to both sides of that. Eben
Moglen as source of advice in "is it OK to convert to GPL-only"? And
you seriously rely on morality of that? Theo's rants aside, if you
have to rely on SFLC for licensing decisions... Ouch.
On Sat, Sep 01, 2007 at 08:36:24PM -0400, Jason Dixon wrote:
> On Sep 1, 2007, at 5:52 PM, Adrian Bunk wrote:
>
>> OK, I begin to understand this, there seem to be three different types
>> of files changed by Jiri's patch:
>> 1. dual licenced files planned to make GPL-only
>> 2. previously dual licenced files with a too recent version used planned
>> to make GPL-only
>> 3. never dual licenced files planned to make GPL-only
>>
>> For files under 1. and 2. Reyk did contribute to dual licenced code
>> without touching the licence, but I missed that there's also code unter 3.
>>
>> So there is a problem, but not with the code under 1. (unless you plan
>> to change the semantics of the word "alternatively"), the problem is
>> with some headers under 2. plus the code under 3.
>
> The BSD license plainly states:
>
> "Permission to use, copy, modify, and/or distribute this software for any
> purpose with or without fee is hereby granted, provided that the above
> copyright notice and this permission notice appear in all copies."
>
> Once the grantor (Reyk) releases his code under that license, it must
> remain. You are free to derive work and redistribute under your license,
> but the original copyright and license permission remains intact. Many
> other entities (Microsoft, Apple, Sun, etc) have used BSD code and have no
> problem understanding this. Why is this so difficult for the Linux brain
> share to absorb?
>
> As a former Linux advocate and current OpenBSD user/developer, I'm appalled
> that fellow open-source developers would see fit to cavalierly disregard
> the rights of the original copyright holder. You wield the GPL when it
> suits you, and trample the courtesies of non-GPL developers just because
> you [think you] can. As bad as Jiri's offense was, it pales to the
> impudence displayed by Alan Cox, one of the so-called defenders of free
> software.
>
> Shame on you all.
Jiri's patch would have wrongly not only removed the BSD statement from
dual licenced files but also from not dual licenced files.
This was a mistake in this patch (that was never merged into the tree)
neither Jiri nor Alan noticed.
The only disagreement is about the following:
Theo claimed boldly in the email that started this thread on
linux-kernel it would "break the law" to choose one licence for dual
licenced code like the following:
/* $OpenBSD: ath.c,v 1.63 2007/05/09 16:41:14 reyk Exp $ */
/* $NetBSD: ath.c,v 1.37 2004/08/18 21:59:39 dyoung Exp $ */
/*-
* Copyright (c) 2002-2004 Sam Leffler, Errno Consulting
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer,
* without modification.
* 2. Redistributions in binary form must reproduce at minimum a disclaimer
* similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any
* redistribution must be conditioned upon including a substantially
* similar Disclaimer requirement for further binary redistribution.
* 3. Neither the names of the above-listed copyright holders nor the names
* of any contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* NO WARRANTY
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY
* AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY,
* OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
* IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
* THE POSSIBILITY OF SUCH DAMAGES.
*/
> Jason Dixon
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
Suppose you saw some other variant of *nix that had some code you wanted
to use, but there was a gaping security hole in it. Wouldn't you patch
it before you incorporated it? and would it be your fault if this fix
made the code not work with the original?
We took the code and fixed a gaping security vulnerability that appeared
within the opening comment. We DO care who does what with our code, and
we fully intend to cover our balls.
The problem is yours to fix. If you actually care, use a license that
SAYS you care. Right now there's a big /* I don't give a shit */ on top
of every BSD file. We took you at your word and assumed you didn't. Now
its too late and you suddenly care, don't you?
Constantine A. Murenin wrote:
> Indeed, it's upsetting that people like Luis Rodriguez push for the
> lawyers to be involved to (fight?) an open source project. Why, may I
> ask?
Is it not self-evident? Legal review is the sane course of action, when
legal issues are the bone of contention.
That said, Linux people are far more pragmatic than FSF people, and
often disagree with the FSF. I would not take an FSF lawyer's word as
Gospel.
Theo manages to confuse "Linux" and "FSF" quite often, but that's
characteristic of his muddled thinking and inexperience.
Jeff
On Sat, Sep 01, 2007 at 09:58:26PM -0400, Casey Dahlin wrote:
> Suppose you saw some other variant of *nix that had some code you wanted
> to use, but there was a gaping security hole in it. Wouldn't you patch
> it before you incorporated it? and would it be your fault if this fix
> made the code not work with the original?
>
> We took the code and fixed a gaping security vulnerability that appeared
> within the opening comment. We DO care who does what with our code, and
> we fully intend to cover our balls.
Who's "we" and would you mind showing your contributions to the tree?
git doesn't seem to find them in 2.5.0-to-current for some reason...
On Sep 1, 2007, at 9:58 PM, Casey Dahlin wrote:
> Suppose you saw some other variant of *nix that had some code you
> wanted to use, but there was a gaping security hole in it. Wouldn't
> you patch it before you incorporated it? and would it be your fault
> if this fix made the code not work with the original?
>
> We took the code and fixed a gaping security vulnerability that
> appeared within the opening comment. We DO care who does what with
> our code, and we fully intend to cover our balls.
Since when is this *your* code? Oh that's right, when Jiri decided
to steal it by deleting Reyk's copyright and license. Oh wait,
that's already been corrected. What was your point again?
> The problem is yours to fix. If you actually care, use a license
> that SAYS you care. Right now there's a big /* I don't give a shit
> */ on top of every BSD file. We took you at your word and assumed
> you didn't. Now its too late and you suddenly care, don't you?
The BSD license, in effect, says that we care about good code. We
allow anyone to use it. The only stipulation is that the copyright
and license permissions must remain intact.
Are you trying to be hateful, or are you really this ignorant?
---
Jason Dixon
DixonGroup Consulting
http://www.dixongroup.net
On Sat, Sep 01, 2007 at 08:36:24PM -0400, Jason Dixon wrote:
>
> On Sep 1, 2007, at 5:52 PM, Adrian Bunk wrote:
>
>> OK, I begin to understand this, there seem to be three different types
>> of files changed by Jiri's patch:
>> 1. dual licenced files planned to make GPL-only
>> 2. previously dual licenced files with a too recent version used planned
>> to make GPL-only
>> 3. never dual licenced files planned to make GPL-only
>>
>> For files under 1. and 2. Reyk did contribute to dual licenced code
>> without touching the licence, but I missed that there's also code unter 3.
>>
>> So there is a problem, but not with the code under 1. (unless you plan
>> to change the semantics of the word "alternatively"), the problem is
>> with some headers under 2. plus the code under 3.
>
> The BSD license plainly states:
>
> "Permission to use, copy, modify, and/or distribute this software for any
> purpose with or without fee is hereby granted, provided that the above
> copyright notice and this permission notice appear in all copies."
>
> Once the grantor (Reyk) releases his code under that license, it must
> remain. You are free to derive work and redistribute under your license,
> but the original copyright and license permission remains intact. Many
> other entities (Microsoft, Apple, Sun, etc) have used BSD code and have no
> problem understanding this. Why is this so difficult for the Linux brain
> share to absorb?
>
> As a former Linux advocate and current OpenBSD user/developer, I'm appalled
> that fellow open-source developers would see fit to cavalierly disregard
> the rights of the original copyright holder. You wield the GPL when it
> suits you, and trample the courtesies of non-GPL developers just because
> you [think you] can. As bad as Jiri's offense was, it pales to the
> impudence displayed by Alan Cox, one of the so-called defenders of free
> software.
>From http://www.mac.linux-m68k.org/docs/macpaper.php
"Always be the second operating system port to an undocumented platform.
The sterling work done by the OpenBSD/Mac team was a huge help to the
Linux project. I'm also happy to say that while half of the world may
sit on usenet advocacy groups throwing manure the relationship between
the Linux and BSD Macintosh teams has always been one of mutual
co-operation. Together we advance our detective work and knowledge of
the Macintosh platforms to the good of all Macintosh users dumped"
Alan Cox circa 1999.
http://lists.freedesktop.org/archives/xorg/2007-August/027419.html
"well I'd be quite happy to see X go GPL but I'm aware
thats not the intention of the project ;)"
Alan Cox circa 2007.
What changed? Why are you guys setting out to break all of the
work underpinning UNIX and the Internet done in the 80s at
Berkeley? The reason the protocols and infrastructure took
off in the first place is due to liberal licenses that let everyone
be involved, not wrapping things up in more restrictions and lawyers.
On Sun, 02 Sep 2007 03:55:37 +0200, Adrian Bunk said:
> Jiri's patch would have wrongly not only removed the BSD statement from
> dual licenced files but also from not dual licenced files.
>
> This was a mistake in this patch (that was never merged into the tree)
> neither Jiri nor Alan noticed.
You know, we *could* have solved this a *hell* of a lot easier if people quit
flaming about it, and we did something *productive* instead.
Like submit a corrected patch. :)
On 01/09/07, [email protected] <[email protected]> wrote:
> On Sun, 02 Sep 2007 03:55:37 +0200, Adrian Bunk said:
>
> > Jiri's patch would have wrongly not only removed the BSD statement from
> > dual licenced files but also from not dual licenced files.
> >
> > This was a mistake in this patch (that was never merged into the tree)
> > neither Jiri nor Alan noticed.
>
> You know, we *could* have solved this a *hell* of a lot easier if people quit
> flaming about it, and we did something *productive* instead.
>
> Like submit a corrected patch. :)
Dear Valdis,
The idea here is that no patching was needed in the first place --
most of the files are/were BSD-licensed, because they were forked from
OpenBSD.
It is beneficial for the atmosphere of both projects to keep the
licence compatible. If Linux tries to GPL future modifications to
Reyk's code, then OpenBSD would not be able to take back the changes.
But this would not be the case if all modifications to Reyk's code are
continued to be BSD-licensed. This is what this whole issue is about.
My understanding, is that Nick Kossifidis never had a problem with
licensing his changes with a BSD-license, although Jiri Slaby always
used GPLv2.
With the last patch posted by Luis [0], Jiri actually recalled his
original patch and relicensed all of his GPLv2 work under BSD (!);
however, at the very same time, Nick changed his mind, and decided to
relicense his BSD code under GPLv2 (!). (Surprise! Yes, it appears
that both Nick and Jiri decided to switch their licensing positions,
and mutually relicense their work under each other's respective
licence. :) Is everyone ready going to go back and forth now? Does
this whole story still makes any sense to you? :)
I hope that both Nick Kossifidis and Jiri Slaby can agree on licensing
their HAL code with a BSD licence, so that the code remains
licence-compatible with OpenBSD. If there are any unresolved licensing
questions, I personally would be more than happy to answer any such
questions as much as I can, and yes -- I am not a lawyer. :)
Nick, Jiri -- since much of the work on OpenHAL is based on Reyk's
HAL, could you please be so kind as to both agree to licence you
changes in OpenHAL with the same licence as Reyk does in OpenBSD's
ath(4) HAL? This step will be very welcome in the OpenBSD community at
large.
Best regards,
Constantine.
[0] http://marc.info/?l=linux-wireless&m=118857712529898&w=2
On Sun, 02 Sep 2007 01:09:18 EDT, "Constantine A. Murenin" said:
> The idea here is that no patching was needed in the first place --
> most of the files are/were BSD-licensed, because they were forked from
> OpenBSD.
Oh, silly me. For some reason, I had it in my head that Jiri's original
patch actually included some real live *code* in addition to the parts that
changed the licensing text... ;)
> co-operation. Together we advance our detective work and knowledge of
> the Macintosh platforms to the good of all Macintosh users dumped"
>
> Alan Cox circa 1999.
>
> http://lists.freedesktop.org/archives/xorg/2007-August/027419.html
>
> "well I'd be quite happy to see X go GPL but I'm aware
> thats not the intention of the project ;)"
> Alan Cox circa 2007.
>
> What changed?
Nothing that I am aware of. You can't take Linux/Mac68K code back into
BSD either. BSD code is being used according to the BSD licence. You
could adopt a different licence if the way your code is being used
bothers you, thats.
Where I've reused BSD code I've aways tried to contribute it back to the
BSD people or share the knowledge (and the knowledge far more than the
code mattered both ways for Mac68K systems)
I suggest you read drivers/net/wan/syncppp.c, which is I think the only
BSD derived bit of code of mine left in the kernel - and its quite
specific what it says.
Ath5k isn't my code so I don't get to pick. Having the OpenBSD maintainer
make bogus remarks about that doesn't help anyone, especially when he's
wrong and doesn't appear to know about the subject in the first place.
Alan
> - If you receive ISC or BSD licensed code, you may not delete the
> license. Same principle, since the notice says so. It's the law.
> Really.
You can shout this all you like but you would be wrong. You can remove
the licence if you have permission to do so. For the ath c files there
was permission to do so.
> My understanding is that with dual-licensed code, you choose to comply
> with all of the terms of either licence. However, you cannot simply
> remove either of these licences from the code, unless you specifically
> receive such right from the copyright holder (remember, with the
> copyright law, unless the rights are specifically given, they are
> retained). This is what Theo was trying to educate the community on. I
> don't see anything unethical in explaining the legal issues.
Your understanding isn't quite right. One of many things you may get with
dual licensed code is the right to pick a licence from several choices,
you may also get the right to remove some choices from the recipient.
A work that combines GPL and BSD licensed material is not the same as a
work which says I may choose between two licences. If both licences must
always apply (which is a perfectly possible condition to put in a
licence) then putting such a "both" GPL/BSD licence piece of code into
OpenBSD would require any OpenBSD distributed containing it was GPL
licenced when conveyed, which I am *very* sure is not the intent.
Thus what you appear to be doing by putting the ath5k C code in OpenBSD is
conveying it under the BSD licence (making a choice between the two
offered) and conveying a right for parties down the chain to convey it
under one of the licences only.
And as we've already established the header files are quite different.
Doesn't mean its not somewhat rude but illegal and rude are two very
different things.
On Sep 1 2007 18:36, Theo de Raadt wrote:
>
>When companies have taken our wireless device drivers, many many of
>them have given changes and fixes back. Some maybe didn't, but that
>is OK.
For companies it's ok, but for linux people it is not?
(1) You do not know how much of the modifications companies did
are actually returned
(2) You do not know whether the ath5k linux part authors will
give back at a later point (much like companies)
On Sun, Sep 02, 2007 at 01:20:27PM +0200, Igor Sobrado wrote:
> On Sun, 2 Sep 2007, Alan Cox wrote:
>> You can shout this all you like but you would be wrong. You can remove
>> the licence if you have permission to do so. For the ath c files there
>> was permission to do so.
>
> There was permission to do so from Reyk Floeter? Really?
>
>> Your understanding isn't quite right. One of many things you may get with
>> dual licensed code is the right to pick a licence from several choices,
>> you may also get the right to remove some choices from the recipient.
>
> Reyk code was never dual licensed! His code is under truly free licensing
> terms (BSD).
Jiri's patch touched both files containing BSD-only code by Reyk and
code Reyk contributed to leaving the file dual licenced.
>> A work that combines GPL and BSD licensed material is not the same as a
>> work which says I may choose between two licences. If both licences must
>> always apply (which is a perfectly possible condition to put in a
>> licence) then putting such a "both" GPL/BSD licence piece of code into
>> OpenBSD would require any OpenBSD distributed containing it was GPL
>> licenced when conveyed, which I am *very* sure is not the intent.
>>
>> Thus what you appear to be doing by putting the ath5k C code in OpenBSD is
>> conveying it under the BSD licence (making a choice between the two
>> offered) and conveying a right for parties down the chain to convey it
>> under one of the licences only.
>
> I think that Theo explained this point clearly quite a few times in the
> last days.
>
>> And as we've already established the header files are quite different.
>
> Is a simple change in the header files a reason to vindicate the people
> that changed the licensing terms? Obviously, it isn't.
>
>> Doesn't mean its not somewhat rude but illegal and rude are two very
>> different things.
>
> No, because this change is both rude and illegal.
You mixed two completely different things in your email:
1. Jiri's patch (that was never merged into Linux) not only removed the
BSD header from dual licenced files but also from not dual licenced
files.
2. Theo accused Alan that telling people that it was OK to choose one
licence for dual licenced code was "advising people to break the law".
Jiri's patch was legally not OK regarding 1. - there's no discussion
regarding this.
The point 2 is what the email of Theo that was forwarded to linux-kernel
is about and what the discussion is about. That's quite a rude action
by Theo unless he's able to prove that this accusation is correct.
> Igor
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Sun, 2 Sep 2007, Adrian Bunk wrote:
> On Sun, Sep 02, 2007 at 01:20:27PM +0200, Igor Sobrado wrote:
>> Reyk code was never dual licensed! His code is under truly free licensing
>> terms (BSD).
>
> Jiri's patch touched both files containing BSD-only code by Reyk and
> code Reyk contributed to leaving the file dual licenced.
Ok.
> You mixed two completely different things in your email:
>
> 1. Jiri's patch (that was never merged into Linux) not only removed the
> BSD header from dual licenced files but also from not dual licenced
> files.
>
> 2. Theo accused Alan that telling people that it was OK to choose one
> licence for dual licenced code was "advising people to break the law".
>
> Jiri's patch was legally not OK regarding 1. - there's no discussion
> regarding this.
>
> The point 2 is what the email of Theo that was forwarded to linux-kernel
> is about and what the discussion is about. That's quite a rude action
> by Theo unless he's able to prove that this accusation is correct.
When code is multi-licensed it must be distributed under *all* these
licensing terms concurrently. It is easy to understand. Removing (or
changing) the conditions that apply to the program from the source code
and documentation *without* an authorization from all the author(s) is
illegal.
So, a multi-licensed file remains multi-licensed except when all authors
agree about a change in the licensing terms. And it is clear on the BSD
license that a modification of the distribution terms is illegal. It is
the first clause on the BSD license:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer,
* without modification.
So, removing (or changing) the list of conditions on the BSD license is
not allowed.
Igor.
On Sun, 02 Sep 2007 13:20:27 +0200 (CEST)
Igor Sobrado <[email protected]> wrote:
> On Sun, 2 Sep 2007, Alan Cox wrote:
> > You can shout this all you like but you would be wrong. You can remove
> > the licence if you have permission to do so. For the ath c files there
> > was permission to do so.
>
> There was permission to do so from Reyk Floeter? Really?
The code pieces I quoted contained that choice. As far as I am concerned
that is what the discussion was about.
Alan
Igor Sobrado wrote:
> When code is multi-licensed it must be distributed under *all* these
> licensing terms concurrently. It is easy to understand. Removing (or
> changing) the conditions that apply to the program from the source code
> and documentation *without* an authorization from all the author(s) is
> illegal.
The plain English in the dual-license text directly contradicts this
fiction.
Jeff
> So, a multi-licensed file remains multi-licensed except when all authors
> agree about a change in the licensing terms. And it is clear on the BSD
Not strictly true. They can either agree to a change and issue one or
they can convey to other parties the right to change the terms. The GPL
for example does this for version selection.
A multi-licensed work (note work not file - don't assume a file is a
boundary of a work) which conveys the choice of licence (as some bits of
ath5k did) allows a receiving party to choose the licence it wishes.
Failing that OpenBSD would have turned itself GPL by adding that file as
according to your argument "it must be distributed under *all* these
licensing terms concurrently".
Alan
On Sun, 2 Sep 2007, Alan Cox wrote:
> You can shout this all you like but you would be wrong. You can remove
> the licence if you have permission to do so. For the ath c files there
> was permission to do so.
There was permission to do so from Reyk Floeter? Really?
> Your understanding isn't quite right. One of many things you may get with
> dual licensed code is the right to pick a licence from several choices,
> you may also get the right to remove some choices from the recipient.
Reyk code was never dual licensed! His code is under truly free
licensing terms (BSD).
> A work that combines GPL and BSD licensed material is not the same as a
> work which says I may choose between two licences. If both licences must
> always apply (which is a perfectly possible condition to put in a
> licence) then putting such a "both" GPL/BSD licence piece of code into
> OpenBSD would require any OpenBSD distributed containing it was GPL
> licenced when conveyed, which I am *very* sure is not the intent.
>
> Thus what you appear to be doing by putting the ath5k C code in OpenBSD is
> conveying it under the BSD licence (making a choice between the two
> offered) and conveying a right for parties down the chain to convey it
> under one of the licences only.
I think that Theo explained this point clearly quite a few times in the
last days.
> And as we've already established the header files are quite different.
Is a simple change in the header files a reason to vindicate the people
that changed the licensing terms? Obviously, it isn't.
> Doesn't mean its not somewhat rude but illegal and rude are two very
> different things.
No, because this change is both rude and illegal.
Igor
On Sun, 2 Sep 2007, Alan Cox wrote:
>> So, a multi-licensed file remains multi-licensed except when all authors
>> agree about a change in the licensing terms. And it is clear on the BSD
>
> Not strictly true. They can either agree to a change and issue one or
> they can convey to other parties the right to change the terms. The GPL
> for example does this for version selection.
So, under a dual-licensed BSD/GPL code the latter license allows a
developer to remove the GPL license itself and release a single-licensed
BSD code if other parties want to do it?
> A multi-licensed work (note work not file - don't assume a file is a
> boundary of a work) which conveys the choice of licence (as some bits of
> ath5k did) allows a receiving party to choose the licence it wishes.
> Failing that OpenBSD would have turned itself GPL by adding that file as
> according to your argument "it must be distributed under *all* these
> licensing terms concurrently".
I would assume a file as a boundary of a work in the case that file is
under different licensing terms to the rest of the software package. On a
lot of software packages different modules are covered under different
licensing terms.
We can choose what license terms we will honor; however, we do not have
the ability to remove the licensing terms we do not like.
Igor.
On Sun, Sep 02, 2007 at 03:00:46PM +0200, Igor Sobrado wrote:
> On Sun, 2 Sep 2007, Alan Cox wrote:
>
>>> So, a multi-licensed file remains multi-licensed except when all authors
>>> agree about a change in the licensing terms. And it is clear on the BSD
>>
>> Not strictly true. They can either agree to a change and issue one or
>> they can convey to other parties the right to change the terms. The GPL
>> for example does this for version selection.
>
> So, under a dual-licensed BSD/GPL code the latter license allows a
> developer to remove the GPL license itself and release a single-licensed
> BSD code if other parties want to do it?
Exactly.
>> A multi-licensed work (note work not file - don't assume a file is a
>> boundary of a work) which conveys the choice of licence (as some bits of
>> ath5k did) allows a receiving party to choose the licence it wishes.
>> Failing that OpenBSD would have turned itself GPL by adding that file as
>> according to your argument "it must be distributed under *all* these
>> licensing terms concurrently".
>
> I would assume a file as a boundary of a work in the case that file is
> under different licensing terms to the rest of the software package. On a
> lot of software packages different modules are covered under different
> licensing terms.
>
> We can choose what license terms we will honor; however, we do not have the
> ability to remove the licensing terms we do not like.
We have the ability if the author explicitely allowed it.
This is the licencing text we are talking about:
/*-
* Copyright (c) 2002-2004 Sam Leffler, Errno Consulting
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer,
* without modification.
* 2. Redistributions in binary form must reproduce at minimum a disclaimer
* similar to the "NO WARRANTY" disclaimer below ("Disclaimer") and any
* redistribution must be conditioned upon including a substantially
* similar Disclaimer requirement for further binary redistribution.
* 3. Neither the names of the above-listed copyright holders nor the names
* of any contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* NO WARRANTY
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTIBILITY
* AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
* THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY,
* OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
* IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
* THE POSSIBILITY OF SUCH DAMAGES.
*/
The author himself offered two _alternatives_ for distributing his code.
> Igor.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
> > Not strictly true. They can either agree to a change and issue one or
> > they can convey to other parties the right to change the terms. The GPL
> > for example does this for version selection.
>
> So, under a dual-licensed BSD/GPL code the latter license allows a
> developer to remove the GPL license itself and release a single-licensed
> BSD code if other parties want to do it?
If the dual licence permits you to select from two alternatives as
appears to be the case in
"* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation."
Then there is no problem in doing exactly what it says and distributing
it under the terms of the GPL v2 and the GPL v2 alone (or indeed the BSD
licence alone). Anyone who took the project code and produced a binary
only proprietary product from it would for example select the BSD licence
alone and convey almost no rights at all to their customer.
> I would assume a file as a boundary of a work in the case that file is
> under different licensing terms to the rest of the software package. On a
Assuming is bad, you should consult caselaw.
> lot of software packages different modules are covered under different
> licensing terms.
>
> We can choose what license terms we will honor; however, we do not have
> the ability to remove the licensing terms we do not like.
If the author has conveyed that right to you, then you may usually do so.
Alan
IANAL, but:
Igor Sobrado <[email protected]> writes:
> So, under a dual-licensed BSD/GPL code the latter license allows a
> developer to remove the GPL license itself and release a
> single-licensed BSD code if other parties want to do it?
Of course. If it wasn't legal, dual BSD/GPL would just be equal
to GPL. Now, dual BSD/GPL equals BSD.
OTOH I'd probable leave the original licence text, something like:
The actual licence conditions:
GPL or BSD or whatever.
Portions of this file were licenced under:
[the original licence text, not valid as a licence for current file]
WRT Atheros driver I'd probably leave the thing as is (i.e., BSD/GPL
= in fact BSD), unless something like 50+% of the code is rewritten -
it's mostly their hard work after all, isn't it? Not legal
requirement, though.
--
Krzysztof Halasa
Krzysztof Halasa wrote:
> WRT Atheros driver I'd probably leave the thing as is (i.e., BSD/GPL
> = in fact BSD), unless something like 50+% of the code is rewritten -
> it's mostly their hard work after all, isn't it? Not legal
> requirement, though.
Yes. This deserves to be reinforced:
There is definite value in sharing the ath5k HAL between OpenBSD and Linux.
Jeff
Igor Sobrado <[email protected]> wrote:
> When code is multi-licensed it must be distributed under *all* these
> licensing terms concurrently.
No. E.g.:
If I don't agree to the GPL (or if I had violated it and therefore have lost
it's privileges), I MUST NOT redistribute it under the GPL because I have no
license to do that, but the BSD license would still allow me to redistribute.
On Sun, 2 Sep 2007, Jeff Garzik wrote:
> Krzysztof Halasa wrote:
>> WRT Atheros driver I'd probably leave the thing as is (i.e., BSD/GPL
>> = in fact BSD), unless something like 50+% of the code is rewritten -
>> it's mostly their hard work after all, isn't it? Not legal
>> requirement, though.
>
> Yes. This deserves to be reinforced:
>
> There is definite value in sharing the ath5k HAL between OpenBSD and Linux.
Of course. Sharing knowledge and efforts can only improve both the GPL
and BSD licensed code. It is important in all cases, but becomes critical
when support from manufacturers is limited or even non existent. In these
cases, shared efforts are required to write successful code.
Cheers,
Igor.
On Sun, Sep 02, 2007 at 03:00:46PM +0200, Igor Sobrado wrote:
> >Not strictly true. They can either agree to a change and issue one or
> >they can convey to other parties the right to change the terms. The GPL
> >for example does this for version selection.
>
> So, under a dual-licensed BSD/GPL code the latter license allows a
> developer to remove the GPL license itself and release a single-licensed
> BSD code if other parties want to do it?
Exactly. That's what dual-licensing is.
[quote]
This is no different from the fact that we have some drivers that are
GPLv2/BSD licensed. Within the kernel, they are GPLv2. But on their own,
you can choose to use them under the BSD license, make your changes to
them, and release them commercially.
And correct - I cannot (and neither can anybody else) then accept those
*non*GPLv2 changes back.
[end quote]
That's from Linus and quite recently.
FWIW, it's damn hard to codify "... and changes to this code should not
change the situation". It's certainly a very good policy and in this
case it's the only sane policy.
[quote]
Actually, normally I *do* have such a trust. It's why I have no problem
with drivers that are dual-GPL/BSD, and in fact, I've told people that I
don't want them to turn them into GPL-only, because that is simply not
polite.
[end quote]
Same posting from Linus. And that's much more relevant to shooting the patch
in question down (and IMO it ought to be shot down) than references to
legality.
This has been pretty interesting for me to watch as I distribute my
isp driver under a dual license (at least the portions of it which are
common with the *BSD and Solaris ports) that is almost identical to
Sam's verbiage.
I'll admit that I hadn't thought about whether redistribution included
the ability to modify the header (and thus the text of the licensing
as I had written) or not. On balance I'd say I believe that the
arguments for, on redistribution, picking one or the other license
makes sense and honored my general intent.
This allows people who modify the code (and presumably improve it) a
"chef's choice" based on where they're serving the meal.
IANAL, but I believe that none of this keeps me from continuing to put
a dual license on stuff I leave up for distribution, or changing that
to restricting the code to Martian Triathalon winners or what have
you.
> - If you receive dual licensed code, you may not delete the license
> you don't like and then distribute it. It has to stay, because you
> may not edit someone's else's license -- which is a three-part legal
> document (For instance: Copyright notice, BSD, followed by GPL).
This is absolute nonsense. By default, you can remove a license if you want
to. This is why both the GPL and the BSD licenses have clauses requiring you
to leave them in.
A file that is under a dual license may be used under either the GPL license
or the BSD license. Neither license requires you to retain the *other*
license in the file. So there is absolutely no reason you cannot remove one
license or the other. To argue otherwise is to argue that you need to comply
with *both* licenses in a dual-licensed file to get the rights granted by
either, and that's nonsense.
You cannot, of course, modify a license and expect your modified license to
apply to protectable elements you didn't author. And anyone who receives
modified versions of the file still has all the rights the original authors
grant them.
Let's perform a thought experiment for a moment. Suppose the BSD license
explicitly said you could remove the licensing clause if you wanted to.
Would you still argue that you couldn't remove it even though it says you
can? Well, the GPL says you can modify anything you want to, except *THAT*
license. This means you can remove any other license notifications you want.
Note that your license editing or removing has no effect on the rights
people actually get except to your code.
DS
Alan Cox wrote:
> The ath5k C file in question (not the headers) seems to give recipients
> permission to further convey the work under a choice of two licences.
Correct.
> It doesn't say they must redistribute under both.
Correct. They need the right to redistribute the work, and they may obtain
that right from either license.
> So I appear to have a
> right to convey the work under the GPL to a third party, who from me
> receives no right to use it except under the GPL.
Here's where your train goes off the rails. They do not receive any right to
use it from you. They receive a license to use it under the GPL from the
original author. Please read GPL section 6.
" 6. Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License."
The GPL does not give you *any* right to extend anyone a license to code you
did not author. (Nor can it as such an extension would have to be done in
writing in most countries.) When you distribute a GPL'd work, the right to
use every creative element in that work is licensed to the recipients
directly from their respective authors. Under no circumstances does the GPL
ever give you the ability to license someone else's work to a third party.
> * Alternatively, this software may be distributed under the terms of the
> * GNU General Public License ("GPL") version 2 as published by the Free
> * Software Foundation.
>
> The choice appears to be delegated to the recipient very clearly and
> very specifically by the licencing on the file. It does not say that I
> must convey the work under both licences. It quite specifically says I may
> convey the work under whichever of the two I prefer (and probably both if
> I wish). Clearly if that had not been the intent it would not have
> included the clause giving the choice.
Either license can grant you the right to distribute it, but how you get the
rights to distribute has *NO* effect on the recipient. They receive a lawful
copy and any rights the original author grants them under a license from
that original author. You have no power to grant or modify rights to the
original work.
This is a common misunderstanding.
Note that you may remove the text of either license from a dual-licensed
file and redistribute under the other license because neither license
requires you to retain the other license and both licenses give you the
right otherwise to modify as you wish. But the removal of a license from a
file has no effect on the grant of license. Your recipients still get a dual
license to those protectable elements in the file that were placed under a
dual license. You cannot stop the automatic grant.
DS
(by the way, text in caps surrounded by *'s is meant to indicate vocal stress,
not volume)
On Sunday 02 September 2007 22:01:18 David Schwartz wrote:
<snip>
> > So I appear to have a
> > right to convey the work under the GPL to a third party, who from me
> > receives no right to use it except under the GPL.
>
> Here's where your train goes off the rails. They do not receive any right
> to use it from you. They receive a license to use it under the GPL from the
> original author. Please read GPL section 6.
>
> " 6. Each time you redistribute the Program (or any work based on the
> Program), the recipient automatically receives a license from the
> original licensor to copy, distribute or modify the Program subject to
> these terms and conditions. You may not impose any further
> restrictions on the recipients' exercise of the rights granted herein.
> You are not responsible for enforcing compliance by third parties to
> this License."
>
> The GPL does not give you *any* right to extend anyone a license to code
> you did not author. (Nor can it as such an extension would have to be done
> in writing in most countries.) When you distribute a GPL'd work, the right
> to use every creative element in that work is licensed to the recipients
> directly from their respective authors. Under no circumstances does the GPL
> ever give you the ability to license someone else's work to a third party.
However, this is not what is happening here. Jiri has made changes that he has
licensed solely under the GPLv2. This means that he now becomes the
licensor - not Mr. Floeter or Mr. Leffler. *BUT* *ONLY* of the version of the
code containing his changes.
Mr. Floeter *CAN* request that his code be removed from said fork - his code
is solely licensed (AFAICT and IIRC) under the BSD/ISC license and was only
covered by the dual-license because it was integrated into a work that
carried said dual-license. (I'm not sure how well such a revocation would
work in reality, but it is Mr. Floeters right.)...
(Sam Leffler could do the same - but I'm not sure how well that would carry.)
> > * Alternatively, this software may be distributed under the terms of the
> > * GNU General Public License ("GPL") version 2 as published by the Free
> > * Software Foundation.
> >
> > The choice appears to be delegated to the recipient very clearly and
> > very specifically by the licencing on the file. It does not say that I
> > must convey the work under both licences. It quite specifically says I
> > may convey the work under whichever of the two I prefer (and probably
> > both if I wish). Clearly if that had not been the intent it would not
> > have included the clause giving the choice.
>
> Either license can grant you the right to distribute it, but how you get
> the rights to distribute has *NO* effect on the recipient. They receive a
> lawful copy and any rights the original author grants them under a license
> from that original author. You have no power to grant or modify rights to
> the original work.
Correct. Doesn't apply in the case of the code in question (unless the changes
that were made are so tiny as to not be copyrightable).
In this case the code is question is a modified version, which means that the
right to distribute said modified version now originates with the person
holding the copyright on the modifications. (Though their right to distribute
the code, in such a situation, is lessened quite a bit by the text of the
license they received the code under)
> This is a common misunderstanding.
No misunderstanding, really.
Alan seems to have given a bad example that doesn't apply to the situation
that is being discussed.
> Note that you may remove the text of either license from a dual-licensed
> file and redistribute under the other license because neither license
> requires you to retain the other license and both licenses give you the
> right otherwise to modify as you wish. But the removal of a license from a
> file has no effect on the grant of license. Your recipients still get a
> dual license to those protectable elements in the file that were placed
> under a dual license. You cannot stop the automatic grant.
Agreed. When re-distributing an un-modified copy of a work. When distributing
a modified work, the "work" has the license that the person who made the
modifications places on it. But individual files and pieces of code will
still retain their original license - this is how it works.
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
David Schwartz wrote:
> Either license can grant you the right to distribute it, but how you get the
> rights to distribute has *NO* effect on the recipient. They receive a lawful
> copy and any rights the original author grants them under a license from
> that original author. You have no power to grant or modify rights to the
> original work.
Secondary parties have the power to grant or modify rights, if delegated
to them by the original author.
Relicensing and transfer of rights happens all the time. How do you
think most music gets into consumer hands?
Jeff
On Sep 3 2007 04:58, Jeff Garzik wrote:
>
> Relicensing and transfer of rights happens all the time. How do you think
> most music gets into consumer hands?
uh, p2p? :)
Jeff Garzik wrote:
> Secondary parties have the power to grant or modify rights, if
> delegated
> to them by the original author.
Yes, but this has to be done in writing and neither the BSD nor the GPL
license claim to allow this.
> Relicensing and transfer of rights happens all the time. How do you
> think most music gets into consumer hands?
Through written agreements to permit this. Neither the BSD nor the GPL
license do. You cannot revoke someone's GPL rights to code you did not
write because they flow directly from the original author to the recipient.
The GPL is quite clear, you do not grant a license to anyone else's code,
the original author does. This is why the GPL does not need to be in writing.
DS
David Schwartz wrote:
> Jeff Garzik wrote:
>
>> Secondary parties have the power to grant or modify rights, if
>> delegated
>> to them by the original author.
>
> Yes, but this has to be done in writing and neither the BSD nor the GPL
> license claim to allow this.
Standard dual license texts do.
Jeff
> Mr. Floeter *CAN* request that his code be removed from said fork
> - his code
> is solely licensed (AFAICT and IIRC) under the BSD/ISC license
> and was only
> covered by the dual-license because it was integrated into a work that
> carried said dual-license. (I'm not sure how well such a revocation would
> work in reality, but it is Mr. Floeters right.)...
No. Neither the BSD license nor the GPL license permit you to revoke rights.
Mr. Floeter's code is still available under the BSD/ISC license. The BSD
license does not require you to make derived works available under a BSD
license. *His* code is still available under a BSD/ISC license, of course,
but the changed code is not.
Read the BSD license. It does not require changes to be made available under
a compatible license. This is the main difference between the BSD and GPL
licenses.
Note that it would be an error to remove the BSD license text, as the BSD
license requires you to keep it and you still need the BSD license to grant
you distribution rights to the original work. However, the license does not
apply to protectable aspects of the code not placed under the BSD license by
their original author, and it is important to add a note to that effect.
DS
> > Yes, but this has to be done in writing and neither the BSD nor the GPL
> > license claim to allow this.
> Standard dual license texts do.
>
> Jeff
No, they don't. They simply state that *you* may obtain the right to
modify/distribute the work from either license at your option. They do not
say anything about relicensing. Again, such an agreement would have to be in
writing.
What license a distributor uses to acquire the rights to modify or
distribute a work has no effect whatsoever on the rights the recipient has
to the work. The GPL is absolutely clear about this (section 6).
DS
On Monday 03 September 2007 05:48:00 David Schwartz wrote:
> > Mr. Floeter *CAN* request that his code be removed from said fork
> > - his code
> > is solely licensed (AFAICT and IIRC) under the BSD/ISC license
> > and was only
> > covered by the dual-license because it was integrated into a work that
> > carried said dual-license. (I'm not sure how well such a revocation would
> > work in reality, but it is Mr. Floeters right.)...
>
> No. Neither the BSD license nor the GPL license permit you to revoke
> rights. Mr. Floeter's code is still available under the BSD/ISC license.
> The BSD license does not require you to make derived works available under
> a BSD license. *His* code is still available under a BSD/ISC license, of
> course, but the changed code is not.
Doesn't matter if the BSD license or the GPL *PERMITS* it or not. The fact
remains that the person making a work available under *ANY* form of copyright
license has the right to revoke said grant of license to anyone. The GPL
codifies certain situations in which the person would not, personally, have
to revoke the license, but does not limit the original copyright holders
rights (in that regard) in any way.
The BSD/ISC license has none of the automatic conditions of the GPL, but it
also cannot remove the copyright holder(s) from exercising their rights.
(And no, I haven't spoken to a lawyer about this - I did, however, ask a
recently graduated law-school student where I could look for case-law and the
text of the actual laws. What I got was some background on US copyright law
itself and an agreement that a copyright license does not - and can not -
affect the person holding the copyright)
> Read the BSD license. It does not require changes to be made available
> under a compatible license. This is the main difference between the BSD and
> GPL licenses.
Have done so. And that is the only part of the license that I actually don't
like.
> Note that it would be an error to remove the BSD license text, as the BSD
> license requires you to keep it and you still need the BSD license to grant
> you distribution rights to the original work. However, the license does not
> apply to protectable aspects of the code not placed under the BSD license
> by their original author, and it is important to add a note to that effect.
Agreed, and I've never claimed otherwise. (Nor has anyone else. I believe the
closest that anyone has come was Alan Cox saying (and I'm going to paraphrase
it because I don't think he ever stated it well) "If you've made changes to a
file that carries a dual BSD/GPL license and your changes are GPL only, the
file can no longer be distributed under the BSD license at all. So it is safe
to remove the headers from that individual file."
There is no way that a license on a constituent file can alter or affect the
license on the whole project (if it is different). It can "muddy the waters",
but that is about as far as I can see it going.
(I realize I may have said different, originally, but you'll have to forgive
me. I was not in the best of moods (or the best state of mind) to be making a
completely rational argument when I did such.)
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Daniel Hazelton <[email protected]> writes:
> The fact
> remains that the person making a work available under *ANY* form of
> copyright
> license has the right to revoke said grant of license to anyone.
Not after the licence has been given and accepted (and there might be
restrictions), unless of course the licence contained such reservation.
--
Krzysztof Halasa
On Monday 03 September 2007 14:26:29 Krzysztof Halasa wrote:
> Daniel Hazelton <[email protected]> writes:
> > The fact
> > remains that the person making a work available under *ANY* form of
> > copyright
> > license has the right to revoke said grant of license to anyone.
>
> Not after the licence has been given and accepted (and there might be
> restrictions), unless of course the licence contained such reservation.
I hate to belabor the point, but you seem to be making the mistake of "The
license applies to the copyright holder" that I've seen a lot of people make
(and kept quiet about).
The person holding the copyright has all the legal standing to revoke a
license grant at any time. Licenses such as the GPL are not signed contracts,
and that means there are limits to what effect they can have on the copyright
holder.
If the license was of the "signed contract" type, or contained text stating
that the copyright holder was giving up all rights of revocation (etc...) I
could agree with you. As it stands, no "Open Source" license that I have seen
used on a major project contains any part that does that. In fact, the GPL is
the only license I can name (offhand) that even touches on the rights of the
copyright holder - and then it is in the form of "If you do X, Y or Z all
rights granted under this license are automatically revoked".
That is an "automatic clause" - not a limitation stating that the copyright
holder can only revoke under those conditions. The person holding the
copyright has quite a few rights - more than people believe - and not even
the most generous of Open Source licenses (except those that contain text
like "granted in perpetuity" or similar) even come close to being exempt from
the holder of the copyright not being able to summarily revoke a given
persons license.
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Daniel Hazelton <[email protected]> writes:
> I hate to belabor the point, but you seem to be making the mistake of "The
> license applies to the copyright holder"
Of course not.
> The person holding the copyright has all the legal standing to revoke a
> license grant at any time.
Based on?
> Licenses such as the GPL are not signed contracts,
> and that means there are limits to what effect they can have on the
> copyright
> holder.
Perhaps that is your local laws, but I'd be surprised anyway.
Do you sign contracts when shopping, or are you (and the shop)
allowed to "revoke" the transaction after it's made (I'm not
talking about shop's return policy)?
Is what you wrote only valid WRT licences?
Which country has such laws BTW?
In Poland, I can't just go and "revoke" a "statement of will"
if I haven't explicite reserved a right to do so.
Obviously I can act contrary to the statement, and be held liable.
--
Krzysztof Halasa
On Monday 03 September 2007 15:33:01 Krzysztof Halasa wrote:
> Daniel Hazelton <[email protected]> writes:
> > I hate to belabor the point, but you seem to be making the mistake of
> > "The license applies to the copyright holder"
>
> Of course not.
I'll take this at face value - I might have mis-parsed your earlier statements
wrong.
> > The person holding the copyright has all the legal standing to revoke a
> > license grant at any time.
>
> Based on?
US Copyright law. A copyright holder, regardless of what license he/she may
have released the work under, can still revoke the license for a specific
person or group of people. (There are some exceptions, but they do not apply
to the situation that is being discussed)
> > Licenses such as the GPL are not signed contracts,
> > and that means there are limits to what effect they can have on the
> > copyright
> > holder.
>
> Perhaps that is your local laws, but I'd be surprised anyway.
I wouldn't. The US Legal system is rather twisted.
> Do you sign contracts when shopping, or are you (and the shop)
> allowed to "revoke" the transaction after it's made (I'm not
> talking about shop's return policy)?
A purchase is separate from a grant of copyright under license. If I purchase
a book I have the right to read the book and I have the right to sell the
book to someone else, but I have no other rights to it. But if I purchase
something, there are rights that go along with said purchase.
Under a license such as the GPL (I can't say the same for the BSD/ISC
license - I haven't spent enough time studying it to know for sure) no money
need change hands for access to the program *and* source. All the rights that
anyone *BESIDES* the copyright holder have to the program and/or source comes
from license. But since money has not changed hands, there is no further set
of rights or guarantees - the copyright holder has not, in general, sold a
copy of the work or granted any guarantee that the license will not be
revoked.
> Is what you wrote only valid WRT licences?
Yes. For contracts there is slightly different set of laws in play.
> Which country has such laws BTW?
The USA.
>
> In Poland, I can't just go and "revoke" a "statement of will"
> if I haven't explicite reserved a right to do so.
> Obviously I can act contrary to the statement, and be held liable.
Ah, see - in the US the license(s) in question (and licenses in general) are
grants of rights, not a "statements of will". If there are to be any limits
on the rights of the copyright holder, under US law, they have to be
explicitly stated in the license itself. (Truthfully, in the US a license
should be read with an implicit "All rights reserved")
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Daniel Hazelton wrote:
> On Monday 03 September 2007 14:26:29 Krzysztof Halasa wrote:
>> Daniel Hazelton <[email protected]> writes:
>>> The fact
>>> remains that the person making a work available under *ANY* form of
>>> copyright
>>> license has the right to revoke said grant of license to anyone.
>> Not after the licence has been given and accepted (and there might be
>> restrictions), unless of course the licence contained such reservation.
>
> I hate to belabor the point, but you seem to be making the mistake of "The
> license applies to the copyright holder" that I've seen a lot of people make
> (and kept quiet about).
I believe you are making the mistake that the license on code has
anything to do with what the author chooses to do in the future.
Releasing something as BSD does not force the author to do anything in
the future with his code, and he/she could add and relicence as he/she
feels fit. HOWEVER, that particular code has already been released as
BSD, and the author no longer has control over that release.
> The person holding the copyright has all the legal standing to revoke a
> license grant at any time. Licenses such as the GPL are not signed contracts,
> and that means there are limits to what effect they can have on the copyright
> holder.
I believe you are confusing the fact that an author can decide to
release code under another license, with the existence of code under
that earlier license. The license grant comes from THE CODE (which
bears a license), not THE AUTHOR. I can use GPL code I get in the mail
because the license on the work says I can do so, not because I
contacted the author and got a specific grant. If such a grant were
only verbal, your theory might hold, but that doesn't apply to any OSS
software under discussion here.
If your legal theory were true, I could sell you a book and then later
demand that you destroy it. I could also release something as public
domain, and then later rescind that (I still hold the copyright on what
I produced), and charge money from anyone who used it. I think its safe
to say that this does not happen in practice. Please provide some
examples to the contrary or caselaw if you want to convince me otherwise.
Furthermore, BSD/GPL software could not really exist under your legal
theory; A programmer who wrote 30 year old core BSD code could wake up
tomorrow and decide to require all BSD derivatives to remove his code or
pay him for it (and the next day he could change the price again). Open
source software would not exist if such a liability were true, and
companies like Sun could not be built up off of derivatives of it.
Linux 0.01 is still available under a pre-GPL license if you can find a
copy, and neither Linus (nor anyone else) can change that.
> If the license was of the "signed contract" type, or contained text stating
> that the copyright holder was giving up all rights of revocation (etc...) I
> could agree with you. As it stands, no "Open Source" license that I have seen
> used on a major project contains any part that does that. In fact, the GPL is
> the only license I can name (offhand) that even touches on the rights of the
> copyright holder - and then it is in the form of "If you do X, Y or Z all
> rights granted under this license are automatically revoked".
>
> That is an "automatic clause" - not a limitation stating that the copyright
> holder can only revoke under those conditions. The person holding the
> copyright has quite a few rights - more than people believe - and not even
> the most generous of Open Source licenses (except those that contain text
> like "granted in perpetuity" or similar) even come close to being exempt from
> the holder of the copyright not being able to summarily revoke a given
> persons license.
There are plenty of rights, but retroactive changes to the license terms
of something you've already distributed is not one of them.
- Jim Bruce
Daniel Hazelton <[email protected]> writes:
> US Copyright law. A copyright holder, regardless of what license he/she may
> have released the work under, can still revoke the license for a specific
> person or group of people. (There are some exceptions, but they do not apply
> to the situation that is being discussed)
Oh come on, I thought some small country in maybe central Africa,
but certainly not USA.
What you write would essentially mean GPL (and any other such licence)
is invalid in the USA.
The licence is basically a promise not to sue. It wouldn't make any
sense to promise if you could revoke at will.
> Ah, see - in the US the license(s) in question (and licenses in general) are
> grants of rights, not a "statements of will".
Right here grants of rights are some sort of statements of will.
> (Truthfully, in the US a license
> should be read with an implicit "All rights reserved")
Actually (and I think it's the same in the USA), a copyrighted work
has an implicit "all rights reserved". A licence is just exception.
--
Krzysztof Halasa
On Tuesday 04 September 2007 04:50:34 James Bruce wrote:
> Daniel Hazelton wrote:
> > On Monday 03 September 2007 14:26:29 Krzysztof Halasa wrote:
> >> Daniel Hazelton <[email protected]> writes:
> >>> The fact
> >>> remains that the person making a work available under *ANY* form of
> >>> copyright
> >>> license has the right to revoke said grant of license to anyone.
> >>
> >> Not after the licence has been given and accepted (and there might be
> >> restrictions), unless of course the licence contained such reservation.
> >
> > I hate to belabor the point, but you seem to be making the mistake of
> > "The license applies to the copyright holder" that I've seen a lot of
> > people make (and kept quiet about).
>
> I believe you are making the mistake that the license on code has
> anything to do with what the author chooses to do in the future.
> Releasing something as BSD does not force the author to do anything in
> the future with his code, and he/she could add and relicence as he/she
> feels fit. HOWEVER, that particular code has already been released as
> BSD, and the author no longer has control over that release.
I may be mistaken, but it has always been my understanding that, unless you
specifically waive your rights, they are automatically retained. (Under the
law in the US, at least).
Hence, a copyright holder can do such, where the license has not been acquired
by money changing hands.
(And actually, my above statement isn't rendered false by your rebuttal - it
still appears that the person I replied to believes that a copyright license
applies to the person holding the copyright in the same manner it applies to
the person receiving the item under said license. Though I will admit it if I
am wrong - publicly)
> > The person holding the copyright has all the legal standing to revoke a
> > license grant at any time. Licenses such as the GPL are not signed
> > contracts, and that means there are limits to what effect they can have
> > on the copyright holder.
>
> I believe you are confusing the fact that an author can decide to
> release code under another license, with the existence of code under
> that earlier license. The license grant comes from THE CODE (which
> bears a license), not THE AUTHOR. I can use GPL code I get in the mail
> because the license on the work says I can do so, not because I
> contacted the author and got a specific grant. If such a grant were
> only verbal, your theory might hold, but that doesn't apply to any OSS
> software under discussion here.
The license is a direct grant from the author. If the author so wished, he/she
could pull the license - either entirely or in part. About the only caveat is
that the author would have to publish and attempt to contact everyone who may
have acquired the item under that license to inform them of such a change -
this does make it difficult, hell, makes it nearly impossible, but it can be
done. (IANAL, but this does appear to be what the law says)
> If your legal theory were true, I could sell you a book and then later
> demand that you destroy it. I could also release something as public
> domain, and then later rescind that (I still hold the copyright on what
> I produced), and charge money from anyone who used it. I think its safe
> to say that this does not happen in practice. Please provide some
> examples to the contrary or caselaw if you want to convince me otherwise.
Actually, no. A purchase does automatically grant the rights inherent in
ownership - but that is a *PURCHASE*. Mere transfer of an item with no
exchange of money cannot convey those rights. As far as the 'public domain'
argument goes... That smells of a straw-man and is as different from a grant
of license as it is from a purchase. When you release something into the
public domain you are waiving *ALL* of your rights as copyright holder.
(Which, I am told, cannot be done in Germany and some other countries)
> Furthermore, BSD/GPL software could not really exist under your legal
> theory; A programmer who wrote 30 year old core BSD code could wake up
> tomorrow and decide to require all BSD derivatives to remove his code or
> pay him for it (and the next day he could change the price again). Open
> source software would not exist if such a liability were true, and
> companies like Sun could not be built up off of derivatives of it.
> Linux 0.01 is still available under a pre-GPL license if you can find a
> copy, and neither Linus (nor anyone else) can change that.
He could, but AFAICT, thirty years ago BSD was still run entirely by UC
Berkely and any copyrights that might be held are held entirely by UC Berkely
and not the individuals that contributed to such. (Whats more, a 30 year old
version of BSD doesn't meet the requirements of the AT&T agreement, so its
only legal in-so-far as it massively predates that agreement (and the lawsuit
which spawned it) :)
And yes, Linus actually could revoke the license on any copy of Linux from
before he started merging code written by other people into the code-base.
After the first time he merged another persons code he lost that unilateral
right. (He could, still, revoke the right to *his* portions of the code,
however)
> > If the license was of the "signed contract" type, or contained text
> > stating that the copyright holder was giving up all rights of revocation
> > (etc...) I could agree with you. As it stands, no "Open Source" license
> > that I have seen used on a major project contains any part that does
> > that. In fact, the GPL is the only license I can name (offhand) that even
> > touches on the rights of the copyright holder - and then it is in the
> > form of "If you do X, Y or Z all rights granted under this license are
> > automatically revoked".
> >
> > That is an "automatic clause" - not a limitation stating that the
> > copyright holder can only revoke under those conditions. The person
> > holding the copyright has quite a few rights - more than people believe -
> > and not even the most generous of Open Source licenses (except those that
> > contain text like "granted in perpetuity" or similar) even come close to
> > being exempt from the holder of the copyright not being able to summarily
> > revoke a given persons license.
>
> There are plenty of rights, but retroactive changes to the license terms
> of something you've already distributed is not one of them.
I'm not talking about retro-active changes. I'm talking about the copyright
holder exercising their right to cancel said license in regards to either a
specific person, a specific group of people or everyone. Cancelling a license
is not altering the terms - it's stating "I am removing this product from
release under this license entirely".
(Though I have never said doing such is *EASY*, and I never said that doing
such would be free from people throwing lawsuits around. It's because of
those two facts that it has only rarely ever been done - even by large
companies.)
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
On Tuesday 04 September 2007 09:27:02 Krzysztof Halasa wrote:
> Daniel Hazelton <[email protected]> writes:
> > US Copyright law. A copyright holder, regardless of what license he/she
> > may have released the work under, can still revoke the license for a
> > specific person or group of people. (There are some exceptions, but they
> > do not apply to the situation that is being discussed)
>
> Oh come on, I thought some small country in maybe central Africa,
> but certainly not USA.
US Law is a twisted maze - you wouldn't believe the contradictions that exist
between different sections of the US Federal Code. (And its worse as you move
down to the State and the Local levels)
> What you write would essentially mean GPL (and any other such licence)
> is invalid in the USA.
Nope. The GPL is an explicit grant of rights and is fully legal and active as
it stands.
> The licence is basically a promise not to sue. It wouldn't make any
> sense to promise if you could revoke at will.
If I was to revoke the license on something I held copyright to, I'd be forced
to make an attempt to contact everyone that may have received a copy of the
work under that license before I could ever begin filing lawsuits. This
process will take at least a month - more if the various localities where
someone might be living has laws about what constitutes an attempt to
contact. (For instance, here in Pennsylvania an attempt to contact is taking
out large format classified ad's in every newspaper in the area where the
person is known to reside - or statewide if the region is not known. The ad's
have to run for a minimum of one week)
This means that it'd take no less than five weeks - and might take as much as
six months - before I could begin filing lawsuits. (And even then I'd have to
have proof that the person in question was violating my copyright at the time
the lawsuit was filed)
> > Ah, see - in the US the license(s) in question (and licenses in general)
> > are grants of rights, not a "statements of will".
>
> Right here grants of rights are some sort of statements of will.
Difference in terminology ?
A "Grant of Rights" is where you say 'Normally only I could do this, but I am
giving you the legal right to do it as well'. A "statement of will" is 'This
is what I want to have happen, in perpetuity'. In the US, a "statement of
will" can include or imply a "Grant of Rights" and vice-versa, but they are
separate entities.
> > (Truthfully, in the US a license
> > should be read with an implicit "All rights reserved")
>
> Actually (and I think it's the same in the USA), a copyrighted work
> has an implicit "all rights reserved". A licence is just exception.
And? The fact remains that "All Rights Reserved" means "I am reserving all
rights I do not specifically grant or waive". ie: If a license doesn't
state 'The licenser hereby waives the right to revoke this license at any
time' then that right hasn't been lost. (A license acquired through a
purchase - as might apply to a novel - is a lot different. And contracts are
a different beast entirely)
DRH
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Daniel Hazelton wrote:
> On Tuesday 04 September 2007 09:27:02 Krzysztof Halasa wrote:
>
>>Daniel Hazelton <[email protected]> writes:
>>
>>>US Copyright law. A copyright holder, regardless of what license he/she
>>>may have released the work under, can still revoke the license for a
>>>specific person or group of people. (There are some exceptions, but they
>>>do not apply to the situation that is being discussed)
The OpenBSD policy page doesn't agree with you:
"...That means that having granted a permission, the copyright holder
can not retroactively say that an individual or class of individuals are
no longer granted those permissions. Likewise should the copyright
holder decide to "go commercial" he can not revoke permissions already
granted for the use of the work as distributed, though he may impose
more restrictive permissions in his future distributions of that work."
http://www.openbsd.org/policy.html
Chris
Chris Friesen writes:
> Daniel Hazelton wrote:
>> On Tuesday 04 September 2007 09:27:02 Krzysztof Halasa wrote:
>>
>>>Daniel Hazelton <[email protected]> writes:
>>>
>>>>US Copyright law. A copyright holder, regardless of what license he/she
>>>>may have released the work under, can still revoke the license for a
>>>>specific person or group of people. (There are some exceptions, but they
>>>>do not apply to the situation that is being discussed)
>
> The OpenBSD policy page doesn't agree with you:
>
> "...That means that having granted a permission, the copyright holder
> can not retroactively say that an individual or class of individuals
> are no longer granted those permissions. Likewise should the copyright
> holder decide to "go commercial" he can not revoke permissions already
> granted for the use of the work as distributed, though he may impose
> more restrictive permissions in his future distributions of that work."
>
> http://www.openbsd.org/policy.html
By my reading, this is supported by 17 USC 203(a)(3):
(3) Termination of the grant may be effected at any time during a
period of five years beginning at the end of thirty-five years
from the date of execution of the grant; or, if the grant covers
the right of publication of the work, the period begins at the
end of thirty-five years from the date of publication of the
work under the grant or at the end of forty years from the date
of execution of the grant, whichever term ends earlier.
(from http://www.law.cornell.edu/uscode/html/uscode17/usc_sec_17_00000203----000-.html )
I would be interested to see what other legal basis is alleged as
grounds to rescind a license.
Michael
On Tue, 4 Sep 2007, Chris Friesen wrote:
> Daniel Hazelton wrote:
>> On Tuesday 04 September 2007 09:27:02 Krzysztof Halasa wrote:
>>
>>> Daniel Hazelton <[email protected]> writes:
>>>
>>>> US Copyright law. A copyright holder, regardless of what license he/she
>>>> may have released the work under, can still revoke the license for a
>>>> specific person or group of people. (There are some exceptions, but they
>>>> do not apply to the situation that is being discussed)
>
> The OpenBSD policy page doesn't agree with you:
>
> "...That means that having granted a permission, the copyright holder
> can not retroactively say that an individual or class of individuals are
> no longer granted those permissions. Likewise should the copyright
> holder decide to "go commercial" he can not revoke permissions already
> granted for the use of the work as distributed, though he may impose
> more restrictive permissions in his future distributions of that work."
>
> http://www.openbsd.org/policy.html
>
>
> Chris
> -
There are other enforceability issues as well. For instance in the
US, Copyright Law applies as soon as something is written. So,
does Copyright Law apply if I write, "You cannot read this."
Of course, it's a trivial example. Revocation of a license to
read a work is absurd. Using this theory, once somebody's
written "work" has been distributed under some license, a
different license would likely be regarded as unenforceable
by a court.
Cheers,
Dick Johnson
Penguin : Linux version 2.6.22.1 on an i686 machine (5588.30 BogoMips).
My book : http://www.AbominableFirebug.com/
_
****************************************************************
The information transmitted in this message is confidential and may be privileged. Any review, retransmission, dissemination, or other use of this information by persons or entities other than the intended recipient is prohibited. If you are not the intended recipient, please notify Analogic Corporation immediately - by replying to this message or by sending an email to [email protected] - and destroy all copies of this information, including any attachments, without reading or disclosing them.
Thank you.
On Tuesday 04 September 2007 15:44:31 Michael Poole wrote:
> Chris Friesen writes:
> > Daniel Hazelton wrote:
> >> On Tuesday 04 September 2007 09:27:02 Krzysztof Halasa wrote:
> >>>Daniel Hazelton <[email protected]> writes:
> >>>>US Copyright law. A copyright holder, regardless of what license he/she
> >>>>may have released the work under, can still revoke the license for a
> >>>>specific person or group of people. (There are some exceptions, but
> >>>> they do not apply to the situation that is being discussed)
> >
> > The OpenBSD policy page doesn't agree with you:
> >
> > "...That means that having granted a permission, the copyright holder
> > can not retroactively say that an individual or class of individuals
> > are no longer granted those permissions. Likewise should the copyright
> > holder decide to "go commercial" he can not revoke permissions already
> > granted for the use of the work as distributed, though he may impose
> > more restrictive permissions in his future distributions of that work."
> >
> > http://www.openbsd.org/policy.html
>
> By my reading, this is supported by 17 USC 203(a)(3):
>
> (3) Termination of the grant may be effected at any time during a
> period of five years beginning at the end of thirty-five years
> from the date of execution of the grant; or, if the grant covers
> the right of publication of the work, the period begins at the
> end of thirty-five years from the date of publication of the
> work under the grant or at the end of forty years from the date
> of execution of the grant, whichever term ends earlier.
>
> (from
> http://www.law.cornell.edu/uscode/html/uscode17/usc_sec_17_00000203----000-
>.html )
Ah, I am both right and wrong, it seems. Apparently you have to wait anywhere
form 35 to 40 years, and then you only have a five year window. Seems damned
strange to me, but oh well.
(I'd totally forgotten that part of the law - or my mind decided to play
tricks on me.)
DRH
PS: See, I will admit it when I'm shown evidence that I'm wrong :)
--
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Daniel Hazelton <[email protected]> writes:
>> Actually (and I think it's the same in the USA), a copyrighted work
>> has an implicit "all rights reserved". A licence is just exception.
>
> And? The fact remains that "All Rights Reserved" means "I am reserving all
> rights I do not specifically grant or waive".
Sure.
That means you can't use a work without obtaining a licence first
(unless you are the copyright holder and don't need (can't have)
the licence).
I.e., as the author, you don't have to write those "copyright xxx"
and the work is still protected.
> ie: If a license doesn't
> state 'The licenser hereby waives the right to revoke this license at any
> time' then that right hasn't been lost.
Yes, it hasn't been *lost*. It just never existed :-)
(with exception of those things starting 40 or so years after the
original publication, I think we can safely skip this part).
Obviously, the copyright holder is free to distribute the work
under a different licence next time (but (s)he has to make sure it's
ok with licences (s)he uses, if it's a derived work).
--
Krzysztof Halasa
On Mon 2007-09-03 04:58:58, Jeff Garzik wrote:
> David Schwartz wrote:
> >Either license can grant you the right to distribute
> >it, but how you get the
> >rights to distribute has *NO* effect on the recipient.
> >They receive a lawful
> >copy and any rights the original author grants them
> >under a license from
> >that original author. You have no power to grant or
> >modify rights to the
> >original work.
>
> Secondary parties have the power to grant or modify
> rights, if delegated to them by the original author.
>
> Relicensing and transfer of rights happens all the time.
> How do you think most music gets into consumer hands?
License is only a promise not to sue. Only original author is permitted to
sue in BSD/GPL case. David seems right here.
If Linus releases GPL/BSD program, you choose GPL and distribute it to
me, and I put it into proprietary evil app, I am _ok_.
You can't sue me, because you are not copyright holder.
Linus can't sue me, because he BSD licensed it.
=> I'm fine. Yep, copyright law is strange, and you may be right
outside U.S.
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html