GPL V3 and Linux

GPL V3 and Linux

[Jeff V. Merkey]

Cudos to Stallman, The patent retaliation clause is exactly what has 
been missing.  The inclusion of custom binaries was a little vague, but
the net of it is that the end user can combine the separate parts, and 
have the freedom to do so given the GPL3 terms.  Any concensus on 
whether Linux will move to GPL3?   I support adoption and congrats to 
Stallman -- A++++.

Jeff

Re: GPL V3 and Linux

[Stephen Hemminger]

On Fri, 20 Jan 2006 09:49:44 -0700
"Jeff V. Merkey" <jmerkey@wolfmountaingroup.com> wrote:

> Cudos to Stallman, The patent retaliation clause is exactly what has 
> been missing.  The inclusion of custom binaries was a little vague, but
> the net of it is that the end user can combine the separate parts, and 
> have the freedom to do so given the GPL3 terms.  Any concensus on 
> whether Linux will move to GPL3?   

No consensus exists, and it would require agreement from all the copyright
holders.

-- 
Stephen Hemminger <shemminger@osdl.org>
OSDL http://developer.osdl.org/~shemminger

Re: GPL V3 and Linux

[Patrick McLean]

Stephen Hemminger wrote:
> On Fri, 20 Jan 2006 09:49:44 -0700
> "Jeff V. Merkey" <jmerkey@wolfmountaingroup.com> wrote:
> 
>> Cudos to Stallman, The patent retaliation clause is exactly what has 
>> been missing.  The inclusion of custom binaries was a little vague, but
>> the net of it is that the end user can combine the separate parts, and 
>> have the freedom to do so given the GPL3 terms.  Any concensus on 
>> whether Linux will move to GPL3?   
> 
> No consensus exists, and it would require agreement from all the copyright
> holders.
> 

I don't think the kernel is going to move to v3, it's licensed 
specifically as v2, this is from the top of COPYING:

 >  Also note that the only valid version of the GPL as far as the kernel
 >  is concerned is _this_ particular version of the license (ie v2, not
 >  v2.2 or v3.x or whatever), unless explicitly otherwise stated.

Also, given that several of the copyright holders in the kernel are 
dead, I don't think we will be able to obtain permission.

Re: GPL V3 and Linux

[Jeff V. Merkey]

Patrick McLean wrote:

> Stephen Hemminger wrote:
>
>> On Fri, 20 Jan 2006 09:49:44 -0700
>> "Jeff V. Merkey" <jmerkey@wolfmountaingroup.com> wrote:
>>
>>> Cudos to Stallman, The patent retaliation clause is exactly what has 
>>> been missing.  The inclusion of custom binaries was a little vague, but
>>> the net of it is that the end user can combine the separate parts, 
>>> and have the freedom to do so given the GPL3 terms.  Any concensus 
>>> on whether Linux will move to GPL3?   
>>
>>
>> No consensus exists, and it would require agreement from all the 
>> copyright
>> holders.
>>
>
> I don't think the kernel is going to move to v3, it's licensed 
> specifically as v2, this is from the top of COPYING:
>
> >  Also note that the only valid version of the GPL as far as the kernel
> >  is concerned is _this_ particular version of the license (ie v2, not
> >  v2.2 or v3.x or whatever), unless explicitly otherwise stated.
>
> Also, given that several of the copyright holders in the kernel are 
> dead, I don't think we will be able to obtain permission.


I can do a ceremony and call them with an eagle bone whistle and a 
Califonia Condor Feather.  We can then ask them directly.
GPL2 is fine if the kernel stays that way for my projects.  moving 
forward, the patent retaliation clause is a great idea. 

Jeff

>
> -
> To unsubscribe from this list: send the line "unsubscribe 
> linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: GPL V3 and Linux

[Jeff V. Merkey]

>>
>> Also, given that several of the copyright holders in the kernel are 
>> dead, I don't think we will be able to obtain permission.
>
>
>
> I can do a ceremony and call them with an eagle bone whistle and a 
> Califonia Condor Feather.  We can then ask them directly.
> GPL2 is fine if the kernel stays that way for my projects.  moving 
> forward, the patent retaliation clause is a great idea.
> Jeff


I called them at the sacred fire in my ceremony room with music from the 
bones of an eagle.  They said whatever concensus is reached
by the majority if fine with them.     One of them said something about 
a woman wearing a crazy hat with a white ostrich feather in it
he did not care for, but that was it.


Jeff

>
>>
>> -
>> To unsubscribe from this list: send the line "unsubscribe 
>> linux-kernel" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at  http://www.tux.org/lkml/
>>

Re: GPL V3 and Linux

[Alan Cox]

On Gwe, 2006-01-20 at 14:34 -0500, Patrick McLean wrote:
> I don't think the kernel is going to move to v3, it's licensed 
> specifically as v2, this is from the top of COPYING:

It may well do, or bits of it may well do but it is rather early to
speculate.

>  >  Also note that the only valid version of the GPL as far as the kernel
>  >  is concerned is _this_ particular version of the license (ie v2, not
>  >  v2.2 or v3.x or whatever), unless explicitly otherwise stated.
> 
> Also, given that several of the copyright holders in the kernel are 
> dead, I don't think we will be able to obtain permission.

It isn't clear that this will be a problem. Very few people specifically
put their code v2 only, and Linus edit of the top copying file was not
done with permission of other copyright holders anyway so really only
affects his code if it is valid at all.

What finally happens is going to depend almost entirely on whether the
GPL v3 is a sane license or not and on consensus, and it is *way* too
early to figure that out.

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Patrick McLean wrote:
> Stephen Hemminger wrote:
>
>
> Also, given that several of the copyright holders in the kernel are 
> dead, I don't think we will be able to obtain permission.
>
>

Makes me wonder if something should be done to address the issue of dead 
copyright holders. Not sure what but maybe there should be a clause in 
GPL3 addressing that?

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Marc Perkel wrote:

>
>
> Patrick McLean wrote:
>
>> Stephen Hemminger wrote:
>>
>>
>> Also, given that several of the copyright holders in the kernel are 
>> dead, I don't think we will be able to obtain permission.
>>
>>
>
> Makes me wonder if something should be done to address the issue of 
> dead copyright holders. Not sure what but maybe there should be a 
> clause in GPL3 addressing that?


Their heirs would have two years to bring a cause of action if they 
object.  Proper notice could be served by posting a notice on the 
internet at kernel.org
that their code is being redistributed under GPL3.  I note that the 
general notice in the code states "GPL2 or any later version of the 
license".  Given this
language, it is highly likely the remaining code can proceed under a new 
license without incident since it can be assumed they already agreed by
having this general notice posted at kernel.org for many years.   I 
think the point is moot.  Legally, there is exposure if their successors 
or owners
of their estates bring action.  Those outside the US would of course be 
subject to the laws of their jurisdiction.  An attorney at FSF needs to 
review their
code and render an opinion, but I think it will not be a problem.

Jeff

>
> -
> To unsubscribe from this list: send the line "unsubscribe 
> linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

>> Makes me wonder if something should be done to address the issue of 
>> dead copyright holders. Not sure what but maybe there should be a 
>> clause in GPL3 addressing that?
>
>
>
> Their heirs would have two years to bring a cause of action if they 
> object.  Proper notice could be served by posting a notice on the 
> internet at kernel.org
> that their code is being redistributed under GPL3.  I note that the 
> general notice in the code states "GPL2 or any later version of the 
> license".  Given this
> language, it is highly likely the remaining code can proceed under a 
> new license without incident since it can be assumed they already 
> agreed by
> having this general notice posted at kernel.org for many years.   I 
> think the point is moot.  Legally, there is exposure if their 
> successors or owners
> of their estates bring action.  Those outside the US would of course 
> be subject to the laws of their jurisdiction.  An attorney at FSF 
> needs to review their
> code and render an opinion, but I think it will not be a problem.
>
> Jeff
>
>>
NOTE:  Under the Doctrine of Esstopel, if you proceed on this basis and 
two years pass without their heirs bringing an action of some sort, then 
under this
legal doctrine, the rights to use their code under GPLv3 would in all 
probability pass consitutional muster.  Again, someone needs to run this 
by an attorney at the FSF and get a formal legal opinion rendered.  The 
Doctrine of Esstopel basically says that if you use something for some 
period of time, and no one
objects, then you obtain certain rights to use it permenantly.  Not 
wanting to disrespect the wishes of the dead, I would attempt to contact 
the successors of
their estates in any event and obtain permission, and if not possible, 
then proceed.

Jeff

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Jeff V. Merkey wrote:
>
>>> Makes me wonder if something should be done to address the issue of 
>>> dead copyright holders. Not sure what but maybe there should be a 
>>> clause in GPL3 addressing that?
>>
>>
>>
>> Their heirs would have two years to bring a cause of action if they 
>> object.  Proper notice could be served by posting a notice on the 
>> internet at kernel.org
>> that their code is being redistributed under GPL3.  I note that the 
>> general notice in the code states "GPL2 or any later version of the 
>> license".  Given this
>> language, it is highly likely the remaining code can proceed under a 
>> new license without incident since it can be assumed they already 
>> agreed by
>> having this general notice posted at kernel.org for many years.   I 
>> think the point is moot.  Legally, there is exposure if their 
>> successors or owners
>> of their estates bring action.  Those outside the US would of course 
>> be subject to the laws of their jurisdiction.  An attorney at FSF 
>> needs to review their
>> code and render an opinion, but I think it will not be a problem.
>>
>> Jeff
>>
>>>
> NOTE:  Under the Doctrine of Esstopel, if you proceed on this basis 
> and two years pass without their heirs bringing an action of some 
> sort, then under this
> legal doctrine, the rights to use their code under GPLv3 would in all 
> probability pass consitutional muster.  Again, someone needs to run 
> this by an attorney at the FSF and get a formal legal opinion 
> rendered.  The Doctrine of Esstopel basically says that if you use 
> something for some period of time, and no one
> objects, then you obtain certain rights to use it permenantly.  Not 
> wanting to disrespect the wishes of the dead, I would attempt to 
> contact the successors of
> their estates in any event and obtain permission, and if not possible, 
> then proceed.
>
> Jeff

Is it possible to have Linux be mostly GPL3 with parts of it GPL2? Or is 
that just too insane to deal with?

Re: GPL V3 and Linux - Dead Copyright Holders

[Kyle Moffett]

On Jan 25, 2006, at 14:24:13, Marc Perkel wrote:
> Is it possible to have Linux be mostly GPL3 with parts of it GPL2?  
> Or is that just too insane to deal with?

Well given that parts of the kernel are GPLv2-only, other parts are  
GPLv2+, other parts are GPL/BSD, etc, I can't see how somebody using  
a GPLv3-only or GPLv3+ license for some other part would be  
problematic.  If anything, the multiple licensing provides additional  
code protection; we get the advantages of all the licenses, but if  
any one license is found to be invalid, it does not break the  
protection of the body of code itself.

Cheers,
Kyle Moffett

Re: GPL V3 and Linux - Dead Copyright Holders

[linux-os (Dick Johnson)]

On Wed, 25 Jan 2006, Kyle Moffett wrote:

> On Jan 25, 2006, at 14:24:13, Marc Perkel wrote:
>> Is it possible to have Linux be mostly GPL3 with parts of it GPL2?
>> Or is that just too insane to deal with?
>
> Well given that parts of the kernel are GPLv2-only, other parts are
> GPLv2+, other parts are GPL/BSD, etc, I can't see how somebody using
> a GPLv3-only or GPLv3+ license for some other part would be
> problematic.  If anything, the multiple licensing provides additional
> code protection; we get the advantages of all the licenses, but if
> any one license is found to be invalid, it does not break the
> protection of the body of code itself.
>
> Cheers,
> Kyle Moffett
>

The original GPL said something about:
"You may not impose any further restrictions on the recipients'
exercise of the rights granted herein." (Section 6).
Then, that __exact__ code was redistributed under Version 2
which further restricted rights, then additional versions
which further restricted rights. Now you are planning to
add additional restrictions? I don't think the present
so-called license would pass muster in any sane court in
the United States after the original licensed code was
plagiarized into a new binding license.

Simple test. Pretend the code was a music chart. Music
charts have been copyrighted since the start of the
copyright office. You write some music and, in its
copyright notice, you license anybody to use it as
long as they don't claim that they wrote it. Then
some licensing agency comes along and writes a new
license, effectively claiming ownership by claiming
control (the legal word is conversion). Do you think
for a moment that any court of law would uphold the
new license?

All of Linux has undergone such a conversion and it is
effectively owned by the "Free Software Foundation, Inc."
Of course RMS didn't tell you this when he appropriated
it, but it's done.

If code was written to be distributed under a certain
set of rules, just like sheet-music, nobody but the
writer or his assigns is allowed to change those
distribution rules at a later date. If those rules
are changed, they are invalid, i.e., unenforceable.

You want new rules, you rewrite the kernel from scratch
under the new rules and, you must not produce a derived
work (which has many meanings) in the process or the
new license is unenforceable as well.


Cheers,
Dick Johnson
Penguin : Linux version 2.6.13.4 on an i686 machine (5589.66 BogoMips).
Warning : 98.36% of all statistics are fiction.
.

****************************************************************
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 DeliveryErrors@analogic.com - and destroy all copies of this information, including any attachments, without reading or disclosing them.

Thank you.

Re: GPL V3 and Linux - Dead Copyright Holders

[Chase Venters]

On Wed, 25 Jan 2006, linux-os \(Dick Johnson\) wrote:
>
> The original GPL said something about:
> "You may not impose any further restrictions on the recipients'
> exercise of the rights granted herein." (Section 6).
> Then, that __exact__ code was redistributed under Version 2
> which further restricted rights, then additional versions
> which further restricted rights. Now you are planning to
> add additional restrictions? I don't think the present
> so-called license would pass muster in any sane court in
> the United States after the original licensed code was
> plagiarized into a new binding license.
>

Try doing your homework. GPL v1 says:

> Each version is given a distinguishing version number.  If the Program
> specifies a version number of the license which applies to it and "any
> later version", you have the option of following the terms and conditions
> either of that version or of any later version published by the Free
> Software Foundation.  If the Program does not specify a version number
> of the license, you may choose any version ever published by the Free
> Software Foundation.

This means that when the code went GPL v1 -> GPL v2, the transition was 
permissible. Linux v1.0 shipped with the GPL v2. It did not ship with a
separate clause specifying that "You may only use *this* version of the GPL"
as it now does. (I haven't done any research to find out when this clause 
was added, but it was after the transition to v2).

I'm not sure what you're trying to imply about "conversion" or FSF 
"owning" Linux. Choosing to release your software under the GPL, even when 
the GPL is authored by a third party, does not make said third party the 
copyright owner of your work.

If a migration to v3 were to occur, the only potential hairball I see is 
if someone objected on the grounds that they contributed code to a version 
of the kernel Linus had marked as "GPLv2 Only". IANAL.

- Chase

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Wed, 25 Jan 2006, Chase Venters wrote:
> 
> This means that when the code went GPL v1 -> GPL v2, the transition was
> permissible. Linux v1.0 shipped with the GPL v2. It did not ship with a
> separate clause specifying that "You may only use *this* version of the GPL"
> as it now does. (I haven't done any research to find out when this clause was
> added, but it was after the transition to v2).

Bzzt. Look closer.

The Linux kernel has _always_ been under the GPL v2. Nothing else has ever 
been valid.

The "version 2 of the License, or (at your option) any later version" 
language in the GPL copying file is not - and has never been - part of the 
actual License itself. It's part of the _explanatory_ text that talks 
about how to apply the license to your program, and it says that _if_ you 
want to accept any later versions of the GPL, you can state so in your 
source code.

The Linux kernel has never stated that in general. Some authors have 
chosen to use the suggested FSF boilerplate (including the "any later 
version" language), but the kernel in general never has.

In other words: the _default_ license strategy is always just the 
particular version of the GPL that accompanies a project. If you want to 
license a program under _any_ later version of the GPL, you have to state 
so explicitly. Linux never did.

So: the extra blurb at the top of the COPYING file in the kernel source 
tree was added not to _change_ the license, but to _clarify_ these points 
so that there wouldn't be any confusion. 

The Linux kernel is under the GPL version 2. Not anything else. Some 
individual files are licenceable under v3, but not the kernel in general. 

And quite frankly, I don't see that changing. I think it's insane to 
require people to make their private signing keys available, for example. 
I wouldn't do it. So I don't think the GPL v3 conversion is going to 
happen for the kernel, since I personally don't want to convert any of my 
code. 

> If a migration to v3 were to occur, the only potential hairball I see is if
> someone objected on the grounds that they contributed code to a version of the
> kernel Linus had marked as "GPLv2 Only". IANAL.

No. You think "v2 or later" is the default. It's not. The _default_ is to 
not allow conversion.

Conversion isn't going to happen.

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Chase Venters]

On Wed, 25 Jan 2006, Linus Torvalds wrote:
>
> On Wed, 25 Jan 2006, Chase Venters wrote:
>>
>> This means that when the code went GPL v1 -> GPL v2, the transition was
>> permissible. Linux v1.0 shipped with the GPL v2. It did not ship with a
>> separate clause specifying that "You may only use *this* version of the GPL"
>> as it now does. (I haven't done any research to find out when this clause was
>> added, but it was after the transition to v2).
>
> Bzzt. Look closer.
>
> The Linux kernel has _always_ been under the GPL v2. Nothing else has ever
> been valid.

I see. That makes perfect sense given that the GPL v2 is dated 1991... 
this slipped my filters when I was grokking Dick's comment:

>>> The original GPL said something about:
>>> "You may not impose any further restrictions on the recipients'
>>> exercise of the rights granted herein." (Section 6).
>>> Then, that __exact__ code was redistributed under Version 2
>>> which further restricted rights,

> The "version 2 of the License, or (at your option) any later version"
> language in the GPL copying file is not - and has never been - part of the
> actual License itself. It's part of the _explanatory_ text that talks
> about how to apply the license to your program, and it says that _if_ you
> want to accept any later versions of the GPL, you can state so in your
> source code.

I wasn't actually referring to the explanatory text; rather, clause 9 in 
the section "TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND 
MODIFICATION". I suppose though upon reading clause 9 again the phrase "If 
the Program does not specify a version number of this License, you may 
choose any version ever published by the Free Software Foundation." is 
kind of confusing.

Does the header of the GPL in COPYING ("GNU General Public 
License; Version 2, June 1991") count as the "Program" specifying a 
version number of the license?

> The Linux kernel has never stated that in general. Some authors have
> chosen to use the suggested FSF boilerplate (including the "any later
> version" language), but the kernel in general never has.

Agreed.

> In other words: the _default_ license strategy is always just the
> particular version of the GPL that accompanies a project. If you want to
> license a program under _any_ later version of the GPL, you have to state
> so explicitly. Linux never did.

If the header of the GPL counts as the "Program's" specification of 
the version number, I suppose you're right. I just don't really understand 
why the language allowing _any_ version of the GPL if the version number 
isn't specified then... when would any project ever publish a version of 
the GPL license that has been modified to remove any mention of a version 
number?

> So: the extra blurb at the top of the COPYING file in the kernel source
> tree was added not to _change_ the license, but to _clarify_ these points
> so that there wouldn't be any confusion.
>
> The Linux kernel is under the GPL version 2. Not anything else. Some
> individual files are licenceable under v3, but not the kernel in general.
>
> And quite frankly, I don't see that changing. I think it's insane to
> require people to make their private signing keys available, for example.
> I wouldn't do it. So I don't think the GPL v3 conversion is going to
> happen for the kernel, since I personally don't want to convert any of my
> code.
>

Fair enough. I'm not trying to really argue for or against the kernel 
switching versions... just trying to address some posts I've seen on LKML 
that seem to get the licensing issue _really_ wrong (perhaps I'm now in 
that group.)

Cheers,
Chase

Re: GPL V3 and Linux - Dead Copyright Holders

[Paul Jakma]

On Wed, 25 Jan 2006, Linus Torvalds wrote:

> In other words: the _default_ license strategy is always just the 
> particular version of the GPL that accompanies a project. If you 
> want to license a program under _any_ later version of the GPL, you 
> have to state so explicitly. Linux never did.

That's not what section 9 seems to say. The default is "any version 
you like".

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
panic: kernel segmentation violation. core dumped		(only kidding)

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Paul Jakma wrote:

> On Wed, 25 Jan 2006, Linus Torvalds wrote:
>
>> In other words: the _default_ license strategy is always just the 
>> particular version of the GPL that accompanies a project. If you want 
>> to license a program under _any_ later version of the GPL, you have 
>> to state so explicitly. Linux never did.
>
>
> That's not what section 9 seems to say. The default is "any version 
> you like".


Right.

:-)

Jeff

>
> regards,

Re: GPL V3 and Linux - Dead Copyright Holders

[Filip Brcic]

[-- Attachment #1: Type: text/plain, Size: 1044 bytes --]

Дана Thursday 26 January 2006 18:59, Paul Jakma је написао(ла):
> On Wed, 25 Jan 2006, Linus Torvalds wrote:
> > In other words: the _default_ license strategy is always just the
> > particular version of the GPL that accompanies a project. If you
> > want to license a program under _any_ later version of the GPL, you
> > have to state so explicitly. Linux never did.
>
> That's not what section 9 seems to say. The default is "any version
> you like".

That's right, but

 Also note that the only valid version of the GPL as far as the kernel
 is concerned is _this_ particular version of the license (ie v2, not
 v2.2 or v3.x or whatever), unless explicitly otherwise stated.

Linux specifies version GPLv2 and only v2. Therefore, for Linux the GPLv2 is 
the default.

-- 
Filip Brcic <brcha@users.sourceforge.net>
WWWeb: http://purl.org/NET/brcha/home/
Jabber: brcha@kdetalk.net
Jabber: brcha@elitesecurity.org
Jabber: fbrcic@gmail.com
ICQ# 40994923
Yahoo! brcha
MSN: brcha@users.sourceforge.net

[-- Attachment #2: Type: application/pgp-signature, Size: 324 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Chase Venters wrote:

> On Thu, 26 Jan 2006, Filip Brcic wrote:
>
>> ???? Thursday 26 January 2006 18:59, Paul Jakma ?? ???????(??):
>>
>>> On Wed, 25 Jan 2006, Linus Torvalds wrote:
>>>
>>>> In other words: the _default_ license strategy is always just the
>>>> particular version of the GPL that accompanies a project. If you
>>>> want to license a program under _any_ later version of the GPL, you
>>>> have to state so explicitly. Linux never did.
>>>
>>>
>>> That's not what section 9 seems to say. The default is "any version
>>> you like".
>>
>>
>> That's right, but
>>
>> Also note that the only valid version of the GPL as far as the kernel
>> is concerned is _this_ particular version of the license (ie v2, not
>> v2.2 or v3.x or whatever), unless explicitly otherwise stated.
>>
>> Linux specifies version GPLv2 and only v2. Therefore, for Linux the 
>> GPLv2 is
>> the default.
>>
>
> Well, my understanding is that this clause wasn't always in COPYING. 
> If not for section 9 of the GPL, then the default would have always 
> been GPLv2 only.
>
> But since this clause was added after some time, one could argue that 
> some code in Linux, even lacking a specific "or any later version" 
> boilerplate, could be licensed under GPLv1, GPLv2, GPLv3, etc.
>
> However, as I stated before -- since this clause is now present, the 
> hairball going to GPLv3 would be copyright holders that submitted code 
> under the GPLv2 Only heading. Since Linus added this clause, and has 
> no doubt joined in many others submitting code since it was added, 
> portions of the kernel *are* GPLv2 Only; hence, it would be 
> impractical to legally migrate to GPLv3.
>
> I'll save from weighing in on whether or not GPLv3 is a good idea -- 
> this is just my evaluation of the facts I see before us.
>
> Cheers,
> Chase

Linus is posturing. I can go back to numerous previous versions when he 
and stallman were "buddy buddy" and the language was open
and said "any later version". Well, here's the gotcha. Any version 
released before Linus said this is GPL 2, 3 or later. As of today, all new
versions are GPLv2. That's how the law works. So 2.6.15 forward is GPLv2 
only. Linus cannot re-release previous Linux versions after he
already posted this NOTICE in COPYING, which he did and left the 
language pen like this. So it's up to the recevier of the code whether
its GPLv2 or GPLv3 or whatever, but those releases which appeared with 
COPYING stating this language are whatever GPL license you
want.

Jeff

Re: GPL V3 and Linux - Dead Copyright Holders

[Olivier Galibert]

On Thu, Jan 26, 2006 at 09:55:14AM -0700, Jeff V. Merkey wrote:
> Linus is posturing. I can go back to numerous previous versions when he 
> and stallman were "buddy buddy" and the language was open
> and said "any later version". Well, here's the gotcha. Any version 
> released before Linus said this is GPL 2, 3 or later. As of today, all new
> versions are GPLv2. That's how the law works. So 2.6.15 forward is GPLv2 
> only.

Since I'm sure you seriously did your homework, I'll just assume you
mistyped "2.4.0-test8".

  http://marc.theaimsgroup.com/?l=linux-kernel&m=96844578906748&w=2

  OG.

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> Linus is posturing. I can go back to numerous previous versions when he
> and stallman were "buddy buddy" and the language was open
> and said "any later version". Well, here's the gotcha. Any version
> released before Linus said this is GPL 2, 3 or later. As of today, all new
> versions are GPLv2. That's how the law works. So 2.6.15 forward is GPLv2
> only. Linus cannot re-release previous Linux versions after he
> already posted this NOTICE in COPYING, which he did and left the
> language pen like this. So it's up to the recevier of the code whether
> its GPLv2 or GPLv3 or whatever, but those releases which appeared with
> COPYING stating this language are whatever GPL license you
> want.
>
> Jeff

	Linus can't put additional restrictions on code he didn't write. If the
authors licensed it under the GPL version 2 and "any later version", Linus
can't re-release it under a more restrictive license. Read section 6
carefully:

  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.

	Notice that the code is licensed to you by the "original licensor", not by
the distributor. The inability to choose a later GPL version is definitely a
"further restriction".

	At least, that's how I read it.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Dave Jones]

On Thu, Jan 26, 2006 at 06:15:54PM -0800, David Schwartz wrote:

 > 	Linus can't put additional restrictions on code he didn't write. If the
 > authors licensed it under the GPL version 2 and "any later version", Linus
 > can't re-release it under a more restrictive license. Read section 6
 > carefully:

I suggest you read section 9, even more carefully. It's not an additional
restriction, it's a option that the license provides, which Linus chose.

Also read the preface of COPYING in the root directory of the kernel source.
(That has been there for a _long_ time btw).

		Dave

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> I suggest you read section 9, even more carefully. It's not an additional
> restriction, it's a option that the license provides, which Linus chose.

	Linus can choose the license for his own code. He can also set the rules
for contributions to Linux and set up the assumed licensing terms when code
is offered for inclusion. He cannot, however, change the license on any code
he did not write. He cannot even grant a license to any code he did not
write.

	From section 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.

	The grant of license does not come from the person doing the distributing,
it comes from the original licensor, which is the original author of each
portion. The license grant is automatic, so you can't modify it or affect it
in any way.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Dave Jones]

On Thu, Jan 26, 2006 at 06:37:28PM -0800, David Schwartz wrote:

 > He cannot, however, change the license on any code he did not write.

Explain how exactly Linus is 'changing' anything ?
The license of the project is clearly stated in the top-level
of the tree. 

If the person submitting code (to *any* project) doesn't read the
license that's the submitters problem, not the project maintainer.

		Dave

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> On Thu, Jan 26, 2006 at 06:37:28PM -0800, David Schwartz wrote:

>  > He cannot, however, change the license on any code he did not write.

> Explain how exactly Linus is 'changing' anything ?
> The license of the project is clearly stated in the top-level
> of the tree.

	I never said that he is or was. I said that he cannot.

> If the person submitting code (to *any* project) doesn't read the
> license that's the submitters problem, not the project maintainer.

	Definitely. And I believe that you have every right to assume that code
that is submitted to you without contradictory information is submitted
under the same license as governs the project it is a contribution to.

	The point is, nothing can change the license of code contributed under the
GPL except a relicense from the original author. The GPL is a special type
of license wherein you always get your license to use from the original
author. So nobody can ever change the license of any code you submit under
the GPL. This is spelled out clearly in section 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." (emphasis added)

	So when you receive a copy of Linux, each piece of code and modification
thereto is licensed to you from its particular original author. No author or
distributor has any power to change that (other than to offer you their own
code or contributions under a different license).

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Diego Calleja]

El Thu, 26 Jan 2006 18:37:28 -0800,
"David Schwartz" <davids@webmaster.com> escribió:

> is offered for inclusion. He cannot, however, change the license on any code
> he did not write. He cannot even grant a license to any code he did not

And he's not doing it, the COPYING file applies for all the code which 
doesn't specifies its own license.


Notice that COPYING has this: "Also note that the only valid version of
the GPL as far as the kernel  is concerned is _this_ particular version
of the license (ie v2, not  v2.2 or v3.x or whatever),
unless explicitly otherwise stated."
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

(IOW, people should care about their own code, but obviously people
usually licenses their code under the same license the project uses,
except some drivers that use a dual-license scheme etc)

RE: GPL V3 and Linux - Dead Copyright Holders

[Arjan van de Ven]

On Thu, 2006-01-26 at 18:15 -0800, David Schwartz wrote:
> > Linus is posturing. I can go back to numerous previous versions when he
> > and stallman were "buddy buddy" and the language was open
> > and said "any later version". Well, here's the gotcha. Any version
> > released before Linus said this is GPL 2, 3 or later. As of today, all new
> > versions are GPLv2. That's how the law works. So 2.6.15 forward is GPLv2
> > only. Linus cannot re-release previous Linux versions after he
> > already posted this NOTICE in COPYING, which he did and left the
> > language pen like this. So it's up to the recevier of the code whether
> > its GPLv2 or GPLv3 or whatever, but those releases which appeared with
> > COPYING stating this language are whatever GPL license you
> > want.
> >
> > Jeff
> 
> 	Linus can't put additional restrictions on code he didn't write. If the
> authors licensed it under the GPL version 2 and "any later version", Linus

you use the word "AND". Everyone else uses the word "OR". The word OR
means you have a choice, basically the code is dual licensed with GPLv2
and "any later version", and you can pick whichever one you want.
However if you combine it with the kernel, the v2 is chosen for you,
just by virtue of the license of the code that is v2-only. Eg: dual
licensed code in the kernel automatically picks the GPLv2 side.


If the wording was "AND", you would be right, but then both licenses
would have to apply at the same time, which gets really nasty if there
are conflicts between them (and it looks like there are)

Re: GPL V3 and Linux - Dead Copyright Holders

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 1065 bytes --]

On Thu, 26 Jan 2006 18:15:54 PST, David Schwartz said:

> 	Linus can't put additional restrictions on code he didn't write. If the
> authors licensed it under the GPL version 2 and "any later version", Linus
> can't re-release it under a more restrictive license

One thing to remember is that as a *practical* matter, there are *two* sets
of copyrights attached to the code (at least under US law).  Each contributor
has their rights to their code - Alan Cox has vast tracts of code, I have
a few dozen lines of bugfixes, and so on.

But equally important is the copyright that Linus has on the kernel as
a whole as a a compilation (see 17 USC 103).  As a practical matter, if you
don't like Linus's license for the compilation, you're stuck with small sections
under the fair use rules.  Similarly, Linus can't include anything under his
compilation unless the licenses are compatible...

Also, it means that even if you did manage to get all the copyright holders
for the code to agree to "GPL v2 or v3", Linus still holds the trump card on
the compilation.....

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[Olivier Galibert]

On Thu, Jan 26, 2006 at 06:15:54PM -0800, David Schwartz wrote:
> 	Linus can't put additional restrictions on code he didn't write. If the
> authors licensed it under the GPL version 2 and "any later version", Linus
> can't re-release it under a more restrictive license.

Yes he can.  The authors licensed the code under _multiple_ licenses
(even if some do not exist yet, which can be amusing, legally), each
of the existing one(s) allowing redistribution if you accept it.  He
does not have to accept _all_ of them to redistribute, _one_ of them
is enough.  And none of them allow to put the code under a different
license, in contrast to say the LGPL.

  OG.

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> On Thu, Jan 26, 2006 at 06:15:54PM -0800, David Schwartz wrote:

> > 	Linus can't put additional restrictions on code he didn't
> > write. If the
> > authors licensed it under the GPL version 2 and "any later
> > version", Linus
> > can't re-release it under a more restrictive license.

> Yes he can.

	No. He can't.

> The authors licensed the code under _multiple_ licenses
> (even if some do not exist yet, which can be amusing, legally), each
> of the existing one(s) allowing redistribution if you accept it.

	Correct. However, all the GPL-based ones grant rights *automatically* upon
distribution. "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"

> He
> does not have to accept _all_ of them to redistribute, _one_ of them
> is enough.

	Correct. However, the recipient gets the benefit of all GPL-style licenses.
That's what clause 6 says, particularly the "automatically" part and the
"from the original licensor" part.

> And none of them allow to put the code under a different
> license, in contrast to say the LGPL.

	Correct. The grant of GPL license is automatic on distribution, and it is
from the original author to the recipient.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Olivier Galibert]

On Fri, Jan 27, 2006 at 11:50:43AM -0800, David Schwartz wrote:
> 	Correct. However, all the GPL-based ones grant rights *automatically* upon
> distribution.

Of course not.  The license only applies if you accept it, and you
don't have to.  It's not an EULA.

  OG.

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> On Fri, Jan 27, 2006 at 11:50:43AM -0800, David Schwartz wrote:

> > Correct. However, all the GPL-based ones grant rights
> > *automatically* upon distribution.

> Of course not.  The license only applies if you accept it, and you
> don't have to.  It's not an EULA.

	You are correct that the license only applies if you accept it, but you are
incorrect if you think that you can fail to accept the GPL and thus restrict
the rights of those you distribute works to under some other license.

	Perhaps you're thinking that if I take a work that's dual-licensed under
the GPL and some other license, then distribute it under that other license,
the recipients don't get the GPL rights. That is not correct. GPL section 6
clearly states that the license is between the original author and the
ultimate recipient. The distributor has no say or control over this
automatic grant of license.

	Section 6 clears states that this license is:

	1) Automatic, not subject to anyone's discretion.

	2) From the original licensor, not the distributor.

	3) Under these terms and conditions, that is, the GPL.

	The case is the same if you modify the work and then distribute it. All
recipients still receive GPL rights from the original authors to the
respective contributions of those authors. You cannot stop it. It is
automatic, and it is from the original authors who agreed to give GPL rights
to all recipients when they placed their works under the GPL.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

David Schwartz wrote:

>>On Fri, Jan 27, 2006 at 11:50:43AM -0800, David Schwartz wrote:
>>    
>>
>
>  
>
>>>Correct. However, all the GPL-based ones grant rights
>>>*automatically* upon distribution.
>>>      
>>>
>
>  
>
>>Of course not.  The license only applies if you accept it, and you
>>don't have to.  It's not an EULA.
>>    
>>
>
>	You are correct that the license only applies if you accept it, but you are
>incorrect if you think that you can fail to accept the GPL and thus restrict
>the rights of those you distribute works to under some other license.
>
>	Perhaps you're thinking that if I take a work that's dual-licensed under
>the GPL and some other license, then distribute it under that other license,
>the recipients don't get the GPL rights. That is not correct. GPL section 6
>clearly states that the license is between the original author and the
>ultimate recipient. The distributor has no say or control over this
>automatic grant of license.
>
>	Section 6 clears states that this license is:
>
>	1) Automatic, not subject to anyone's discretion.
>
>	2) From the original licensor, not the distributor.
>
>	3) Under these terms and conditions, that is, the GPL.
>
>	The case is the same if you modify the work and then distribute it. All
>recipients still receive GPL rights from the original authors to the
>respective contributions of those authors. You cannot stop it. It is
>automatic, and it is from the original authors who agreed to give GPL rights
>to all recipients when they placed their works under the GPL.
>
>	DS
>  
>
This language constitutes conversion and violates about a dozen laws in 
a dozen states. This whole recipricol rights
grant language would have to be litigated, and I don't think anyone will 
know the outcome. I think it would end up
with the Judge saying "You guys take your stuff and go, and you GPL 
people take your stuff and go, and neither one of
you can distribute the others IP".

Jeff

Re: GPL V3 and Linux - Dead Copyright Holders

[Roland Kuhn]

[-- Attachment #1: Type: text/plain, Size: 2815 bytes --]

Hi David!

On 27 Jan 2006, at 03:15, David Schwartz wrote:

>
>> Linus is posturing. I can go back to numerous previous versions  
>> when he
>> and stallman were "buddy buddy" and the language was open
>> and said "any later version". Well, here's the gotcha. Any version
>> released before Linus said this is GPL 2, 3 or later. As of today,  
>> all new
>> versions are GPLv2. That's how the law works. So 2.6.15 forward is  
>> GPLv2
>> only. Linus cannot re-release previous Linux versions after he
>> already posted this NOTICE in COPYING, which he did and left the
>> language pen like this. So it's up to the recevier of the code  
>> whether
>> its GPLv2 or GPLv3 or whatever, but those releases which appeared  
>> with
>> COPYING stating this language are whatever GPL license you
>> want.
>>
>> Jeff
>
> 	Linus can't put additional restrictions on code he didn't write.  
> If the
> authors licensed it under the GPL version 2 and "any later  
> version", Linus
> can't re-release it under a more restrictive license. Read section 6
> carefully:
>
>   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.
>
> 	Notice that the code is licensed to you by the "original  
> licensor", not by
> the distributor. The inability to choose a later GPL version is  
> definitely a
> "further restriction".
>
> 	At least, that's how I read it.

Excuse my ignorant jumping into this thread. Correct me if I'm wrong:  
Linus takes code provided by others under "GPLv2 or any later  
version" and decides to comply to the rules of v2, which allow him to  
distribute the work under v2 only. Incidentally v2 is also the less  
strict than v3, so all Linus does is take the greatest freedom he is  
offered (and pass it on, of course).

The problem would start only if someone insisted on GPLv3 for his/her  
code and wanted to get it merged in Linus' tree. I think the only one  
to clarify this point is Linus, and he has already done so.

Ciao,
                     Roland

--
TU Muenchen, Physik-Department E18, James-Franck-Str., 85748 Garching
Telefon 089/289-12575; Telefax 089/289-12570
--
UNIX was not designed to stop you from doing stupid things, because that
would also stop you from doing clever things.
	-Doug Gwyn
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GS/CS/M/MU d-(++) s:+ a-> C+++ UL++++ P+++ L+++ E(+) W+ !N K- w--- M 
+ !V Y+
PGP++ t+(++) 5 R+ tv-- b+ DI++ e+++>++++ h---- y+++
------END GEEK CODE BLOCK------





[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 186 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[Paul Jakma]

On Thu, 26 Jan 2006, Filip Brcic wrote:

> Linux specifies version GPLv2 and only v2. Therefore, for Linux the 
> GPLv2 is the default.

Note that I didn't say otherwise. I was merely commenting on Linus' 
interpretation of the GPL generally.

Btw, with respect to the Linux kernel "v2 only" preamble, Alan Cox's 
email earlier in this thread is worth reading.

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
Man's horizons are bounded by his vision.

Re: GPL V3 and Linux - Dead Copyright Holders

[Diego Calleja]

El Thu, 26 Jan 2006 17:59:01 +0000 (GMT),
Paul Jakma <paul@clubi.ie> escribió:

> That's not what section 9 seems to say. The default is "any version 
> you like".

That's not exactly what it says.

The real text is (emphasis mine): "_IF_ the Program does not specify
a version number of this License, you may choose any version ever
published by the Free Software Foundation."


Before that it says: "Each version is given a distinguishing version
number.  If the Program specifies a version number of this License
which applies to it and "any later version", you have the option of
following the terms and conditions either of that version or of any
later version published by the Free Software Foundation"


IOW, such rules will aply if no version has been specified or 
if has the "any later version" addon. Which is not the case:
I can read "GNU GENERAL PUBLIC LICENSE, Version 2, June 1991".

Re: GPL V3 and Linux - Dead Copyright Holders

[Chase Venters]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1512 bytes --]

On Thu, 26 Jan 2006, Diego Calleja wrote:

> El Thu, 26 Jan 2006 17:59:01 +0000 (GMT),
> Paul Jakma <paul@clubi.ie> escribió:
>
>> That's not what section 9 seems to say. The default is "any version
>> you like".
>
> That's not exactly what it says.
>
> The real text is (emphasis mine): "_IF_ the Program does not specify
> a version number of this License, you may choose any version ever
> published by the Free Software Foundation."
>
>
> Before that it says: "Each version is given a distinguishing version
> number.  If the Program specifies a version number of this License
> which applies to it and "any later version", you have the option of
> following the terms and conditions either of that version or of any
> later version published by the Free Software Foundation"
>
>
> IOW, such rules will aply if no version has been specified or
> if has the "any later version" addon. Which is not the case:
> I can read "GNU GENERAL PUBLIC LICENSE, Version 2, June 1991".
>

It's possible there may be something I'm not understanding. I do recognize 
that the license has a version number on it.

But then again, so did GPLv1. And it also had the same section 9 clause.

So the question is - why would the GPL need a clause that says "You can 
use any version of the GPL if the Program does not specify a version" when 
every official version of the GPL includes a version number? Are they 
expecting authors to strip the version number header in order to somehow 
take advantage of section 9?

Cheers,
Chase

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

There seems to be some confusion about licensing. I'm just going to see 
if I can define the problem and the issues.

First - some people think all of Linux is under GPLv2 - but some people 
seem to think it's really GPLv2 or later. That needs to be resolved. Can 
different parts of Linux be controlled by multiple licenses. If so - 
that could create confusion because someone would have to agree to all 
the licenses within Linux in order to use it. The alternative is to say 
it's all GPLv2 and exclude GPLv3 from inclusion. Do we want to do that.

Second - is GPLv3 Linux compatible. If Linux were to start over today 
would it pick GPLv2 or GPLv3? Is there anything in GPLv3 that is not 
Linux compatible. I would at least like to see GPLv3 (final draft) to be 
100% Linux compatible.

Suppose GPLv3 were Linux compatible and many existing authors and new 
authors adopted GPLv3 but dead authors and some stubborn people and 
people who can't be found are still at GPLv2. Lets also assume that 
critical parts of Linux code are licensed in both worlds. What dos that 
mean? Does that mean that GPLv3 prevails?

This is something that might be worth doing some serious legal work on 
because if we do it wrong it could bite us hard in the future. But I 
want to try to properly raise the question here so that we all at least 
understand the problem.

My 2 centz ....

Re: GPL V3 and Linux - Dead Copyright Holders

[Chase Venters]

On Thu, 26 Jan 2006, Marc Perkel wrote:

> There seems to be some confusion about licensing. I'm just going to see if I 
> can define the problem and the issues.
>
> First - some people think all of Linux is under GPLv2 - but some people seem 
> to think it's really GPLv2 or later. That needs to be resolved. Can different 
> parts of Linux be controlled by multiple licenses. If so - that could create 
> confusion because someone would have to agree to all the licenses within 
> Linux in order to use it. The alternative is to say it's all GPLv2 and 
> exclude GPLv3 from inclusion. Do we want to do that.

I don't think there are any questions about what the "whole" of Linux is 
governed under. It's governed by whatever the most 'restrictive' license 
usage is, which is 'GPLv2 Only'. (The fact that GPLv2 Only doesn't apply 
to the whole kernel or perhaps even to some past kernels doesn't matter - 
you still can't package the whole as GPLv3)

The only discussion I see taking place any longer is basically irrelevant 
to Linux and GPLv3 - it concerns whether or not other, older kernel 
releases were legally GPLv2 Only or not. A disagreement on either the 
intent, mechanism or action of parts of the GPL license, if you will.

> Second - is GPLv3 Linux compatible. If Linux were to start over today would 
> it pick GPLv2 or GPLv3? Is there anything in GPLv3 that is not Linux 
> compatible. I would at least like to see GPLv3 (final draft) to be 100% Linux 
> compatible.

What does Linux compatible mean? Linus expressed some frustrations with 
some of the new terms, so perhaps in that sense the new license is not 
compatible.

If I'm reading him right, he doesn't want to restrain vendors like TiVO 
from using the kernel and subsequently refusing to provide the public with 
encryption keys necessary to build kernels to run on that hardware (for 
example).

> Suppose GPLv3 were Linux compatible and many existing authors and new authors 
> adopted GPLv3 but dead authors and some stubborn people and people who can't 
> be found are still at GPLv2. Lets also assume that critical parts of Linux 
> code are licensed in both worlds. What dos that mean? Does that mean that 
> GPLv3 prevails?

It's mental masturbation at this point. Suppose Linus approved of going to 
GPLv3... there may be some little technical gotchas in terms of "dead 
authors", but in that hypothetical, how many people would really be 
worried that the descendents of these dead authors would actively act to 
stop Linux from being distributed under the new license?

The bigger issue is that no matter where you fall on GPL's section 9, 
_lots_ of code is undoubtedly GPLv2 only. These living authors would have 
to agree to GPLv3, and as we now know, Linus does not.

> This is something that might be worth doing some serious legal work on 
> because if we do it wrong it could bite us hard in the future. But I want to 
> try to properly raise the question here so that we all at least understand 
> the problem.

Why? I think it's pretty much a dead issue.

Cheers,
Chase

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Chase Venters wrote:
> On Thu, 26 Jan 2006, Marc Perkel wrote:
>
>> There seems to be some confusion about licensing. I'm just going to 
>> see if I can define the problem and the issues.
>>
>> First - some people think all of Linux is under GPLv2 - but some 
>> people seem to think it's really GPLv2 or later. That needs to be 
>> resolved. Can different parts of Linux be controlled by multiple 
>> licenses. If so - that could create confusion because someone would 
>> have to agree to all the licenses within Linux in order to use it. 
>> The alternative is to say it's all GPLv2 and exclude GPLv3 from 
>> inclusion. Do we want to do that.
>
> I don't think there are any questions about what the "whole" of Linux 
> is governed under. It's governed by whatever the most 'restrictive' 
> license usage is, which is 'GPLv2 Only'. (The fact that GPLv2 Only 
> doesn't apply to the whole kernel or perhaps even to some past kernels 
> doesn't matter - you still can't package the whole as GPLv3)
If I write some code and that code becomes a critical part of the linux 
kernel and my code is GPLv3 then no one could use Linux unless they 
removed my code. (And all GPLv3 code) - Thus the inclusion of GPLv3 code 
would force the whole kernel to be effectively GPLv3. Unless Linus says 
nothing gets included unless it's GPLv2.
>
> The only discussion I see taking place any longer is basically 
> irrelevant to Linux and GPLv3 - it concerns whether or not other, 
> older kernel releases were legally GPLv2 Only or not. A disagreement 
> on either the intent, mechanism or action of parts of the GPL license, 
> if you will.
>
>> Second - is GPLv3 Linux compatible. If Linux were to start over today 
>> would it pick GPLv2 or GPLv3? Is there anything in GPLv3 that is not 
>> Linux compatible. I would at least like to see GPLv3 (final draft) to 
>> be 100% Linux compatible.
>
> What does Linux compatible mean? Linus expressed some frustrations 
> with some of the new terms, so perhaps in that sense the new license 
> is not compatible.
In this context Linux compatible means a license that Linus is happy with.
>
> If I'm reading him right, he doesn't want to restrain vendors like 
> TiVO from using the kernel and subsequently refusing to provide the 
> public with encryption keys necessary to build kernels to run on that 
> hardware (for example).
Yes - so perhaps GPLv3 is not Linux compatible. If so - then GPLv3 needs 
to be changed or excluded.
>
>> Suppose GPLv3 were Linux compatible and many existing authors and new 
>> authors adopted GPLv3 but dead authors and some stubborn people and 
>> people who can't be found are still at GPLv2. Lets also assume that 
>> critical parts of Linux code are licensed in both worlds. What dos 
>> that mean? Does that mean that GPLv3 prevails?
>
> It's mental masturbation at this point. Suppose Linus approved of 
> going to GPLv3... there may be some little technical gotchas in terms 
> of "dead authors", but in that hypothetical, how many people would 
> really be worried that the descendents of these dead authors would 
> actively act to stop Linux from being distributed under the new license?
Lawyers of big corporations. Greedy decendents who pull an SCO.
>
> The bigger issue is that no matter where you fall on GPL's section 9, 
> _lots_ of code is undoubtedly GPLv2 only. These living authors would 
> have to agree to GPLv3, and as we now know, Linus does not.
But if the Kernel contains GPLv3 code then that part of the kernel is 
subject to those restrictions. If the code is critical then it 
affectively makes Linux subjet to GPLv3. Perhaps the only way to avoid 
that is to require the same license for all kernel code?
>
>> This is something that might be worth doing some serious legal work 
>> on because if we do it wrong it could bite us hard in the future. But 
>> I want to try to properly raise the question here so that we all at 
>> least understand the problem.
>
> Why? I think it's pretty much a dead issue.
>
>
When lawyers get involved there is no such thing as a dead issue ..... sigh!

Re: GPL V3 and Linux - Dead Copyright Holders

[Chase Venters]

On Thu, 26 Jan 2006, Marc Perkel wrote:
> If I write some code and that code becomes a critical part of the linux 
> kernel and my code is GPLv3 then no one could use Linux unless they removed 
> my code. (And all GPLv3 code) - Thus the inclusion of GPLv3 code would force 
> the whole kernel to be effectively GPLv3. Unless Linus says nothing gets 
> included unless it's GPLv2.
>
> [more to this effect]

I assume that Linus's dissatisfaction with the GPLv3 means that this 
licensing characteristic is implied. I suppose it might be valuable to 
have an explicit declaration on this issue.

As for the rest of my message, my remarks and assumptions about how Linux 
is governed by the most restrictive clause apply to today's kernel (ie, no 
GPLv3 code at all).

I was honestly hoping that this debate wouldn't ignite until when we were 
much closer to having the real, final license. Or perhaps that it would 
ignite with the purpose of participating in the development of GPLv3. 
Neither seems to be the case, and so notwithstanding your concers about 
people merging in GPLv3 code, if the issue's not dead already, it's 
probably frozen until the real, live GPLv3 gets released.

Cheers,
Chase

Re: GPL V3 and Linux - Dead Copyright Holders

[linux-os (Dick Johnson)]

On Thu, 26 Jan 2006, Chase Venters wrote:

> On Thu, 26 Jan 2006, Marc Perkel wrote:
>> If I write some code and that code becomes a critical part of the linux
>> kernel and my code is GPLv3 then no one could use Linux unless they removed
>> my code. (And all GPLv3 code) - Thus the inclusion of GPLv3 code would force
>> the whole kernel to be effectively GPLv3. Unless Linus says nothing gets
>> included unless it's GPLv2.
>>
>> [more to this effect]
>
> I assume that Linus's dissatisfaction with the GPLv3 means that this
> licensing characteristic is implied. I suppose it might be valuable to
> have an explicit declaration on this issue.
>
> As for the rest of my message, my remarks and assumptions about how Linux
> is governed by the most restrictive clause apply to today's kernel (ie, no
> GPLv3 code at all).
>
> I was honestly hoping that this debate wouldn't ignite until when we were
> much closer to having the real, final license. Or perhaps that it would
> ignite with the purpose of participating in the development of GPLv3.
> Neither seems to be the case, and so notwithstanding your concers about
> people merging in GPLv3 code, if the issue's not dead already, it's
> probably frozen until the real, live GPLv3 gets released.
>

No. It's dead. Linus says it's dead. It's dead.

QUOTE..."

No. You think "v2 or later" is the default. It's not. The _default_ is to
not allow conversion.

Conversion isn't going to happen.

 		Linus

"...ENDQUOTE

> Cheers,
> Chase

Attempting to change existing licensing can invalidate all
licensing. The less done, the better.

Cheers,
Dick Johnson
Penguin : Linux version 2.6.13.4 on an i686 machine (5589.66 BogoMips).
Warning : 98.36% of all statistics are fiction.
.

****************************************************************
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 DeliveryErrors@analogic.com - and destroy all copies of this information, including any attachments, without reading or disclosing them.

Thank you.

Re: GPL V3 and Linux - V3 adds new restrictions

[Marc Perkel]

Trying to look at this from a legal point of view. GPLv3 might actually 
contradict GPLv2.

GPLv3 is more RESTRICTIVE than v2. With v2 you didn't have the new 
anti-DRM and anti-patent restrictions. The original license says 
somewhere that you can't change the license to be more restrictive.

None of us like patents and DRM but language that places new 
restrictions on software might not be GPLv2 compatible. Stallman might 
need to call his new license something else than GPL if he's going to 
add language that adds restrictions.

I can see an argument where GPLv2 prohibits GPLv3.

Re: GPL V3 and Linux - V3 adds new restrictions

[sean]

On Thu, 26 Jan 2006 14:28:45 -0800
Marc Perkel <marc@perkel.com> wrote:

> Trying to look at this from a legal point of view. GPLv3 might actually 
> contradict GPLv2.
> 
> GPLv3 is more RESTRICTIVE than v2. With v2 you didn't have the new 
> anti-DRM and anti-patent restrictions. The original license says 
> somewhere that you can't change the license to be more restrictive.
> 
> None of us like patents and DRM but language that places new 
> restrictions on software might not be GPLv2 compatible. Stallman might 
> need to call his new license something else than GPL if he's going to 
> add language that adds restrictions.
> 
> I can see an argument where GPLv2 prohibits GPLv3.
> 

As the _owner_ of the code, you can set whatever license(s) you choose.  
You don't lose your copyright just because you've granted the rest of 
us the right to use your code under the terms of the GPL.

For instance, licensing your code to the world as GPL does not mean you 
can't also license it to another group of people under XYZ terms.  The 
XYZ license may be totally incompatible with the GPL and have many more 
restrictions (a prohibition against distribution for instance).  Anyone
who has agreed to the XYZ license with you, could also go grab a copy
of the GPL'd code, but would then have to abide by _all_ the GPL 
conditions.

Similarly, the owner of each piece of Linux code is free to make the 
code he released under GPLv2 available now under v3 without any
conflict or violation of the prohibition against additional restrictions.

Of course, older versions of the code would still be available 
under v2 as well; not even the owner can revoke that.

Sean

Re: GPL V3 and Linux - V3 adds new restrictions

[sean]

On Thu, 26 Jan 2006 14:28:45 -0800
Marc Perkel <marc@perkel.com> wrote:

> Trying to look at this from a legal point of view. GPLv3 might actually 
> contradict GPLv2.
> 
> GPLv3 is more RESTRICTIVE than v2. With v2 you didn't have the new 
> anti-DRM and anti-patent restrictions. The original license says 
> somewhere that you can't change the license to be more restrictive.
> 
> None of us like patents and DRM but language that places new 
> restrictions on software might not be GPLv2 compatible. Stallman might 
> need to call his new license something else than GPL if he's going to 
> add language that adds restrictions.
> 
> I can see an argument where GPLv2 prohibits GPLv3.
> 

As the _owner_ of the code, you can set whatever license(s) you choose.  
You don't lose your copyright just because you've granted the rest of 
us the right to use your code under the terms of the GPL.

For instance, licensing your code to the world as GPL does not mean you 
can't also license it to another group of people under XYZ terms.  The 
XYZ license may be totally incompatible with the GPL and have many more 
restrictions (a prohibition against distribution for instance).  Anyone
who has agreed to the XYZ license with you, could also go grab a copy
of the GPL'd code, but would then have to abide by _all_ the GPL 
conditions.

Similarly, the owner of each piece of Linux code is free to make the 
code he released under GPLv2 available now under v3 without any
conflict or violation of the prohibition against additional restrictions.

Of course, older versions of the code would still be available 
under v2 as well; not even the owner can revoke that.

Sean

Re: GPL V3 and Linux - V3 adds new restrictions

[Linus Torvalds]

On Thu, 26 Jan 2006, Marc Perkel wrote:
>
> Trying to look at this from a legal point of view. GPLv3 might actually
> contradict GPLv2.

The GPLv2 is explicitly written so that _nothing_ else than a GPLv2 
license can be compatible with it.

At most you can dual-license and effectively allow extra rights that way, 
but the point is that the GPLv2 is always the limit for any restrictions 
(and it's written so that anybody can always take any but the GPLv2 
freedoms away - so if you dual-license, your licensees can always decide 
to _only_ honor the GPLv2, and ignore any other license).

That's why, in order to re-license anythingt from GPLv2 to anything else, 
the license grant itself must make it clear that the "anything else" was 
always permissible in the first place. Which is why the FSF had only two 
"outs": either people explicitly do the "..or any later version" thing, 
_or_ people don't mention the version of the GPL at all, in which case any 
version will do.

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Diego Calleja]

El Thu, 26 Jan 2006 12:57:27 -0600 (CST),
Chase Venters <chase.venters@clientec.com> escribió:

> So the question is - why would the GPL need a clause that says "You can 
> use any version of the GPL if the Program does not specify a version" when 
> every official version of the GPL includes a version number? Are they 
> expecting authors to strip the version number header in order to somehow 
> take advantage of section 9?

That's what I though. The only sane reason I can find is that many
projects don't even include a copy of the GPL and just say "this
is licensed under the GNU Public License"

Re: GPL V3 and Linux - Dead Copyright Holders

[Paul Jakma]

On Thu, 26 Jan 2006, Diego Calleja wrote:

> The real text is (emphasis mine): "_IF_ the Program does not 
> specify a version number of this License, you may choose any 
> version ever published by the Free Software Foundation."

> Before that it says: "Each version is given a distinguishing 
> version number.  If the Program specifies a version number of this 
> License which applies to it and "any later version", you have the 
> option of following the terms and conditions either of that version 
> or of any later version published by the Free Software Foundation"

> IOW, such rules will aply if no version has been specified or if 
> has the "any later version" addon. Which is not the case: I can 
> read "GNU GENERAL PUBLIC LICENSE, Version 2, June 1991".

That refers to the version of the GPL licence. That is specific to 
the licence document and distinct from any text where "the Program 
specifies a version number of this License which applies to it ..." 
(one suggestion for doing so is given: include a Preamble).

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
"And I want a bike and a monkey and a friend for the monkey."

 	--Ralph Wiggum
 	  Brother's Little Helper (Episode AABF22)

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Linus Torvalds wrote:
> No. You think "v2 or later" is the default. It's not. The _default_ is to 
> not allow conversion.
>
> Conversion isn't going to happen.
>
> 		Linus
>
>   
Just for clarification. What you are saying is that anyone who insists 
on contributing to the kernel under GPLv3 - that code would be 
prohibited from being included in the kernel? That to contribute to the 
kernel you must contribute under the terms presently in place?

I'm just asking for clarification.

I don't have a GPLv3 opinion yet since it's a changing document. But I 
do see a need for a consistent licensing of the components of Linux to 
avoid confusion and I'm worried that if any GPLv3 gets in then it will 
be all GPLv3 so you would have to prohibity that if you are going to 
stop Linux from becoming GPLv3 by default.

Re: GPL V3 and Linux - Dead Copyright Holders

[Al Viro]

On Thu, Jan 26, 2006 at 02:16:43PM -0800, Marc Perkel wrote:
> Just for clarification. What you are saying is that anyone who insists 
> on contributing to the kernel under GPLv3 - that code would be 
> prohibited from being included in the kernel? That to contribute to the 
> kernel you must contribute under the terms presently in place?

"terms compatible with those for code already in place".  Which is obviously
true for any project, since otherwise you end up with result that can not
be distributed at all.

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Thu, 26 Jan 2006, Marc Perkel wrote:
>
> Just for clarification. What you are saying is that anyone who insists on
> contributing to the kernel under GPLv3 - that code would be prohibited from
> being included in the kernel? That to contribute to the kernel you must
> contribute under the terms presently in place?

No. We actually have a lot of code that is more widely licensed than just 
GPLv2. There's the GPL/BSD code, and there's a lot of files that have the 
".. or any later version" addendum which means that they are GPLv3 
compatible.

The only thing that the kernel requires is that since the majority of the 
code is actually GPLv2-only, that in order for you to be able to link with 
the code, your license has to be GPLv2-compatible.

A "GPLv3 _only_" license is not compatible with GPLv2, since v3 adds new 
limitations to re-distribution. But what you can do is to dual-license the 
code - the same way we've had GPL/BSD dual licenses. Of course, that 
effectively becomes the same as "GPLv2" with the "any later version" 
clause, but if you like the v3 in _particular_, you can actually mention 
it specifically (ie you can dual-license under "v2 _or_ v3", but without 
the "any later version" wording if you want).

Note that the Linux kernel has had the clarification that the "by default, 
we're version-2 _only_" for a long time, and that limitation is not a new 
thing.

You can argue that I should have made that clear on "Day 1" (back in 1992, 
when the original switch to the GPL happened), but the fact is, all of the 
development for the last five or more years has been done with that "v2 
only, unless otherwise stated" (I forget exactly when it happened, but it 
was before we even started using BK, so it's a loong time ago).

Also, this has been discussed before, and anybody who felt that they 
didn't want to have the "v2 only" limitation has been told to add the "or 
any later version" thing to their own code, so nobody can claim that I 
restricted their licensing. 

So to recap:

 - Linux has been v2-only for a _loong_ time, long before there was even 
   any choice of licenses. That explicit "v2 only" thing was there at 
   least for 2.4.0, which is more than five years ago. So this is not some 
   sudden reaction to the current release of GPLv3. This has been there 
   quite _independently_ of the current GPLv3 discussion.

 - if you disagree with code you write, you can (and always have been 
   able) to say so, and dual-license in many different ways, including 
   using the "or later version" language. But that doesn't change the fact 
   that others (a _lot_ of others) have been very much aware of the "v2 
   only" rule for the kernel, and that most of the Linux kernel sources 
   are under that rule.

 - People argue that Linux hasn't specified a version, and that by virtue 
   of paragraph 9, you'd be able to choose any version you like. I 
   disagree. Linux has always specified the version: I don't put the 
   license in the source code, the source code just says

	Copyright (C) 1991-2002 Linux Torvalds

   and the license is in the COPYING file, which has ALWAYS been v2. Even 
   before (for clarification reasons) it explicitly said so.

   In other words, that "if no version is mentioned" simply isn't even an 
   argument. That's like arguing that "if no license is mentioned, it's 
   under any license you want", which is crap. If no license is mentioned, 
   you don't have any license at all to use it. The license AND VERSION 
   has always been very much explicit: linux/COPYING has been there since 
   1992, and it's been the _version_2_ of the license since day 1.

   People can argue against that any way they like. In the end, the only 
   way you can _really_ argue against it is in court. Last I saw, 
   intentions mattered more than any legalistic sophistry. The fact that 
   Linux has been distributed with a specific version of the GPL is a big 
   damn clue, and the fact that I have made my intentions very clear over 
   several years is another HUGE clue.

 - I don't see any real upsides to GPLv3, and I do see potential 
   downsides. Things that have been valid under v2 are no longer valid 
   under v3, so changing the license has real downsides.

Quite frankly, _if_ we ever change to GPLv3, it's going to be because 
somebody convinces me and other copyright holders to add the "or any later 
license" to all files, just because v3 really is so much better. It 
doesn't seem likely, but hey, if somebody shows that the GPLv2 is 
unconsitutional (hah!), maybe something like that happens. 

So I'm not _entirely_ dismissing an upgrade, but quite frankly, to upgrade 
would be a huge issue. Not just I, but others that have worked on Linux 
over the last five to ten years would have to agree on it. In contrast, 
staying with GPLv2 is a no-brainer: we've used it for almost 15 years, and 
it's worked fine, and nobody needs any convincing.

And that really is a big issue: GPLv2 is a perfectly fine license. It has 
worked well for us for fourteen years, nothing really changed with the 
introduction of GPLv3. The fact that there is a newer license to choose 
from doesn't detract from the older ones.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Michael Buesch]

[-- Attachment #1: Type: text/plain, Size: 232 bytes --]

On Saturday 28 January 2006 02:33, you wrote:
> 	Copyright (C) 1991-2002 Linux Torvalds
                                    ^
And I thought I was the only one doing this mistake
from time to time. :P

-- 
Greetings Michael.

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Michael Buesch wrote:

>On Saturday 28 January 2006 02:33, you wrote:
>  
>
>>	Copyright (C) 1991-2002 Linux Torvalds
>>    
>>
>                                    ^
>And I thought I was the only one doing this mistake
>from time to time. :P
>
>  
>
I do it to all the time.  Freudian Slip.

:-)

Jeff

Re: GPL V3 and Linux - Dead Copyright Holders

[Simon Oosthoek]

Linus Torvalds wrote:
> The Linux kernel is under the GPL version 2. Not anything else. Some
>  individual files are licenceable under v3, but not the kernel in
> general.

I believe that if v2 and v3 turn out to be incompatible, it would be
quite hard to rationalise v3+ licensed files inside the kernel. So when
people want their code to be in the kernel and still be v3+ compatible,
they should probably dual license it, or include a specific section
saying that the code can be licensed under v2 only if in the context of
the Linux kernel.

> And quite frankly, I don't see that changing. I think it's insane to
>  require people to make their private signing keys available, for
> example. I wouldn't do it. So I don't think the GPL v3 conversion is
> going to happen for the kernel, since I personally don't want to
> convert any of my code.

I'm not sure this is the correct interpretation of the current draft. I
assume you're referring to this part:

GPLv3-draft1:
 > (...)
> Complete Corresponding Source Code also includes any encryption or
> authorization codes necessary to install and/or execute the source
> code of the work, perhaps modified by you, in the recommended or
> principal context of use, such that its functioning in all
> circumstances is identical to that of the work, except as altered by
> your modifications. It also includes any decryption codes necessary
> to access or unseal the work's output. Notwithstanding this, a code
> need not be included in cases where use of the work normally implies
> the user already has it.
> (...)

I'd interpret that as forcing people who try to hide their code or make 
it difficult to get at the source code to not be able to do that. I'm 
not sure this would affect the Linux kernel at all and I don't think it 
would require any of your private keys to be disclosed at all. If you 
would sign or encrypt the kernel distribution with your private key, 
everyone would need to have access to your public key, but that's the 
whole idea anyway.

Cheers

Simon

Re: GPL V3 and Linux - Dead Copyright Holders

[Al Viro]

On Fri, Jan 27, 2006 at 11:46:17AM +0100, Simon Oosthoek wrote:
> Linus Torvalds wrote:
> >The Linux kernel is under the GPL version 2. Not anything else. Some
> > individual files are licenceable under v3, but not the kernel in
> >general.
> 
> I believe that if v2 and v3 turn out to be incompatible, it would be
> quite hard to rationalise v3+ licensed files inside the kernel. So when
> people want their code to be in the kernel and still be v3+ compatible,
> they should probably dual license it, or include a specific section
> saying that the code can be licensed under v2 only if in the context of
> the Linux kernel.

Bzzert.   "GPLv2 only in the context of the Linux kernel" is incompatible
with GPLv2 and means that resulting kernel is impossible to distribute.

Re: GPL V3 and Linux - Dead Copyright Holders

[Simon Oosthoek]

Al Viro wrote:
> On Fri, Jan 27, 2006 at 11:46:17AM +0100, Simon Oosthoek wrote:
>> Linus Torvalds wrote:
>>> The Linux kernel is under the GPL version 2. Not anything else. Some
>>> individual files are licenceable under v3, but not the kernel in
>>> general.
>> I believe that if v2 and v3 turn out to be incompatible, it would be
>> quite hard to rationalise v3+ licensed files inside the kernel. So when
>> people want their code to be in the kernel and still be v3+ compatible,
>> they should probably dual license it, or include a specific section
>> saying that the code can be licensed under v2 only if in the context of
>> the Linux kernel.
> 
> Bzzert.   "GPLv2 only in the context of the Linux kernel" is incompatible
> with GPLv2 and means that resulting kernel is impossible to distribute.

really? if it was dual licensed (that's what I meant, perhaps the "or" 
should be an "and"? ;-), v2 in the kernel and v3(or any later version, 
etc.), if the code is used outside of the kernel, it would "fall back 
to" v3+ as soon as it's taken out of the kernel and used in something else.

If I'd want to contribute code to the kernel, I'd have to comply with 
the license of the kernel, which is v2 of the GPL. If I would actually 
prefer my code to be licensed under v3 or higher, I'd have to specify 
that my code is only licensed under v2 for the kernel to humour Linus 
Torvalds and respect the license of the kernel, but in all other ways 
the code is used, I only grant a license to copy under the conditions of 
the GPL v3 or higher. I don't see why that would affect the distribution 
of the kernel at all.

Cheers

Simon


-- 
phone:(+31|0)53 4810319
fax:  (+31|0)53 4810333
simon.oosthoek@ti-wmc.nl
http://www.ti-wmc.nl/

Re: GPL V3 and Linux - Dead Copyright Holders

[Olivier Galibert]

On Fri, Jan 27, 2006 at 03:00:53PM +0100, Simon Oosthoek wrote:
> If I'd want to contribute code to the kernel, I'd have to comply with 
> the license of the kernel, which is v2 of the GPL. If I would actually 
> prefer my code to be licensed under v3 or higher, I'd have to specify 
> that my code is only licensed under v2 for the kernel to humour Linus 
> Torvalds and respect the license of the kernel, but in all other ways 
> the code is used, I only grant a license to copy under the conditions of 
> the GPL v3 or higher. I don't see why that would affect the distribution 
> of the kernel at all.

"GPLv2 only for the kernel" is a different license than "GPLv2" and is
incompatible with GPLv2.

  OG.

Re: GPL V3 and Linux - Dead Copyright Holders

[Simon Oosthoek]

Olivier Galibert wrote:
> On Fri, Jan 27, 2006 at 03:00:53PM +0100, Simon Oosthoek wrote:
>> If I'd want to contribute code to the kernel, I'd have to comply with 
>> the license of the kernel, which is v2 of the GPL. If I would actually 
>> prefer my code to be licensed under v3 or higher, I'd have to specify 
>> that my code is only licensed under v2 for the kernel to humour Linus 
>> Torvalds and respect the license of the kernel, but in all other ways 
>> the code is used, I only grant a license to copy under the conditions of 
>> the GPL v3 or higher. I don't see why that would affect the distribution 
>> of the kernel at all.
> 
> "GPLv2 only for the kernel" is a different license than "GPLv2" and is
> incompatible with GPLv2.
>

hmm, so if I want to contribute to the kernel, but prefer my code to be 
licensed under GPLv3 or higher, I would be unable to submit it to the 
kernel unless I "lower my standards" to GPLv2 or higher?

If someone wants to use that code elsewhere, he can take it from the 
kernel and re-use it in a GPLv2 project and further, which may violate 
the terms of GPLv3 and would therefore conflict with my interests.

Of course, this may be purely theoretical, but it could block people 
from submitting code to the kernel in order avoid the v2 GPL license.

 From this, I think it could be concluded that it might turn out to be 
quite harmful if code under GPLv3+ cannot be combined with the linux 
kernel...

/Simon

PS, IANAL

Re: GPL V3 and Linux - Dead Copyright Holders

[Al Viro]

On Fri, Jan 27, 2006 at 03:42:29PM +0100, Simon Oosthoek wrote:
> hmm, so if I want to contribute to the kernel, but prefer my code to be 
> licensed under GPLv3 or higher, I would be unable to submit it to the 
> kernel unless I "lower my standards" to GPLv2 or higher?
> 
> If someone wants to use that code elsewhere, he can take it from the 
> kernel and re-use it in a GPLv2 project and further, which may violate 
> the terms of GPLv3 and would therefore conflict with my interests.
> 
> Of course, this may be purely theoretical, but it could block people 
> from submitting code to the kernel in order avoid the v2 GPL license.
> 
> From this, I think it could be concluded that it might turn out to be 
> quite harmful if code under GPLv3+ cannot be combined with the linux 
> kernel...

That's GPL working *as* *intended*.  No, you can't create a derivative
of GPLv2 program and prohibit the use of your modifications in other
GPLv2 programs.  It's not just "Linus won't accept it upstream"; it's
"you can't even distribute such fork yourself".  And it's 100% intentional -
that's what GPL is about.

As for the harm...  We somehow survived years of similar "harmful" situation.
No, you can't put proprietary code in kernel and prohibit other GPLv2
projects to reuse it.  Yes, it would theoretically turn some authors of
extremely valuable (but never materializing) code away.  And yes, there had
been very similar whining and dire warnings.  Nothing new here...

Re: GPL V3 and Linux - Dead Copyright Holders

[Simon Oosthoek]

Al Viro wrote:
> 
> That's GPL working *as* *intended*.  No, you can't create a derivative
> of GPLv2 program and prohibit the use of your modifications in other
> GPLv2 programs.  It's not just "Linus won't accept it upstream"; it's
> "you can't even distribute such fork yourself".  And it's 100% intentional -
> that's what GPL is about.

I think I understand the GPL, but I'm not a lawyer :-)

> As for the harm...  We somehow survived years of similar "harmful" situation.
> No, you can't put proprietary code in kernel and prohibit other GPLv2
> projects to reuse it.  Yes, it would theoretically turn some authors of
> extremely valuable (but never materializing) code away.  And yes, there had
> been very similar whining and dire warnings.  Nothing new here...

The comparison between proprietary and GPLv3 code is disconcerting and 
strangely funny in this way :-/

What I think *is* new here is that some people would consider GPLv3 a 
logical extension of GPLv2 (in the sense of enhancing freedom), but if 
it turns out that it's so different from GPLv2 that GPLv3 code cannot be 
mixed into the v2 kernel, which does stem from the same philosophy.

I suppose I'll just wait for the next draft and check again...

thanks for the feedback!

/Simon

-- 
phone:(+31|0)53 4810319
fax:  (+31|0)53 4810333
simon.oosthoek@ti-wmc.nl
http://www.ti-wmc.nl/

Re: GPL V3 and Linux - Dead Copyright Holders

[Chris Bergeron]

Simon Oosthoek wrote:
> Al Viro wrote:
>> On Fri, Jan 27, 2006 at 11:46:17AM +0100, Simon Oosthoek wrote:
>>> Linus Torvalds wrote:
>>>> The Linux kernel is under the GPL version 2. Not anything else. Some
>>>> individual files are licenceable under v3, but not the kernel in
>>>> general.
>>> I believe that if v2 and v3 turn out to be incompatible, it would be
>>> quite hard to rationalise v3+ licensed files inside the kernel. So when
>>> people want their code to be in the kernel and still be v3+ compatible,
>>> they should probably dual license it, or include a specific section
>>> saying that the code can be licensed under v2 only if in the context of
>>> the Linux kernel.
>>
>> Bzzert.   "GPLv2 only in the context of the Linux kernel" is 
>> incompatible
>> with GPLv2 and means that resulting kernel is impossible to distribute.
> If I would actually prefer my code to be licensed under v3 or higher, 
> I'd have to specify that my code is only licensed under v2 for the 
> kernel to humour Linus Torvalds and respect the license of the kernel, 
> but in all other ways the code is used, I

I think what the "Bzzert" is implying is that once you license your code 
under the GPL v2 for the kernel use, it's then licensed that way for 
distribution.  You can't prohibit people from using it in other contexts 
with that license or it violates the GPL v2 license and then can't be 
distributed with the kernel, either. If I'm reading that correctly.

To cross license you'd have to allow either license for any use, really, 
as long as the use complies with the license.  Adding 'if the code is 
used outside of the kernel, it would "fall back to" v3+' would violate 
the GPL v2.

-- Chris

Re: GPL V3 and Linux - Dead Copyright Holders

[Kyle Moffett]

On Jan 27, 2006, at 09:00, Simon Oosthoek wrote:
> really? if it was dual licensed (that's what I meant, perhaps the  
> "or" should be an "and"? ;-), v2 in the kernel and v3 (or any later  
> version, etc.), if the code is used outside of the kernel, it would  
> "fall back to" v3+ as soon as it's taken out of the kernel and used  
> in something else.

You cannot say "This code is GPLv2 only in the context of the  
kernel".  You may say "This code is licensed under GPLv2 or GPLv3".   
The reasoning behind this is as follows:  If I take a kernel tree,  
and apply a large patch that removes all of the code except for your  
driver, the result is perfectly legal to distribute under GPLv2  
(because I took a combined GPLv2 source, applied any modification I  
felt like (such as deleting everything except one of the drivers),  
and therefore get a GPLv2 source.

One thing I am not sure about: If you have a kernel which contains  
code licensed under GPLv2 only and code licensed under GPLv3 only,  
would it be redistributable at all?  If so, under what conditions/ 
terms?  IANAL, so take the following with a couple pounds of salt;  
since GPLv2 says something like "You cannot add any additional  
restrictions", and GPLv3 adds additional restrictions, it might be  
true that the composite is not legally redistributable even though  
the two individual parts are, no?

Cheers,
Kyle Moffett

--
They _will_ find opposing experts to say it isn't, if you push hard  
enough the wrong way.  Idiots with a PhD aren't hard to buy.
   -- Rob Landley

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Fri, 27 Jan 2006, Simon Oosthoek wrote:
> 
> really? if it was dual licensed (that's what I meant, perhaps the "or" should
> be an "and"? ;-), v2 in the kernel and v3(or any later version, etc.), if the
> code is used outside of the kernel, it would "fall back to" v3+ as soon as
> it's taken out of the kernel and used in something else.

You very much _can_ dual-license it, and say "for the kernel, we license 
it under GPLv2".

But if you do that, then the GPLv2 licensing in the kernel means that 
somebody else can take it from the kernel, and ti can still be under the 
GPLv2 too (_and_ the GPLv3, for that matter - the kernel GPLv2 license in 
no way takes away the ability to use it in any other way that you've 
dual-licensed it).

In other words, you cannot say "outside of the kernel, you ahve to use the 
GPLv3", because the GPLv2 usage _inside_ of the kernel requires that the 
GPLv2 also be usable outside of it.

But you most definitely _can_ dual-license in general. It's perfectly fine 
to license something under the GPLv2 and allow redistribution with the 
kernel, _and_ have the same code as part of some other project under 
GPLv3.

It's just that you cannot limit the kernel version to just a kernel 
project. Somebody can (at any time) take the Linux kernel, and turning it 
into some totally other GPL-v2-licensed project.

(This really is not at all specific to the GPLv3 - the same thing holds 
for any other dual licensing with the GPL. You  cannot license your 
proprietary code to be "GPL only within the kernel, and proprietary 
anywhere else". The GPL inherently requires that the code can be used in 
another project).

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Horst von Brand]

Simon Oosthoek <simon.oosthoek@ti-wmc.nl> wrote:

[...]

> really? if it was dual licensed (that's what I meant, perhaps the "or"
> should be an "and"? ;-), v2 in the kernel and v3(or any later version,
> etc.), if the code is used outside of the kernel, it would "fall back
> to" v3+ as soon as it's taken out of the kernel and used in something
> else.

Then, due to the GPLv2 license, anybody can take the version in the kernel
and distribute it further under GPLv2. Exactly because GPL is about keeping
such roads open.
-- 
Dr. Horst H. von Brand                   User #22616 counter.li.org
Departamento de Informatica                     Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria              +56 32 654239
Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Fri, 27 Jan 2006, Al Viro wrote:
> 
> Bzzert.   "GPLv2 only in the context of the Linux kernel" is incompatible
> with GPLv2 and means that resulting kernel is impossible to distribute.

Indeed. The GPL (both v2 and v3) disallow restricting usage. 

So certain _code_ can be either v2 or v3 only, but you can't make that 
decision based on how the code is used. 

So you can't license, for example, your code "udner the GPL only for the 
Linux kernel". Trust me, some companies have actually wanted to do exactly 
that - they wanted to distribute their code, but _only_ for the kernel 
(and you'd not be allowed to use it for any other GPL'd project). That 
just cannot fly. It's either GPL, or it isn't. There's no "GPL with the 
following rules".

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Giacomo A. Catenazzi]

Linus Torvalds wrote:
> 
> On Fri, 27 Jan 2006, Al Viro wrote:
> 
>>Bzzert.   "GPLv2 only in the context of the Linux kernel" is incompatible
>>with GPLv2 and means that resulting kernel is impossible to distribute.
> 
> 
> Indeed. The GPL (both v2 and v3) disallow restricting usage. 
> 
> So certain _code_ can be either v2 or v3 only, but you can't make that 
> decision based on how the code is used. 
> 
> So you can't license, for example, your code "udner the GPL only for the 
> Linux kernel". Trust me, some companies have actually wanted to do exactly 
> that - they wanted to distribute their code, but _only_ for the kernel 
> (and you'd not be allowed to use it for any other GPL'd project). That 
> just cannot fly. It's either GPL, or it isn't. There's no "GPL with the 
> following rules".

Not really the same case, but ...

/*
  *  linux/init/version.c
  *
  *  Copyright (C) 1992  Theodore Ts'o
  *
  *  May be freely distributed as part of Linux.
  */

ciao
	cate

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> Not really the same case, but ...
>
> /*
>   *  linux/init/version.c
>   *
>   *  Copyright (C) 1992  Theodore Ts'o
>   *
>   *  May be freely distributed as part of Linux.
>   */

	That should be fixed. If the author didn't intend to release this under
GPLv2, it should be replaced. If he did, it should be clarified.

	Actually, we have 13 years of use. And I think Ted knows that the kernel is
under GPLv2. ;)

	You can't include something with the Linux kernel if it's not under GPLv2.
And you can't put additional restrictions on GPLv2 code. We all know that, I
think.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Fri, 27 Jan 2006, Simon Oosthoek wrote:
> 
> I'm not sure this is the correct interpretation of the current draft. I
> assume you're referring to this part:
>
> [ snipped ]

Yes.

> I'd interpret that as forcing people who try to hide their code or make it
> difficult to get at the source code to not be able to do that.

IF that is the legal interpretation, then yes, I'd agree with you. And no, 
I'm not a lawyer. However, the way I read it, it's not about just not 
being able to hide the object code - it's fundamentally about being able 
to replace and run the object code.

I may indeed be reading it wrong, but I don't think I am. It explicitly 
says "install and/or execute".

So I think it says that if I have a private signing key that "enables" the 
kernel on some hardware of mine, GPLv3 requires that private key to be 
made available for that hardware. Note how that is tied to the _hardware_ 
(or platform - usualyl the checking would be done by firmware, of course), 
not the actual source code of the program.

And that's really what I don't like. I believe that a software license 
should cover the software it licenses, not how it is used or abused - even 
if you happen to disagree with certain types of abuse.

I believe that hardware that limits what their users can do will die just 
becuase being user-unfriendly is not a way to do successful business. Yes, 
I'm a damned blue-eyed optimist, but I'd rather be blue-eyed than consider 
all uses of security technology to necessarily always be bad.

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Chris Adams]

Once upon a time, Linus Torvalds  <torvalds@osdl.org> said:
>I believe that hardware that limits what their users can do will die just 
>becuase being user-unfriendly is not a way to do successful business. Yes, 
>I'm a damned blue-eyed optimist, but I'd rather be blue-eyed than consider 
>all uses of security technology to necessarily always be bad.

I haven't read the GPLv3 draft myself, but I would guess that private
signing key language is aimed squarely at TiVo.  They use a Linux
kernel, but only kernel binaries signed by them will load (so you can't
modify the kernel without jumping through hoops).  Is that
"user-unfriendly"?  Probably so, but only to a small percentage of the
users.

-- 
Chris Adams <cmadams@hiwaay.net>
Systems and Network Administrator - HiWAAY Internet Services
I don't speak for anybody but myself - that's enough trouble.

Re: GPL V3 and Linux - Dead Copyright Holders

[Simon Oosthoek]

On Sat, January 28, 2006 5:39, Linus Torvalds said:
>
>
> On Fri, 27 Jan 2006, Simon Oosthoek wrote:
>>
>> I'm not sure this is the correct interpretation of the current draft. I
>> assume you're referring to this part:
>>
>> [ snipped ]
>
> Yes.
>
>> I'd interpret that as forcing people who try to hide their code or make
>> it difficult to get at the source code to not be able to do that.
>
> IF that is the legal interpretation, then yes, I'd agree with you. And no,
> I'm not a lawyer. However, the way I read it, it's not about just not
> being able to hide the object code - it's fundamentally about being able
> to replace and run the object code.

After some more reading and thinking, I agree with your interpretation.

> I may indeed be reading it wrong, but I don't think I am. It explicitly
> says "install and/or execute".
>
> So I think it says that if I have a private signing key that "enables" the
> kernel on some hardware of mine, GPLv3 requires that private key to be
> made available for that hardware. Note how that is tied to the _hardware_
> (or platform - usualyl the checking would be done by firmware, of course),
> not the actual source code of the program.
>
> And that's really what I don't like. I believe that a software license
> should cover the software it licenses, not how it is used or abused - even
> if you happen to disagree with certain types of abuse.
>
> I believe that hardware that limits what their users can do will die just
> becuase being user-unfriendly is not a way to do successful business. Yes,
> I'm a damned blue-eyed optimist, but I'd rather be blue-eyed than consider
> all uses of security technology to necessarily always be bad.
>

[snip: you don't like further restrictions on what people do with the
binaries]

I suppose this could be the achilles heel of the GPLv2, when a device
could be running perfectly free software, but the compiled work will only
run when signed with a private key of the manufacturer, this would create
a situation where the user/hacker cannot fix bugs on his own machine,
despite having full access to the source and in principle being able to
fix it. After installing the fixed software on the device it will stop
working.

Of course, a benevolent way of using this is to just check whether a
manufacturer supplied kernel is used for warranty reasons, but I suppose
some manufacturers will choose the simple option of not letting any other
software run on the device.

I suppose I'm not a blue eyed optimist (or brown eyed in my case), I can
see this being a real scenario :-(

Another way to fix this in GPLv3 would probably be to require the hardware
and firmware to allow the modified software to remain functional after
modification (barring newly introduced bugs).

Having said all this, GPLv2 is still a pretty good license, but it seems
there could be a real need for further reducing the likelyhood of (free)
software being locked up by DRM or (software) patents.

Cheers

Simon

Re: GPL V3 and Linux - Dead Copyright Holders

[Matthias Urlichs]

Hi, Linus Torvalds wrote:

> And that's really what I don't like. I believe that a software license 
> should cover the software it licenses, not how it is used or abused - even 
> if you happen to disagree with certain types of abuse.
> 
> I believe that hardware that limits what their users can do will die just 
> becuase being user-unfriendly is not a way to do successful business. Yes, 
> I'm a damned blue-eyed optimist

There's one problem with your approach. I probably agree with you about
the dying-out part, in th long term, BUT that leaves the current owners of
said hardware out in the cold.

Say I buy a FooBox. It's advertized as running Linux and foos reasonably
enough, thank you, but I'd like it to bar too, so I want to build a new
kernel and upload it to my box, thereby turning my FooBox into a
FooBarBox. Or maybe there's a bug, and the vendor says "no big deal"
(and I disagree).

I have a reasonable expectation to be able to do so. That's what free
software is all about (among other things). But if the box requires some
TPMish keysigning to be fed with an update, my system has just turned into
an expensive doorstop and/or I have no recourse, and no legal rights WRT
the vendor, whatsoever.

It's perfectly OK if the *owner* of the box says I can't do that and won't
give me the key (say, I'm an employee and the box is not supposed to run
stuff the IT honchos don't like). BUT NOT FOR ANYBODY ELSE.

-- 
Matthias Urlichs   |   {M:U} IT Design @ m-u-it.de   |  smurf@smurf.noris.de
Disclaimer: The quote was selected randomly. Really. | http://smurf.noris.de
 - -
Make it idiot-proof, and someone will breed a better idiot.
	-- Oliver Elphick

Re: GPL V3 and Linux - Dead Copyright Holders

[Florian Weimer]

* Simon Oosthoek:

> GPLv3-draft1:
>> (...)
>> Complete Corresponding Source Code also includes any encryption or
>> authorization codes necessary to install and/or execute the source
>> code of the work, perhaps modified by you, in the recommended or
>> principal context of use, such that its functioning in all
>> circumstances is identical to that of the work, except as altered by
>> your modifications. It also includes any decryption codes necessary
>> to access or unseal the work's output. Notwithstanding this, a code
>> need not be included in cases where use of the work normally implies
>> the user already has it.
>> (...)
>
> I'd interpret that as forcing people who try to hide their code or make 
> it difficult to get at the source code to not be able to do that.

I view it slightly different.  Suppose you produce a device which can
only boot images signed by a certain public key.  In this case, you
can give your users source code, but they can't change it and run it
on the device because the signature does not match.  This is a real
threat to software freedom.

Such technology already exists, see for an example:

<http://java.sun.com/products/jce/doc/guide/HowToImplAProvider.html>

The reasons for that are entirely obscure (because anyone can obtain a
certificate, there is no kind of quality control) and likely only
related to export regulations.  But a similar approach could be used
elsewhere, to exercise more control over which code can run on a
certain platform.

Re: GPL V3 and Linux - Dead Copyright Holders

[Thomas Horsten]

On Wed, 25 Jan 2006, Linus Torvalds wrote:

> And quite frankly, I don't see that changing. I think it's insane to
> require people to make their private signing keys available, for example.
> I wouldn't do it.

Linus, I think you are missing the point here. I see this provision as one
of the most important fixes in GPL v3. It might not be perfect in the way
it is presented in the draft, but the rationale is certainly laudable.

This closes a very serious loophole, that in certain situations renders
the GPLv2 almost worthless. The problem is embedded systems, but
especially cellphones.

A typical smartphone has a baseband processor (running the GSM stack under
a proprietary OS - let's call it the Baseband OS) and an application
processor (running the PDA features and GUI, for example Symbian OS or
Linux, let's call it the Application OS). Newer phones use a microkernel
architecture where the two OS's each run as tasks under a microkernel so
that a single CPU can be shared, but that's not relevant for this point.

The practice for many phone manufacturers is to heavily encrypt and sign
all ROM images that go on the phone. They do this for several reasons, but
mostly to prevent end-users from modifying the baseband code (where things
like SIM-locks and network locks are stored) or changing the IMEI number
of stolen phones.

However, most manufacturers extend this protectin to also encompass the
PDA-side code. This is understandable (to an extent) where the operating
system is proprietary or licensed (e.g. Symbian OS) and not available for
end-users to modify in any case.

But when Linux is used instead (something that most major handset
manufacturers are working on at the moment), this whole picture changes.
Suddenly the whole point of "Free Software" goes away. Because the handset
manufacturer modifies, adapts and uses the Linux kernel (for free),
without having to give anything back to the community. Yes, they have to
make their modified source available, but there is no way a casual user
(read: someone not a member of the phone-unlocking mafia) can update the
kernel with a modified, recompiled version.

But it goes further. The manufacturer is free to embed "trusted
computing" technology into the kernel (e.g. digital signing of all
executables that can run on the phone), as long as the encryption
technology is included in the kernel (they can even embed the
public key in the published sources, and you have no way of
obtaining the secret key) - thereby preventing the user from running ANY
code on the phone not pre-approved by the manufacturer.

Their main motivation for doing this is basically to protect the mobile
network operators' revenue, by preventing (or at least limiting)
installation of things such as VoIP clients and instant messenger systems
that could lead towards an "open TCP/IP" infrastructure that all the
operators fear so much.

The prime example of this is what is already implemented is Symbian OS. In
Version 9, used by the latest handsets from Nokia and Sony Ericsson, it is
impossible to install applications that are not signed by an independent
testing house, something that effectively locks small businesses and
independent hobby coders out of the market. There are even certain
"capabilities", needed for example to implement networking protocols, that
can only be signed if approved by the phone manufacturer.

The only way to prevent Linux from being abused in this way is to require
that the end-users have a method of installing the modified code on the
device for which it was intended (such as your phone). If they don't want
to disclose their primary signing keys, fine, but then they need to make
another method available of replacing any free software used in the device
with a modified (by the end-user) version. I understand the need to
protect the non-GPL'ed "baseband" OS from tampering, but it has to be
possible to replace the GPL'ed software. This is what this provision is
all about (for me), and I think it is an extremely important one.

Of course I have focused on cellphones, but the same is equally true for
other embedded devices, such as digital TV boxes (e.g. PVR's). If I want
to add a nice feature (and share the patch with my friends who bought the
same box) I should be free to do so. Isn't that what Free Software is all
about?

Thomas

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

[This is a repost of the private reply given that lkml wasn't cc'ed]

Thomas Horsten wrote:
>> But when Linux is used instead (something that most major handset
>> manufacturers are working on at the moment), this whole picture changes.
>> Suddenly the whole point of "Free Software" goes away. Because the handset
>> manufacturer modifies, adapts and uses the Linux kernel (for free),
>> without having to give anything back to the community. Yes, they have to
>> make their modified source available, but there is no way a casual user
>> (read: someone not a member of the phone-unlocking mafia) can update the
>> kernel with a modified, recompiled version.

And I would contend that even if they did it wouldn't make much of a
difference. In fact, my reading of GPLv3 is that it would be quite
straight-forward to create GPLv3-compliant consumer exclusion mechanisms.

Here's from an earlier posting I did on the LWN forum discussing the topic:
----------
Prior to booting kernel, let the firmware verify the kernel's signature.
Whether or not the signature is valid, boot the thing. But, if it isn't
valid, lock up or disable or just don't enable those hardware components
that control access to the "stuff" -- and do this in a way that requires
a hard reset for another pass at lock/don't-lock. Then package all the
proprietary/DRM software into individual applications. So everything just
runs as normal up to the time these apps try to access the hardware,
and now because it's disabled, the device is useless.

Using the words of the draft, there are no codes for you to "install and/
or execute" the work. In fact, you're free to install and execute any
kernel you wish. Plus, the kernel's "functioning in all circumstances is
identical". It's just that hardware beneath it just doesn't work any more,
and that, consequently, the proprietary applications running on top can't
do anything because the hardware is disabled.

There you have it, GPL3-compatible DRM -- to the best of my understanding
of course, and mind you I'm not a lawyer.
----------

Just to make sure there is no confusion: note that both signed and
unsigned kernels behave identically here. It's the user-space applications
now that fail when attempting to access some piece of hardware. The
classic case being a mmap'ed register window, therefore both the signed
and unsigned kernel can map it to user-space (i.e. no modification in
behavior for the GPLv3'ed kernel), but the applications' read/writes on
said registers don't work on the non-signed kernel.

Like I said in the LWN forum thread, I do believe things such as DRM
are worth fighting for, but I really think the GPL is the wrong venue.

>> The only way to prevent Linux from being abused in this way is to require
>> that the end-users have a method of installing the modified code on the
>> device for which it was intended (such as your phone). If they don't want
>> to disclose their primary signing keys, fine, but then they need to make
>> another method available of replacing any free software used in the device
>> with a modified (by the end-user) version. I understand the need to
>> protect the non-GPL'ed "baseband" OS from tampering, but it has to be
>> possible to replace the GPL'ed software. This is what this provision is
>> all about (for me), and I think it is an extremely important one.

And entirely useless if you ask me. A GPLv3 could dictate that you could
replace the kernel, but it isn't a hardware license nor a license for
user-space applications -- both of which's direct interaction could not
be dictated by the license of separate "work".

So, in the end, even if this were applied, you would find yourself in a
world where you can replace kernels, but that apart from having access
to running programs on a given architecture, play around with the RAM,
and access entirely unimportant hardware, you can't do very much.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Thomas Horsten]

On Sat, 28 Jan 2006, Karim Yaghmour wrote:

> Just to make sure there is no confusion: note that both signed and
> unsigned kernels behave identically here. It's the user-space applications
> now that fail when attempting to access some piece of hardware. The
> classic case being a mmap'ed register window, therefore both the signed
> and unsigned kernel can map it to user-space (i.e. no modification in
> behavior for the GPLv3'ed kernel), but the applications' read/writes on
> said registers don't work on the non-signed kernel.

That would be a dubious circumvention. Remember that the GPLv3 is still a
draft - the wording can (and should probably) be improved to make it clear
that the system as a whole must behave identically if a modified version
of the GPL'ed software is used.

> Like I said in the LWN forum thread, I do believe things such as DRM
> are worth fighting for, but I really think the GPL is the wrong venue.

It's a good place to start putting pressure on the OEM's. If they can
choose between heavy DRM'ed and closed hardware, and pay millions in
license fees, or get the software they need for free in return for
dropping the restrictions, some are bound to choose the free route. This
is where the fight begins.

Thomas

Re: GPL V3 and Linux - Dead Copyright Holders

[Peter Read]

> It's a good place to start putting pressure on the OEM's. If they can
> choose between heavy DRM'ed and closed hardware, and pay millions in
> license fees, or get the software they need for free in return for
> dropping the restrictions, some are bound to choose the free route. This
> is where the fight begins.
>

It'd be nice to think that licensing fees wouldn't make a difference
as people would understand that locking themselves into something
they're not allowed to exert any control over is not a good idea...  &
so oem's would have to chose due to revenue loss rather than simply
'licensing savings'.

Unfortunately I doubt we're anywhere near close enough yet, so I'll
get my head out of cloud nine...

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Thomas Horsten wrote:
> That would be a dubious circumvention. Remember that the GPLv3 is still a
> draft - the wording can (and should probably) be improved to make it clear
> that the system as a whole must behave identically if a modified version
> of the GPL'ed software is used.

As a software license, GPLv3 can dictate the usage rules for software
distributed under it, but it can't dictate the usage terms of hardware
and software independently developed (ex.: DRM'ed hardware and
proprietary user-space applications). No wording could erase that.
And what is suggest is not "circumvention", it's just not something
GPLv3 could cover.

> It's a good place to start putting pressure on the OEM's. If they can
> choose between heavy DRM'ed and closed hardware, and pay millions in
> license fees, or get the software they need for free in return for
> dropping the restrictions, some are bound to choose the free route. This
> is where the fight begins.

DRM is a symptom, not the root problem. The root problem is that someone
somewhere has been convinced that they are loosing money because of
"pirates". Sure, middle-men (record labels) have crafted the reality to
fit their needs. But the latter would not be able to operate would the
former not exist. There's only one thing "artists" will dislike more than
the fear of not making money: it's actually not making any. So if you
want to get rid of DRM, the solution is to not buy any DRM'ed stuff.
Every time you buy a region-coded DVD or a "protected" CD you are
feeding the DRM cycle. To stop it, just don't buy any of it.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Glauber de Oliveira Costa]

On 1/30/06, Karim Yaghmour <karim@opersys.com> wrote:
>
> Thomas Horsten wrote:
> > That would be a dubious circumvention. Remember that the GPLv3 is still a
> > draft - the wording can (and should probably) be improved to make it clear
> > that the system as a whole must behave identically if a modified version
> > of the GPL'ed software is used.
>
> As a software license, GPLv3 can dictate the usage rules for software
> distributed under it, but it can't dictate the usage terms of hardware
> and software independently developed (ex.: DRM'ed hardware and
> proprietary user-space applications). No wording could erase that.
> And what is suggest is not "circumvention", it's just not something
> GPLv3 could cover.

I may be missing the point here, (In case you're more than welcome to
correct me), but ... Why? Can't a software license restrict the usage
of the software? In which ways do you think the sentence "Don't use in
DRM'ed hardware" differs from sentences like  "Not allowed in country
X",  "Don't use for commercial purposes",  and other alikes ? I think
that saying in which hardware your software can or cannot run is a
pretty valid license term (without messing with the question about it
being the right thing to do here).

Besides that, I pretty much agree with the rest of your mail.

--
Free Software : Technology for a better world
=============================
Glauber de Oliveira Costa
jabber: glommer@jabber.org
=============================

Re: GPL V3 and Linux - Dead Copyright Holders

[Filip Brcic]

[-- Attachment #1: Type: text/plain, Size: 2766 bytes --]

Дана Monday 30 January 2006 20:43, Glauber de Oliveira Costa је написао(ла):
> On 1/30/06, Karim Yaghmour <karim@opersys.com> wrote:
> > As a software license, GPLv3 can dictate the usage rules for software
> > distributed under it, but it can't dictate the usage terms of hardware
> > and software independently developed (ex.: DRM'ed hardware and
> > proprietary user-space applications). No wording could erase that.
> > And what is suggest is not "circumvention", it's just not something
> > GPLv3 could cover.
>
> I may be missing the point here, (In case you're more than welcome to
> correct me), but ... Why? Can't a software license restrict the usage
> of the software? In which ways do you think the sentence "Don't use in
> DRM'ed hardware" differs from sentences like  "Not allowed in country
> X",  "Don't use for commercial purposes",  and other alikes ?

"Don't use in DRM'ed hardware" and "Don't use for commercial purposes" do 
differ from "Not allowed in country X". The first two are essentially the 
same (as far as I am aware, DRM has no use for F/OSS, it is made for 
commercial stuff). They protect the freedom of the software. The third 
(!country) is something different since it attacks the freedom of the 
software. Equivalents (or similar thoughts) would be "Don't use in DRM'ed 
hardware", "Don't use for commercial purposes" and "Use in any country you 
want".

> I think
> that saying in which hardware your software can or cannot run is a
> pretty valid license term (without messing with the question about it
> being the right thing to do here).

I agree that it is a valid term. I wouldn't like to see my program running on 
some obscure DRM'ed hardware when I made it to be free.

As far as I can see, Linus wants to allow usage of Linux in DRM'ed hardware 
(for ex. future mobile phones). He wants to allow usage, but he hopes that 
the customers would make companies disband DRM (and similar 
crypto/obscure/... stuff). If that is the case, I don't agree with his 
oppinion that the customers could do anything. Most of the "customers" have 
no problem with DRM, and look-a-likes. Most of the customers still use m$ win 
and office, and others think that os x is an free/open operating system.

> Besides that, I pretty much agree with the rest of your mail.

I do too, but I don't think that "To stop it, just don't buy any of it." is 
the solution of the problem. It would be a solution if most of the customers 
would do so.

-- 
Filip Brcic <brcha@users.sourceforge.net>
WWWeb: http://purl.org/NET/brcha/home/
Jabber: brcha@kdetalk.net
Jabber: brcha@elitesecurity.org
Jabber: fbrcic@gmail.com
ICQ# 40994923
Yahoo! brcha
MSN: brcha@users.sourceforge.net

[-- Attachment #2: Type: application/pgp-signature, Size: 320 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[Georg C. F. Greve]

[-- Attachment #1: Type: text/plain, Size: 1139 bytes --]

 || On Tue, 31 Jan 2006 02:12:17 -0200
 || Glauber de Oliveira Costa <glommer@gmail.com> wrote: 

 g> So, I think the main point is that free software and Linux has
 g> both to gain being together. And maybe saying no to GPLv3 may be a
 g> too disruptive move. GPLv3 is still a draft. Can't it be discussed
 g> a little among with Linus, some companies representants, and some
 g> guys from fsf?

Yes, indeed discussing is what FSF has allocated an entire year for.

While it is possible that this particular criticism was based on a
misunderstanding, which may be an indicator of need for language
improvement, all criticism should ideally be put into a comprehensive
form and submitted through the gplv3.fsf.org/comments system.

If Linus lacks the time to do it himself, maybe some people on this
list could help gather his points as possible submissions for
improvement?

Regards,
Georg

-- 
Georg C. F. Greve                                 <greve@fsfeurope.org>
Free Software Foundation Europe	                 (http://fsfeurope.org)
Join the Fellowship and protect your freedom!     (http://www.fsfe.org)

[-- Attachment #2: Type: application/pgp-signature, Size: 306 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Filip Brcic wrote:
> I do too, but I don't think that "To stop it, just don't buy any of it." is 
> the solution of the problem. It would be a solution if most of the customers 
> would do so.

Exactly. And this may indeed be a case where the non-techies have decided
for the techies. The tech community (or part of it in this case) cannot
change the world. It can certainly act as a catalyst, and boy has free
software been such a catalyst, but there are other forces at play here,
including consumers who not care about DRM and those that even understand/
accommodate it. Even if licenses prohibited the use of FLOSS in DRM'ed
hardware: where there's a need, there's demand, and where'd demand, there's
supply. IOW, we will continue seeing DRM'ed hardware for the foreseeable
future, and there's nothing any techie (or software license) can do about
it. So, I have to question myself with regards to what use it is to
encumber a successful license with draconian runtime restrictions when
said restrictions will not solve anything anytime in the future --
especially as said restrictions will not effectively block the use of
DRM hardware and corresponding user-space applications as I explained
earlier.

DRM is something worth fighting, but we need something that attacks the
root problem, not its symptoms. In comparison, GPLv2 was indeed
successful in that it attacked the root problem of software distribution
freedom. How it may leverage that by introducing restrictions on symptoms
of another problem still evades me.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Wed, 1 Feb 2006, Karim Yaghmour wrote:
> 
> DRM is something worth fighting, but we need something that attacks the
> root problem, not its symptoms. In comparison, GPLv2 was indeed
> successful in that it attacked the root problem of software distribution
> freedom. How it may leverage that by introducing restrictions on symptoms
> of another problem still evades me.

Side note: the reason GPLv2 is so successful at fighting the root problem 
of using copyright to fight restrictive copyrights is that it makes 
"interesting material" available under a license that forbids further 
restricting it.

I would suggest that anybody who wants to fight DRM practices seriously 
look at the equivalent angle. If you create interesting content, you can 
forbid that _content_ to ever be encrypted or limited. 

In other words, I personally think that the anti-DRM clause is much more 
sensible in the context of the Creative Commons licenses, than in software 
licenses. If you create valuable and useful content that other people want 
to be able to use (catchy tunes, funny animation, good icons), I would 
suggest you protect that _content_ by saying that it cannot be used in any 
content-protection schemes.

Afaik, all the Creative Commons licenses already require that you can't 
use technological measures to restrict the rigts you give with the CC 
licenses. The "Share Alike" license in particular requires all work based 
on it to also be shared alike, ie it has the "GPL feel" to it.

If enough interesting content is licensed that way, DRM eventually becomes 
marginalized. Yes, it takes decades, but that's really no different at all 
from how the GPL works. The GPL has taken decades, and it hasn't 
"marginalized" commercial proprietary software yet, but it's gotten to the 
point where fewer people at least _worry_ about it.

As long as you expect Disney to feed your brain and just sit there on your 
couch, Disney & co will always be able to control the content you see. DRM 
is the smallest part of it - the crap we see and hear every day 
(regardless of any protection) is a much bigger issue.

The GPL already requires source code (ie non-protected content). So the 
GPL already _does_ have an anti-DRM clause as far as the _software_ is 
concerned. If you want to fight DRM on non-software fronts, you need to 
create non-software content, and fight it _there_.

I realize that programmers are bad at content creation. So many 
programmers feel that they can't fight DRM that way. Tough. Spread the 
word instead. Don't try to fight DRM the wrong way.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Linus Torvalds wrote:
> If enough interesting content is licensed that way, DRM eventually becomes 
> marginalized. Yes, it takes decades, but that's really no different at all 
> from how the GPL works. The GPL has taken decades, and it hasn't 
> "marginalized" commercial proprietary software yet, but it's gotten to the 
> point where fewer people at least _worry_ about it.
...
> The GPL already requires source code (ie non-protected content). So the 
> GPL already _does_ have an anti-DRM clause as far as the _software_ is 
> concerned. If you want to fight DRM on non-software fronts, you need to 
> create non-software content, and fight it _there_.
> 
> I realize that programmers are bad at content creation. So many 
> programmers feel that they can't fight DRM that way. Tough. Spread the 
> word instead. Don't try to fight DRM the wrong way.

Bullseye.

It dawned on me after sending earlier emails that I should've mentioned the
responsibility of content creators, but I'm glad I didn't, you've summed
it up quite eloquently.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Mer, 2006-02-01 at 14:31 -0800, Linus Torvalds wrote:
> I realize that programmers are bad at content creation. So many 
> programmers feel that they can't fight DRM that way. Tough. Spread the 
> word instead. Don't try to fight DRM the wrong way.


Programmers tend to be interested in economics for some strange reason.
So read the analyses on DRM. Then stop worrying about the content
companies because they are the losers to the computer industry on this..

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Linus Torvalds wrote:
> 
> The GPL already requires source code (ie non-protected content). So the 
> GPL already _does_ have an anti-DRM clause as far as the _software_ is 
> concerned. If you want to fight DRM on non-software fronts, you need to 
> create non-software content, and fight it _there_.
> 

The point is not only getting access to the source code, but also being 
able to change it. Being able to freely study the code is only half of 
the beauty of the GPL. The other half, being able to change it, can be 
very effectively stopped using DRM.

I agree with your position that the GPL should not be used to prevent 
implementing a DRM system, but I like the fact that it's trying to 
prevent DRM:ing the code itself (e.g. requiring signed binaries to run).

The effect would probably be that no one can implement a DRM system that 
isn't easily broken, but that only demonstrates how flawed the idea of 
DRM is.

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Thu, 2 Feb 2006, Pierre Ossman wrote:
> 
> The point is not only getting access to the source code, but also being able
> to change it. Being able to freely study the code is only half of the beauty
> of the GPL. The other half, being able to change it, can be very effectively
> stopped using DRM.

No it cannot.

Sure, DRM may mean that you can not _install_ or _run_ your changes on 
somebody elses hardware. But it in no way changes the fact that you got 
all the source code, and you can make changes (and use their changes) to 
it. That requirement has always been there, even with plain GPLv2. You 
have the source.

The difference? The hardware may only run signed kernels. The fact that 
the hardware is closed is a _hardware_ license issue. Not a software 
license issue. I'd suggest you take it up with your hardware vendor, and 
quite possibly just decide to not buy the hardware. Vote with your feet. 
Join the OpenCores groups. Make your own FPGA's.

And it's important to realize that signed kernels that you can't run in 
modified form under certain circumstances is not at all a bad idea in many 
cases.

For example, distributions signing the kernel modules (that are 
distributed under the GPL) that _they_ have compiled, and having their 
kernels either refuse to load them entirely (under a "secure policy") or 
marking the resulting kernel as "Tainted" (under a "less secure" policy) 
is a GOOD THING.

Notice how the current GPLv3 draft pretty clearly says that Red Hat would 
have to distribute their private keys so that anybody sign their own 
versions of the modules they recompile, in order to re-create their own 
versions of the signed binaries that Red Hat creates. That's INSANE.

Btw, what about signed RPM archives? How well do you think a secure 
auto-updater would work if it cannot trust digital signatures?

I think a lot of people may find that the GPLv3 "anti-DRM" measures aren't 
all that wonderful after all.

Because digital signatures and cryptography aren't just "bad DRM". They 
very much are "good security" too.

Babies and bathwater..

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Emilio Jesús Gallego Arias]

Linus Torvalds <torvalds@osdl.org> writes:

> On Thu, 2 Feb 2006, Pierre Ossman wrote:
>> 
>> The point is not only getting access to the source code, but also being able
>> to change it. Being able to freely study the code is only half of the beauty
>> of the GPL. The other half, being able to change it, can be very effectively
>> stopped using DRM.
>
> No it cannot.

1.- Distribute a kernel with some DRM built-in under the GPL.

2.- Claim that such kernel is an effective technological measure to
    protect copyright. 

3.- You are no longer free to modify that kernel, (removing the DRM
    module) or you can be sued under the DMCA, for circumventing an
    effective technological measure. It doesn't matter in what
    hardware are you going to run such kernel. The DMCA implicitly
    imposes an additional restriction to the GPL, but as the
    restriction is not imposed directly by the copyright owner, but by
    the law, it's OK as far the GPL is concerned.

Regards,

Emilio

Re: GPL V3 and Linux - Dead Copyright Holders

[Helge Hafting]

Emilio Jesús Gallego Arias wrote:

>Linus Torvalds <torvalds@osdl.org> writes:
>
>  
>
>>On Thu, 2 Feb 2006, Pierre Ossman wrote:
>>    
>>
>>>The point is not only getting access to the source code, but also being able
>>>to change it. Being able to freely study the code is only half of the beauty
>>>of the GPL. The other half, being able to change it, can be very effectively
>>>stopped using DRM.
>>>      
>>>
>>No it cannot.
>>    
>>
>
>1.- Distribute a kernel with some DRM built-in under the GPL.
>
>2.- Claim that such kernel is an effective technological measure to
>    protect copyright. 
>
>3.- You are no longer free to modify that kernel, (removing the DRM
>    module) or you can be sued under the DMCA, for circumventing an
>    effective technological measure. It doesn't matter in what
>    hardware are you going to run such kernel. The DMCA implicitly
>    imposes an additional restriction to the GPL, but as the
>    restriction is not imposed directly by the copyright owner, but by
>    the law, it's OK as far the GPL is concerned.
>  
>
This can't legally happen.  Do the DMCA prevent "circumventing"
even when you have the legal right to make copies of the content?
(If so, then the music industry breaks DMCA when manufacturing
their protected CDs from a protected master . . .)

If so, then step (1) is illegal because it breaks with the GPL. 
Remember, when you distribute something under the GPL, you cannot
impose restrictions.  It is well established that you can't link in
something with a more restrictive commercial licence, for example. 
I think adding that DRM falls in the same trap - if the DMCA really
impose this additional restriction (because DRM-breaking is illegal
even when content copying is not) then you are not allowed to
add that restriction. 

Someone distributing DRM-protected kernels are breaking
copyrigth law then, if the DMCA is so strict.

Helge Hafting

Re: GPL V3 and Linux - Dead Copyright Holders

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 578 bytes --]

On Thu, 02 Feb 2006 11:30:30 +0100, Helge Hafting said:

> This can't legally happen.  Do the DMCA prevent "circumventing"
> even when you have the legal right to make copies of the content?

As a matter of fact, yes.  See 17 USC 1201 and Skylarov v. Adobe - the fact
that a user may have totally legitimate fair-use access to the protected
content is *not* an exemption to the DMCA's anti-circumvention clause.

> (If so, then the music industry breaks DMCA when manufacturing
> their protected CDs from a protected master . . .)

Oohh.. an interesting point.. ;)

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

Re: GPL V3 and Linux - Dead Copyright Holders

[James Bruce]

Emilio Jesús Gallego Arias wrote:
>... 
> 1.- Distribute a kernel with some DRM built-in under the GPL.
> 
> 2.- Claim that such kernel is an effective technological measure to
>     protect copyright. 

You forgot:

2.5- Due to the DMCA, the code now has an additional restriction on
      top of what is already in its license, the GPL v2.  The GPL v2
      forbids additional restrictions, and thus the resulting work
      cannot be distributed.

> 3.- You are no longer free to modify that kernel, (removing the DRM
>     module) or you can be sued under the DMCA, for circumventing an
>     effective technological measure. It doesn't matter in what
>     hardware are you going to run such kernel. The DMCA implicitly
>     imposes an additional restriction to the GPL, but as the
>     restriction is not imposed directly by the copyright owner, but by
>     the law, it's OK as far the GPL is concerned.

If the DRM author(s) are the ones claiming the DRM is an "effective 
technological measure", then they are the ones imposing an additional 
restriction.  Those authors are the ones who can be sued, not the end 
users of the kernel+DRM.  If someone else makes the claim, it carries no 
weight at all, because they are not the author or copyright owner.

Also, remember that the GPL is about DISTRIBUTION only.  You are always 
free to modify whatever you want.  So your #3 is just plain wrong, as 
you still can modify the kernel.  What you can't do is distribute that 
modified version, although that's a meaningless since #2.5 shows you 
couldn't distribute the unmodified version anyway.

The modification vs. distribution point cannot be emphasized enough, as 
far too many people miss this important distinction when thinking about 
the GPL.  Nothing in GPL software becomes magically illegal at any 
point; The only thing that can be legal or not is distribution.  I am 
free to mix Linux, Open Solaris, GCC, and BSD code with the advertising 
clause, and top it off with GNU FDL docs, and do so to my hearts' 
content.  The only thing I *can't* do with that abomination is 
distribute it.

  - Jim Bruce

Re: GPL V3 and Linux - Dead Copyright Holders

[Emilio Jesús Gallego Arias]

James Bruce <bruce@andrew.cmu.edu> writes:

> Emilio Jesús Gallego Arias wrote:
>> ... 1.- Distribute a kernel with some DRM built-in under the GPL.
>> 2.- Claim that such kernel is an effective technological measure to
>>     protect copyright. 
>
> You forgot:
>
> 2.5- Due to the DMCA, the code now has an additional restriction on
>      top of what is already in its license, the GPL v2.  The GPL v2
>      forbids additional restrictions, and thus the resulting work
>      cannot be distributed.

Ok, so to add DRM to GPLed software, the copyright holder has to state
that the DMCA does not apply to such software? Or does the GPL state
that? 

Quoting the GPL:

6.- [...] You may not impose any further restrictions on the recipients'
  exercise of the rights granted herein. [...]

The point is that it is not the copyright holder who is imposing the
restrictions, is the law. For example, the law may impose some export
restrictions, would that void the GPL?

>> 3.- You are no longer free to modify that kernel, (removing the DRM
                                 ^^^^^^
I meant modify and distribute, sorry.
>> [...]
>
> If the DRM author(s) are the ones claiming the DRM is an "effective
> technological measure", then they are the ones imposing an additional

They are only a claiming, not restricting. The restriction would come
from the law when a judge decides that the code in question is an
"effective technological measure".
> restriction.  Those authors are the ones who can be sued, not the end
> users of the kernel+DRM.  If someone else makes the claim, it carries
Ok, I mean a user who tries to exercise all the rights stated in the
GPL, including distribution of the modified code.
> no weight at all, because they are not the author or copyright owner.

Regards,

Emilio

Re: GPL V3 and Linux - Dead Copyright Holders

[Helge Hafting]

Emilio Jesús Gallego Arias wrote:

>James Bruce <bruce@andrew.cmu.edu> writes:
>
>  
>
>>Emilio Jesús Gallego Arias wrote:
>>    
>>
>>>... 1.- Distribute a kernel with some DRM built-in under the GPL.
>>>2.- Claim that such kernel is an effective technological measure to
>>>    protect copyright. 
>>>      
>>>
>>You forgot:
>>
>>2.5- Due to the DMCA, the code now has an additional restriction on
>>     top of what is already in its license, the GPL v2.  The GPL v2
>>     forbids additional restrictions, and thus the resulting work
>>     cannot be distributed.
>>    
>>
>
>Ok, so to add DRM to GPLed software, the copyright holder has to state
>that the DMCA does not apply to such software? Or does the GPL state
>that? 
>  
>
This isn't only about DRM protecting your distributed kernel.
Lets say you want to make a linux-driven home entertainment
device.  And you add DRM - not to protect the kernel you don't
really care about, but in order to use protected content in a
restricted fashion. Perhaps your business also sell DVDs.

Saying that the DRM doesn't apply to the kernel itself won't
help.  Users may, in this case, want to alter the DRM so it
doesn't restrict their use of Cds, DVDs etc.  That however
breaks with the DMCA - even if you allow unrestricted source
code modifications.  So you cannot legally distribute a kernel
with DRM - because DRM comes with a "no tampering" law
which don't work with the GPL.  Businesses can still add
DRM stuff in a proprietary userland app though.

The fact that DMCA law is a restriction imposed by
government rather than the distributor makes no difference.
The distributor implicitly imposes restrictions by linking in DRM sw, just
as the distributor would implicitly impose some restrictions by
linking a proprietary-licenced object into the kernel.

Helge Hafting




>Quoting the GPL:
>
>6.- [...] You may not impose any further restrictions on the recipients'
>  exercise of the rights granted herein. [...]
>
>The point is that it is not the copyright holder who is imposing the
>restrictions, is the law. For example, the law may impose some export
>restrictions, would that void the GPL?
>
>  
>
>>>3.- You are no longer free to modify that kernel, (removing the DRM
>>>      
>>>
>                                 ^^^^^^
>I meant modify and distribute, sorry.
>  
>
>>>[...]
>>>      
>>>
>>If the DRM author(s) are the ones claiming the DRM is an "effective
>>technological measure", then they are the ones imposing an additional
>>    
>>
>
>They are only a claiming, not restricting. The restriction would come
>from the law when a judge decides that the code in question is an
>"effective technological measure".
>  
>
>>restriction.  Those authors are the ones who can be sued, not the end
>>users of the kernel+DRM.  If someone else makes the claim, it carries
>>    
>>
>Ok, I mean a user who tries to exercise all the rights stated in the
>GPL, including distribution of the modified code.
>  
>
>>no weight at all, because they are not the author or copyright owner.
>>    
>>
>
>Regards,
>
>Emilio
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/
>  
>

Re: GPL V3 and Linux - Dead Copyright Holders

[Emilio Jesús Gallego Arias]

Helge Hafting <helge.hafting@aitel.hist.no> writes:

> [...]
> This isn't only about DRM protecting your distributed kernel.
> Lets say you want to make a linux-driven home entertainment
> device.  And you add DRM - not to protect the kernel you don't
> really care about, but in order to use protected content in a
> restricted fashion. Perhaps your business also sell DVDs.

Yes, I was thinking about this situation, not talking about DRMing the
kernel.

I thought that this case wasn't already covered by the GPLv2, and was
one of the points addressed in the GPLv3.

> [...]
> The fact that DMCA law is a restriction imposed by
> government rather than the distributor makes no difference.
> The distributor implicitly imposes restrictions by linking in DRM sw, just
> as the distributor would implicitly impose some restrictions by
> linking a proprietary-licenced object into the kernel.

Umm, the interesting question here if what happens in countries that
haven't implemented the DMCA.

So it seems that this is kinda offtopic for l-k.

Thanks,

Emilio

Re: GPL V3 and Linux - Dead Copyright Holders

[Helge Hafting]

Emilio Jesús Gallego Arias wrote:

>Umm, the interesting question here if what happens in countries that
>haven't implemented the DMCA.
>  
>
In such countries, linking in DRM will be okay. Because then no law
prevents others from removing the DRM at will.  Of course distributing
such a kernel to the U.S. or other DMCA countries won't be possible.

Helge Hafting

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> 1.- Distribute a kernel with some DRM built-in under the GPL.
>
> 2.- Claim that such kernel is an effective technological measure to
>     protect copyright.
>
> 3.- You are no longer free to modify that kernel, (removing the DRM
>     module) or you can be sued under the DMCA, for circumventing an
>     effective technological measure. It doesn't matter in what
>     hardware are you going to run such kernel. The DMCA implicitly
>     imposes an additional restriction to the GPL, but as the
>     restriction is not imposed directly by the copyright owner, but by
>     the law, it's OK as far the GPL is concerned.

	You can't do that. The restriction is not imposed by the law, it's imposed
by the the copyright owner the instant he added an effective technological
measure to protect copyright to the GPL'd code.

	GPLv2 code *cannot* contain any license or copyright enforcement
mechanisms. It can contain code that appears to be such a thing only
provided that the legal position is not that it's such a thing.

	For example, you can make a kernel that will refuse to load kernel modules
that aren't licensed under the GPL. But you cannot prevent anyone from
removing that logic if they want to.

	*If* you add a license enforcement mechanism to some code or *you* declare
that existing code is such a mechanism, then *you* are imposing the
additional restrictions.

	The idea of it being imposed "directly by the copyright owner" is just
something you made up. The GPL does not distinguish between direct and
indirect impositions. In fact, it goes out of its way to make it impossible
to indirectly impose additional conditions. See, for example, section 7.

> The point is that it is not the copyright holder who is imposing the
> restrictions, is the law. For example, the law may impose some export
> restrictions, would that void the GPL?

	Again, see section 6. If you cannot allow unrestricted modification, then
you cannot comply with the GPL. If you cannot comply with the GPL, then you
do not get the rights the GPL might give you. If you need the GPL to give
you the right to distribute, and you cannot comply with the GPL (even though
it's through no fault of your own), then you cannot distribute.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1287 bytes --]



On Thu, 2 Feb 2006, Emilio Jesús Gallego Arias wrote:
> 
> 1.- Distribute a kernel with some DRM built-in under the GPL.
> 
> 2.- Claim that such kernel is an effective technological measure to
>     protect copyright. 
> 
> 3.- You are no longer free to modify that kernel, (removing the DRM
>     module) or you can be sued under the DMCA, for circumventing an
>     effective technological measure.

I don't believe that for a second.

The law is not a mindless computer that just follows orders. It's 
interpreted by human beings.

And the DMCA has been interpreted already in real lawsuits to mean that 
"circumvention" is not "anything you say is circumvention". See the 
decision on garage door openers, for example.

Besides, the people who inserted the DRM code explicitly gave you 
permission to modify it, so the whole point is moot. There's no 
"circumvention".

The fact is, you can always come up with made-up problems, and try to 
solve them. That's what "Throwing out the baby with the bath-water" 
_means_ for christ sake! Using a solution that has _known_ problems 
(GPLv3) for a problem that you made up and don't even know it's real.

In engineering, it's called over-desiging, and it's stupid. In law, it's 
called "billable hours", and it's encouraged.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Emilio Jesús Gallego Arias]

Linus Torvalds <torvalds@osdl.org> writes:

> Besides, the people who inserted the DRM code explicitly gave you 
> permission to modify it, so the whole point is moot. There's no 
> "circumvention".

Yes, it is mostly clear, and indeed GPL3 seems a little bit over
engineered, but that doesn't mean that GPL2 could not have any
loopholes:

1. Release a kernel with builtin DRM for video. (For example HDCP [1])
   Such DRM implementation is released under the GPL2 by copyright
   holder A.

2. Distribute a modified kernel without DRM. Copyright holder A gave
   you permission to do so, by the GPL2, everything is OK.

3. People can backup videos from copyright holder B using the modified
   kernel.

4. Copyright holder B can sue you under the DMCA, for circumventing an
   effective technological measure. It doesn't matter whatever license 
   copyright holder A gave you.

Regards,

Emilio

Footnotes: 
[1]  High-Bandwidth Digital Content Protection. Currently you would
never get a license for a GPL implementation, but it's used as an
hypothetical example.

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Linus Torvalds wrote:
> On Thu, 2 Feb 2006, Pierre Ossman wrote:
>   
>> The point is not only getting access to the source code, but also being able
>> to change it. Being able to freely study the code is only half of the beauty
>> of the GPL. The other half, being able to change it, can be very effectively
>> stopped using DRM.
>>     
>
> No it cannot.
>
> Sure, DRM may mean that you can not _install_ or _run_ your changes on 
> somebody elses hardware. But it in no way changes the fact that you got 
>   

I don't consider things I've bought to be somebody elses hardware. The 
whole attitude of the big manufacturer that kindly gives me permission 
to use their product only how they see fit is very disgusting to me.

> The difference? The hardware may only run signed kernels. The fact that 
> the hardware is closed is a _hardware_ license issue. Not a software 
> license issue. I'd suggest you take it up with your hardware vendor, and 
> quite possibly just decide to not buy the hardware. Vote with your feet. 
> Join the OpenCores groups. Make your own FPGA's.
>   

I'm concerned that in a few years time such systems will be rare and 
hard to come by (possibly even illegal). I find such system pissing all 
over the spirit of the GPL. To me, the GPL has always been about the 
freedom of modifying (in place, not making a clone).

It's a fine line before we are in the territory of restricting what 
software can be used for. But for me this is not about restricting their 
rights as much as it is preventing them from restricting mine.

> And it's important to realize that signed kernels that you can't run in 
> modified form under certain circumstances is not at all a bad idea in many 
> cases.
>
> For example, distributions signing the kernel modules (that are 
> distributed under the GPL) that _they_ have compiled, and having their 
> kernels either refuse to load them entirely (under a "secure policy") or 
> marking the resulting kernel as "Tainted" (under a "less secure" policy) 
> is a GOOD THING.
>   

I dislike the former but the latter is acceptable (and, as you say, in 
some cases desirable). There is a big difference between refusing to run 
and printing/logging warnings.

> Notice how the current GPLv3 draft pretty clearly says that Red Hat would 
> have to distribute their private keys so that anybody sign their own 
> versions of the modules they recompile, in order to re-create their own 
> versions of the signed binaries that Red Hat creates. That's INSANE.
>
> Btw, what about signed RPM archives? How well do you think a secure 
> auto-updater would work if it cannot trust digital signatures?
>
>   

I'm arguing the principle here, not the wording of the current draft. 
Signatures that are required for execution should be covered, those that 
result in warnings should not be. Imagine the shit storm if Red Hat 
decided to ship an rpm that didn't allow packages that weren't signed by 
them.

It's basically about control. I do not find it reasonable to allow the 
vendor control of what goes or not on systems I've bought. They're free 
to put systems in place so they can detect that I've fiddled with it so 
they can deny me support. But if they want to make a completely closed 
system then they'll have to develop it on their own and not use my code. 
"Look but don't touch" is not sufficient for me.

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Gene Heskett]

On Thursday 02 February 2006 04:37, Pierre Ossman wrote:
>Linus Torvalds wrote:
>> On Thu, 2 Feb 2006, Pierre Ossman wrote:
>>> The point is not only getting access to the source code, but also
>>> being able to change it. Being able to freely study the code is
>>> only half of the beauty of the GPL. The other half, being able to
>>> change it, can be very effectively stopped using DRM.
>>
>> No it cannot.
>>
>> Sure, DRM may mean that you can not _install_ or _run_ your changes
>> on somebody elses hardware. But it in no way changes the fact that
>> you got
>
>I don't consider things I've bought to be somebody elses hardware. The
>whole attitude of the big manufacturer that kindly gives me permission
>to use their product only how they see fit is very disgusting to me.
>
>> The difference? The hardware may only run signed kernels. The fact
>> that the hardware is closed is a _hardware_ license issue. Not a
>> software license issue. I'd suggest you take it up with your
>> hardware vendor, and quite possibly just decide to not buy the
>> hardware. Vote with your feet. Join the OpenCores groups. Make your
>> own FPGA's.
>
>I'm concerned that in a few years time such systems will be rare and
>hard to come by (possibly even illegal). I find such system pissing
> all over the spirit of the GPL.

And I too see this scenario developing as the years go by, prodded to 
keep it in motion at every twitch of a muscle to do other wise by the 
likes of M$ because it cements the last brick into his domination of 
the world scene.  It scares me bad enough to keep my powder dry & in 
good supply if you get my drift.  DRM-less hardware will vanish from 
the supply chain unless smuggled in from au or nz.  And I do mean 
smuggled, with severe penalties for being caught at it.

> To me, the GPL has always been about 
> the freedom of modifying (in place, not making a clone).
>
>It's a fine line before we are in the territory of restricting what
>software can be used for. But for me this is not about restricting
> their rights as much as it is preventing them from restricting mine.
>
Right on.

>> And it's important to realize that signed kernels that you can't run
>> in modified form under certain circumstances is not at all a bad
>> idea in many cases.
>>
>> For example, distributions signing the kernel modules (that are
>> distributed under the GPL) that _they_ have compiled, and having
>> their kernels either refuse to load them entirely (under a "secure
>> policy") or marking the resulting kernel as "Tainted" (under a "less
>> secure" policy) is a GOOD THING.
>
>I dislike the former but the latter is acceptable (and, as you say, in
>some cases desirable). There is a big difference between refusing to
> run and printing/logging warnings.
>
>> Notice how the current GPLv3 draft pretty clearly says that Red Hat
>> would have to distribute their private keys so that anybody sign
>> their own versions of the modules they recompile, in order to
>> re-create their own versions of the signed binaries that Red Hat
>> creates. That's INSANE.

Where are the guys in the white jackets when we need them?

>> Btw, what about signed RPM archives? How well do you think a secure
>> auto-updater would work if it cannot trust digital signatures?
>
>I'm arguing the principle here, not the wording of the current draft.
>Signatures that are required for execution should be covered, those
> that result in warnings should not be. Imagine the shit storm if Red
> Hat decided to ship an rpm that didn't allow packages that weren't
> signed by them.
>
>It's basically about control. I do not find it reasonable to allow the
>vendor control of what goes or not on systems I've bought. They're
> free to put systems in place so they can detect that I've fiddled
> with it so they can deny me support. But if they want to make a
> completely closed system then they'll have to develop it on their own
> and not use my code. "Look but don't touch" is not sufficient for me.

Amen.  And on that point, we need a better method to detect that they 
have 'borrowed' FOSS code, and an expensive lawsuit settled in FOSS 
favor to set an enforcement example, but I have no imagination of a 
method they couldn't just as easily remove IF they had a coder worthy 
of the name of coder who wanted to be so larsonous.

>Rgds
>Pierre
>
>
>-
>To unsubscribe from this list: send the line "unsubscribe
> linux-kernel" in the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/

-- 
Cheers, Gene
People having trouble with vz bouncing email to me should add the word
'online' between the 'verizon', and the dot which bypasses vz's
stupid bounce rules.  I do use spamassassin too. :-)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2006 by Maurice Eugene Heskett, all rights reserved.

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Iau, 2006-02-02 at 01:00 -0800, Linus Torvalds wrote:
> Sure, DRM may mean that you can not _install_ or _run_ your changes on 
> somebody elses hardware. 

Last time I checked the Xbox was owned by the person who bought it. Xbox
Linux hits this problem today. So it may affect "your hardware" too
unless you make hardware, which is an unusual and privileged position.

That isn't to say the GPLv3 approach is neccessarily a solution at all.

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Thu, 2 Feb 2006, Alan Cox wrote:

> On Iau, 2006-02-02 at 01:00 -0800, Linus Torvalds wrote:
> > Sure, DRM may mean that you can not _install_ or _run_ your changes on 
> > somebody elses hardware. 
> 
> Last time I checked the Xbox was owned by the person who bought it. Xbox
> Linux hits this problem today. So it may affect "your hardware" too
> unless you make hardware, which is an unusual and privileged position.

Ok, now replace "hardware" by "software", and replace DRM by 
"proprietary", and what's the difference?

The fact is, if you buy proprietary software, you cannot make it do 
everything you want, regardless of of whether you "own" it or not. The 
creator of the software may have designed it so that it only does certain 
things.

Tough. The solution: use open source software.

The same holds true for hardware. If you buy proprietary hardware, you 
cannot make it do everything you want, whether you "own" it or not. The 
manufacturer of the hardware may have designed it so that it only does 
certain things.

Tough. The solution: use open hardware.

The solution is NOT to create a software license that is obviously not 
usable. And the GPLv3 really _is_ obviously not usable for the kernel, 
because it creates insane situations whether the hardware is open or 
closed.

In other words, the problem you state is a problem. But it has nothing to 
do with the GPLv3.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Linus Torvalds wrote:
> On Thu, 2 Feb 2006, Alan Cox wrote:
>
>   
>> On Iau, 2006-02-02 at 01:00 -0800, Linus Torvalds wrote:
>>     
>>> Sure, DRM may mean that you can not _install_ or _run_ your changes on 
>>> somebody elses hardware. 
>>>       
>> Last time I checked the Xbox was owned by the person who bought it. Xbox
>> Linux hits this problem today. So it may affect "your hardware" too
>> unless you make hardware, which is an unusual and privileged position.
>>     
>
> Ok, now replace "hardware" by "software", and replace DRM by 
> "proprietary", and what's the difference?
>
> The fact is, if you buy proprietary software, you cannot make it do 
> everything you want, regardless of of whether you "own" it or not. The 
> creator of the software may have designed it so that it only does certain 
> things.
>
> Tough. The solution: use open source software.
>
> The same holds true for hardware. If you buy proprietary hardware, you 
> cannot make it do everything you want, whether you "own" it or not. The 
> manufacturer of the hardware may have designed it so that it only does 
> certain things.
>
> Tough. The solution: use open hardware.
>
>   

So taking open software and closed hardware and combining it into 
something that I cannot modify is ok by you? But since you support GPLv2 
I take it you do not find it ok to take open software and closed 
software and combine that into something that I cannot modify. 
Personally, I consider both equal violations of the rights I seek to 
protect when I license my code under the GPL.

> The solution is NOT to create a software license that is obviously not 
> usable. And the GPLv3 really _is_ obviously not usable for the kernel, 
> because it creates insane situations whether the hardware is open or 
> closed.
>   

Just to make things clear, are you only having problems with the risk of 
forcing everyone to hand out their signing keys, or are you also opposed 
to preventing the hardware scenario above (provided it could be done 
without side-effects)?

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Thu, 2 Feb 2006, Pierre Ossman wrote:
> 
> So taking open software and closed hardware and combining it into something
> that I cannot modify is ok by you?

But you CAN modify the software part of it. You can run it on other 
hardware.

It boils down to this: we wrote the software. That's the only part _I_ 
care about, and perhaps (at least to me) more importantly, because it's 
the only part we created, it's the only part that I feel we have a moral 
right to control.

I _literally_ feel that we do not - as software developers - have the 
moral right to enforce our rules on hardware manufacturers. We are not 
crusaders, trying to force people to bow to our superior God. We are 
trying to show others that co-operation and openness works better.

That's my standpoint, at least. Always has been. It's the reason I
chose the GPL in the first place (and it's the exact same reason that I 
wrote the original Linux copyright license). I do _software_, and I 
license _software_.

And I realize that others don't always agree with me. That's fine. You 
don't have to. But I licensed my project under a license _I_ agreed with, 
which is the GPLv2. Others who feel differently can license under their 
own licenses. Including, very much, the GPLv3.

I'm not arguing against the GPLv3. 

I'm arguing that the GPLv3 is wrong for _me_, and it's not the license I 
ever chose.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Linus Torvalds wrote:
> I _literally_ feel that we do not - as software developers - have the 
> moral right to enforce our rules on hardware manufacturers. We are not 
> crusaders, trying to force people to bow to our superior God. We are 
> trying to show others that co-operation and openness works better.
>
>   

Then I have to ask, why GPL and not a BSD license? GPL is after all,
forcing our beliefs onto anyone who wishes to benefit from our work.

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Lennart Sorensen]

On Thu, Feb 02, 2006 at 06:54:47PM +0100, Pierre Ossman wrote:
> Then I have to ask, why GPL and not a BSD license? GPL is after all,
> forcing our beliefs onto anyone who wishes to benefit from our work.

The GPL enforced the view on free software, the BSD license does not.
The BSD license lets you do whatever you want pretty much.

If you want to make money from linux, you can do so by being better than
other people at solving problems and making enhancements.  With BSD all
it takes is keeping secrets.  So if you are very good at what you do,
you will be perfectly happy dealing with GPL code and making money
writing enhancements for people, while if you aren't very good, you
could just go write some software using BSD licensed code and sell that,
preferably taking out a software patent while you are at it to prevent
someone else from writing another implementation.

Len Sorensen

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Lennart Sorensen wrote:
> On Thu, Feb 02, 2006 at 06:54:47PM +0100, Pierre Ossman wrote:
>   
>> Then I have to ask, why GPL and not a BSD license? GPL is after all,
>> forcing our beliefs onto anyone who wishes to benefit from our work.
>>     
>
> The GPL enforced the view on free software, the BSD license does not.
> The BSD license lets you do whatever you want pretty much.
>
>   

I am aware of the difference between GPL and BSD. My point was that if
Linus feels that we should not enforce our rules on others then why does
he prefer a license that does just that? If people will come around just
by seeing how well the community works then BSD should be sufficient, or
even public domain.

I don't share this view, which is why I like the DRM ideas in GPLv3
which close something I see as a loophole in GPLv2.

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Horst von Brand]

Pierre Ossman <drzeus-list@drzeus.cx> wrote:
> Lennart Sorensen wrote:
> > On Thu, Feb 02, 2006 at 06:54:47PM +0100, Pierre Ossman wrote:

[...]

> >> Then I have to ask, why GPL and not a BSD license? GPL is after all,
> >> forcing our beliefs onto anyone who wishes to benefit from our work.

Nobody is forcing anybody to use Linux. If you don't agree, use something
else. Which probably has its own strings attached...

> > The GPL enforced the view on free software, the BSD license does not.
> > The BSD license lets you do whatever you want pretty much.

> I am aware of the difference between GPL and BSD. My point was that if
> Linus feels that we should not enforce our rules on others then why does
> he prefer a license that does just that? If people will come around just
> by seeing how well the community works then BSD should be sufficient, or
> even public domain.

> I don't share this view, which is why I like the DRM ideas in GPLv3
> which close something I see as a loophole in GPLv2.

You are free to create your own kernel under your rules, and see how well
that one does. Linux is successful (at least in part) because people who
work on it on the whole agree with its GPLv2, and so do people who use it.

In any case, the split you present is quite something... either BSD's most
everything goes or adding draconian no-DRM-ever-anywhere rules, GPLv3
style.
-- 
Dr. Horst H. von Brand                   User #22616 counter.li.org
Departamento de Informatica                     Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria              +56 32 654239
Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Thu, 2 Feb 2006, Pierre Ossman wrote:
> Linus Torvalds wrote:
> > I _literally_ feel that we do not - as software developers - have the 
> > moral right to enforce our rules on hardware manufacturers. We are not 
> > crusaders, trying to force people to bow to our superior God. We are 
> > trying to show others that co-operation and openness works better.
> 
> Then I have to ask, why GPL and not a BSD license? GPL is after all,
> forcing our beliefs onto anyone who wishes to benefit from our work.

Yes, a lot of people see the GPL as a "crusading" license, and I think 
that's partly because the FSF really has been acting like a crusader.

But I think that one of the main reasons Linux has been successful is that 
I don't think that the Linux community really is into crusading (some 
small parts of it are, but it's not the main reason). I think Linux has 
made the GPL more "socially acceptable", by being a hell of a lot less 
religious about it than the FSF was.

So to me, the point of the GPL is not the "convert the infidels" logic, 
but something totally different:

 - "quid pro quo"

   This is where I started out. My initial reason for my original license 
   (which was also "you must make changes available under the same 
   license") was not crusading, but simple reciprocity. I give out source 
   code - you can use it if you reciprocate.

   In other words, to me, the GPL "give back source" is an issue of 
   fairness. I don't ask for anything more than I give. I ask for source 
   code and the ability to incorporate your changes back into _my_ use, 
   but I don't want to limit _your_ use in any way.

   So in my worldview - not as a crusader - the GPLv2 is _fair_. It asks 
   others to give back exactly what I myself offer: the source code to 
   play with. I don't ask for control over their other projects (be they 
   hardware or software), and I don't ask for control over copyrights 
   (in the kernel, people are _encouraged_ to keep their copyrights, 
   rather than signing them over to me).

   I only ask for exact reciprocity of what I give: the license for me to 
   freely use the changes to source code that I initiated.

   The GPLv3 fundamentally changes that balance, in my opinion. It asks 
   for more than it gives. It no longer asks for just source back, it asks 
   for _control_ over whatever system you used the source in.

See? I think the GPLv3 makes _perfect_ sense as a conversion tool. But as 
a "please reciprocate in kind" tool, the GPLv2 is better.

Now, my very earliest original license (and the GPLv2) fit my notion of 
reciprocity, and as mentioned, that was the reason I _originally_ selected 
that over the BSD license. However, over time, having seen how things 
evolve, I've come to appreciate another aspect of the GPLv2, which is why 
I would never put a project I personally really cared about under the BSD 
license:

 - encouraging merging

   I've come to believe that the BSD license is not a "sustainable" 
   license, because while it encourages (and allows) forking even more 
   than the GPL does, it does not encourage merging the forks back.

   And I've come to the private conclusion that the real value of a fork 
   is lost if you don't have the ability to merge back the end result. Not 
   that all forks should be merged back - most forks are dead ends - but 
   the firm ability to merge back _if_ it turns out to be something other 
   than a dead end.

   The GPL guarantees you the right to both fork _and_ merge the result 
   back - equally, and on both sides. That makes it sustainable. In 
   contrast, the BSD license encourages forking, but also allows for not 
   merging back, and that means that if the project ever gets to the point 
   where there are economic or political reasons to diverge, it _will_ 
   eventually diverge, and there is no counter-acting force at all.

   Now, not all projects have any economic incentives to diverge: there 
   are good reasons to stay on one base, and the costs of forking are 
   bigger than the advantages. So projects like Apache and sendmail have 
   worked fine - the pain of being different (when you're "just" a network 
   service) is generally much higher than the gain of differentiation.

   But just about anywhere else, the "cohesion" of a BSD-licensed project 
   is just lower. You'll have somebody make a commercial release of it, 
   and they'll spend a bit more effort on it, and eventually the original
   freely licensed project will become immaterial.

So long-term, I believe that a GPL'd project is stabler. I believe, for 
example, that the fact that Wine switched over to the LGPL (which shares a 
lot of the cohesion argument) was a very important decision for the 
project, and that they would eventually have otherwise become irrelevant 
and the commercial users of the BSD-licensed code would have taken over.

But note that my second reason is not why I _began_ using the GPLv2, and 
that it's also equally true of the GPLv3 and LGPL. 

Anyway, there are other reasons I like the GPLv2. It's a "known entity" 
and it's been around for a long time.

In fact, even in -92, when I switched to the GPL, that "known factor" part 
was a major secondary reason for switching. I could have tried to just 
change my own license - but I felt it was an advantage to be something 
that people knew about, and not having to explain it and check it with 
lawyers. The fact that the GPLv2 was still "young" back then was nothing 
compared to how wet behind the ears my _own_ license was ;)

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Linus Torvalds wrote:
> On Thu, 2 Feb 2006, Pierre Ossman wrote:
>   
>> Linus Torvalds wrote:
>>     
>>> I _literally_ feel that we do not - as software developers - have the 
>>> moral right to enforce our rules on hardware manufacturers. We are not 
>>> crusaders, trying to force people to bow to our superior God. We are 
>>> trying to show others that co-operation and openness works better.
>>>       
>> Then I have to ask, why GPL and not a BSD license? GPL is after all,
>> forcing our beliefs onto anyone who wishes to benefit from our work.
>>     
>
> Yes, a lot of people see the GPL as a "crusading" license, and I think 
> that's partly because the FSF really has been acting like a crusader.
>
> But I think that one of the main reasons Linux has been successful is that 
> I don't think that the Linux community really is into crusading (some 
> small parts of it are, but it's not the main reason). I think Linux has 
> made the GPL more "socially acceptable", by being a hell of a lot less 
> religious about it than the FSF was.
>
> So to me, the point of the GPL is not the "convert the infidels" logic, 
> but something totally different:
>   

Thank you. I think that completely clarified your position in this.

Personally, I'm more of a "crusader". I don't think it's ok to force
someone into our way of thinking, but when the try to use our work I
feel they have at least a moral obligation to be as open themselves.
Kind of along the lines of the old "Do unto others..."-saying. That also
mean I don't really like bundling proprietary and open software (which I
do for a living so feel free to call me a hypocrite).

This whole DRM:d hardware issue is a bit different though since it seems
to be moving to a point where it cannot be avoided. "Vote with your
wallet" fails when there are so few of us that care about these things.
With they way electronics are packaged nowadays (chip packages that is),
it's getting increasingly difficult to build your own stuff. My fear is
that open source will be something you can only fiddle with on your ten
year old computer from the pre-DRM era.

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Mark v Wolher]

Pierre Ossman wrote:
> Linus Torvalds wrote:
> 
>>On Thu, 2 Feb 2006, Pierre Ossman wrote:
>>  
>>
>>>Linus Torvalds wrote:
>>>    
>>>
>>>>I _literally_ feel that we do not - as software developers - have the 
>>>>moral right to enforce our rules on hardware manufacturers. We are not 
>>>>crusaders, trying to force people to bow to our superior God. We are 
>>>>trying to show others that co-operation and openness works better.
>>>>      
>>>
>>>Then I have to ask, why GPL and not a BSD license? GPL is after all,
>>>forcing our beliefs onto anyone who wishes to benefit from our work.
>>>    
>>
>>Yes, a lot of people see the GPL as a "crusading" license, and I think 
>>that's partly because the FSF really has been acting like a crusader.
>>
>>But I think that one of the main reasons Linux has been successful is that 
>>I don't think that the Linux community really is into crusading (some 
>>small parts of it are, but it's not the main reason). I think Linux has 
>>made the GPL more "socially acceptable", by being a hell of a lot less 
>>religious about it than the FSF was.
>>
>>So to me, the point of the GPL is not the "convert the infidels" logic, 
>>but something totally different:
>>  
> 
> 
> Thank you. I think that completely clarified your position in this.
> 
> Personally, I'm more of a "crusader". I don't think it's ok to force
> someone into our way of thinking, but when the try to use our work I
> feel they have at least a moral obligation to be as open themselves.
> Kind of along the lines of the old "Do unto others..."-saying. That also
> mean I don't really like bundling proprietary and open software (which I
> do for a living so feel free to call me a hypocrite).
> 
> This whole DRM:d hardware issue is a bit different though since it seems
> to be moving to a point where it cannot be avoided. "Vote with your
> wallet" fails when there are so few of us that care about these things.
> With they way electronics are packaged nowadays (chip packages that is),
> it's getting increasingly difficult to build your own stuff. My fear is
> that open source will be something you can only fiddle with on your ten
> year old computer from the pre-DRM era.
> 
> Rgds
> Pierre
> 
> -

Well, i followed this thread on and off and i'd like to add a few words
of mine;

I think the harder one tries to enforce things on ppl the more
resistance is created. The 'free' countries who don't buy that DRM crap
will have a mighty opening in the market by creating machines which
allow you to do whatever you want (think of the region free players).
Just like the music industry is battling the ocean, they can't stop it,
cause ppl like to be free in their choices and creativeness. Yet the
internet music income is increasing which proves them wrong anyway.

I'd rather buy chinese hardware which i can do whatever i want to do
with and/or create software for than something of those who want to
control ppl in what they do and don't. Next thing is when you start your
system a signal is sent that "Drone 87-3KD983S" is now working on his
system.

Just remember, there are alway solutions, but not at the price of freedom.

Greetz,

Mark

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Pierre Ossman wrote:
> This whole DRM:d hardware issue is a bit different though since it seems
> to be moving to a point where it cannot be avoided. "Vote with your
> wallet" fails when there are so few of us that care about these things.
> With they way electronics are packaged nowadays (chip packages that is),
> it's getting increasingly difficult to build your own stuff. My fear is
> that open source will be something you can only fiddle with on your ten
> year old computer from the pre-DRM era.

Forgive me, but I don't subscribe to this cataclysmic scenario. The only
reason it's getting increasingly difficult to build your own stuff is
because the chip manufacturers are interested in making money, and to
do that they have to answer market demand, and market demand nowadays
is for gizmos/systems that have an ever-increasing number of features
and ever-increasing performance requirements. Cell phones are a prime
example. So, yes, it would be totally impossible to think today that
any single group of individuals could reproduce the early Apple era,
but that's just life -- things get complicated with time.

As a techie I love tinkering with things, and I'm always on the lookout
for mainstream electronics which I can hack to get Linux running on
them. At the end of the day, though, one has to come to terms with what
was pointed out earlier: we are dealing with proprietary hardware. And
as much as that may be disappointing, one has to realize that unlike
software, one cannot create and spread open-source hardware like one
can do with open source software in his basement just using free time --
because all the volunteers in the world cannot mint silicon and fab
PCBs and make them available to the rest of humanity.

We can convert all we want, but belief doesn't make money grow on trees.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Pierre Ossman]

Karim Yaghmour wrote:
> Pierre Ossman wrote:
>   
>> This whole DRM:d hardware issue is a bit different though since it seems
>> to be moving to a point where it cannot be avoided. "Vote with your
>> wallet" fails when there are so few of us that care about these things.
>> With they way electronics are packaged nowadays (chip packages that is),
>> it's getting increasingly difficult to build your own stuff. My fear is
>> that open source will be something you can only fiddle with on your ten
>> year old computer from the pre-DRM era.
>>     
>
> Forgive me, but I don't subscribe to this cataclysmic scenario. The only
> reason it's getting increasingly difficult to build your own stuff is
> because the chip manufacturers are interested in making money, and to
> do that they have to answer market demand, and market demand nowadays
> is for gizmos/systems that have an ever-increasing number of features
> and ever-increasing performance requirements. Cell phones are a prime
> example. So, yes, it would be totally impossible to think today that
> any single group of individuals could reproduce the early Apple era,
> but that's just life -- things get complicated with time.
>
>   

That's understandable. I cannot require them to go out of their way to
accommodate my desire for tinkering. When they decide to actually put
time and effort into preventing me from tinkering then I get a bit
annoyed. I still cannot really do much about it though. If people keep
buying the crap then it will continue to be produced. What I can do is
say that such a system is not allowed to be based on my work. It may not
matter in the long run, but at least I'm doing something. Things might
get better by itself, or worse for that matter, but I'd prefer to not
stand idly by, hoping it will go the way I'd like.

I try to not sound like a doomsday prophet, but I tend to get a bit
worked up when I consider what it would be like with this DRM nonsense
taken to the extreme. :)

Rgds
Pierre

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Pierre Ossman wrote:
> That's understandable. I cannot require them to go out of their way to
> accommodate my desire for tinkering. When they decide to actually put
> time and effort into preventing me from tinkering then I get a bit
> annoyed. I still cannot really do much about it though. If people keep
> buying the crap then it will continue to be produced. What I can do is
> say that such a system is not allowed to be based on my work. It may not
> matter in the long run, but at least I'm doing something.

FWIW, let me just say that I fully understand and respect your point of
view. I don't take any enjoyment out of seing some of the stuff I've
done used in such a way either. I just have a different take on how that
best be solved.

> Things might
> get better by itself, or worse for that matter, but I'd prefer to not
> stand idly by, hoping it will go the way I'd like.
> 
> I try to not sound like a doomsday prophet, but I tend to get a bit
> worked up when I consider what it would be like with this DRM nonsense
> taken to the extreme. :)

Going a little bit off topic here allow me to say that I personally
find most of the actual content out there being sold is utter junk.

I don't remember which one it was out of Siskel or Ebbert (and god
rest his soul I don't even remember which one passed away), but one
of them appeared on this documentary once and said something like:
"If you're 30 and you've got a degree, there's absolutely nothing
for you to see at the movie theater." And my impression is that this
applies readily accross the board for all entertainment. I mean, for
pete's sake, these supposed content providers are offering up new
"artists" every other week and they keep pumping out these totally
brain-dead TV shows ... Heck, I can't remember the last time I opened
the TV or listened to the radio and thought: "this is actually worth
my time." So, as far as I'm concerned, the brand names can encrypt
the content all they want, it's probably more interesting to watch
that way anyway.

FWIW, I suspect that part of the reason a lot of these folks are
seeing a decline in sales is mainly because their content isn't
something the clients think is really all that worthy.

Think about it, when is the last time you heard of an artist or
saw one perform on screen or elsewhere and thought to yourself:
"wow, this is really a talented person/group."

And to be quite frank, I have no pitty for artists who boo-hoo in
public about those darn "pirates" taking their income away. The
human species has been pumping out performers, singers, composers,
... artists for thousands of years before anyone was able to
record them. The only difference here is that many modern "artists"
think they have a god-given right that society continue to
sustain their relatively-recent recording-media-funded revenues.

Digital media only means bubble-gum "artists" fade away. True
ones are likely continue being what they are, regardless of the
fate of recorded-media as an income.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Helge Hafting]

Pierre Ossman wrote:

>This whole DRM:d hardware issue is a bit different though since it seems
>to be moving to a point where it cannot be avoided. "Vote with your
>wallet" fails when there are so few of us that care about these things.
>  
>
I do not share this fear.  Well, off-the-shelf complete pc's might end 
up like
that, demanding a signed OS and so on if we take a very pessimistic
view on the future. 

But "vote with your wallet" still works.  There is a huge parts market .
You can buy hundreds of brands of motherboards, processors and
add-on cards.  Strict DRM will never fly in this market!  What if
_one_ manufacturer comes out with a new improved motherboard that only
will boot and run a signed version of windows?  No sales! 

The guy who build is own computer won't take that sort of thing, he'll
find another manufacturer.  Remember the outrage when Intel implemented
cpuID?  They were forced to make that optional.

And if one country encourages DRM hw, then other countries will
ramp up non-DRM hw production as the demand rises.

I think the worst that could possibly happen is that cheap complete pc's
found at supermarkets might have some DRM hw in them.  Stuff that
we might need a few 30-hour debugging sessions to circumvent. And
so we will be forced to assemble our pc's from parts - the way we
usually do anyway for other reasons.

Helge Hafting

Re: GPL V3 and Linux - Dead Copyright Holders

[Ian Kester-Haney]

On 2/2/06, Linus Torvalds <torvalds@osdl.org> wrote:
All this boils down to is a linking argument.  Nvidia releases closed
sorce drivers that link to the kernel as a module.  Many kernel users
have nvidia cards or even ati cards.  Under GPLv3 at first glance
these will have to be implemented without the kernel linking that
gives them their good performance.  Same for DVDs and other Media with
DRM or other copyright measures.  Free Software cannot include the
tools to view this content as GPL in any circumstance and any argument
to the contrary is baseless.

Copyright law has many issues and the GPL cannot address this by
limiting the functionality of GPL projects.  IMHO the fact that I buy
a HDVD or Blu-Ray disk with protected content presumes an agreement to
the specifications and I must abide by the restrictions set by the
manufacturer whether it be a big company or a small firm releasing the
content in that format.  This cannot be onforced in a free software
environment if the ability to break the restrictions is built-in, the
GPL cannot be compatible, that is why closed operating systems like
Windows and MacOS and the BSDs continue to dominate the marketplace. 
Linux may well erode the server market that serves up and produces the
content that ends up being protected, but the end user will not in the
forseeable future be able to use the Open Source Mavement to view such
materials as long as the "percieved" ability to override the
protection is present.  Just an average joes perspective.  I 
personally don't want linux to adopt a new liscence regardless of the
reasons.  New does not always mean better and just because the GPL
leaves the choice of version up to the inventor does not make the
whole concept fluff.  You can't really argue in common sense that the
Section 9 language gives future versions of the GPL the ability to
take over the definitively stated version provided by the grantee.

Thankyou.


> On Thu, 2 Feb 2006, Pierre Ossman wrote:
> >
> > So taking open software and closed hardware and combining it into something
> > that I cannot modify is ok by you?
>
> But you CAN modify the software part of it. You can run it on other
> hardware.
>
> It boils down to this: we wrote the software. That's the only part _I_
> care about, and perhaps (at least to me) more importantly, because it's
> the only part we created, it's the only part that I feel we have a moral
> right to control.
>
> I _literally_ feel that we do not - as software developers - have the
> moral right to enforce our rules on hardware manufacturers. We are not
> crusaders, trying to force people to bow to our superior God. We are
> trying to show others that co-operation and openness works better.
>
> That's my standpoint, at least. Always has been. It's the reason I
> chose the GPL in the first place (and it's the exact same reason that I
> wrote the original Linux copyright license). I do _software_, and I
> license _software_.
>
> And I realize that others don't always agree with me. That's fine. You
> don't have to. But I licensed my project under a license _I_ agreed with,
> which is the GPLv2. Others who feel differently can license under their
> own licenses. Including, very much, the GPLv3.
>
> I'm not arguing against the GPLv3.
>
> I'm arguing that the GPLv3 is wrong for _me_, and it's not the license I
> ever chose.
>
>                         Linus
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: GPL V3 and Linux - Dead Copyright Holders

[Lee Revell]

On Thu, 2006-02-02 at 12:13 -0600, Ian Kester-Haney wrote:
> On 2/2/06, Linus Torvalds <torvalds@osdl.org> wrote:
> All this boils down to is a linking argument.  Nvidia releases closed
> sorce drivers that link to the kernel as a module.  Many kernel users
> have nvidia cards or even ati cards.  Under GPLv3 at first glance
> these will have to be implemented without the kernel linking that
> gives them their good performance.  Same for DVDs and other Media with
> DRM or other copyright measures.  Free Software cannot include the
> tools to view this content as GPL in any circumstance and any argument
> to the contrary is baseless.

What nvidia is doing is already illegal under the GPLv2.

Lee

Re: GPL V3 and Linux - Dead Copyright Holders

[Christopher Friesen]

Lee Revell wrote:

> What nvidia is doing is already illegal under the GPLv2.

I don't think that's been legally proven.

The question is whether it is a derivative work.  If their driver core 
(aka the binary blob) is common to all their drivers (across multiple 
OS's), it could be argued that the binary blob itself is not a 
derivative work.  One could almost view it as firmware.

If they ship the binary blob as well as code that interfaces the binary 
blob with the kernel, and the end-user compiles the code together and 
loads it into the kernel, does that necessarily violate the GPL?

Chris

Re: GPL V3 and Linux - Dead Copyright Holders

[Lee Revell]

On Thu, 2006-02-02 at 13:11 -0600, Christopher Friesen wrote:
> Lee Revell wrote:
> 
> > What nvidia is doing is already illegal under the GPLv2.
> 
> I don't think that's been legally proven.
> 

OK, s/illegal/questionable/ - I don't want to open that can of worms,
this thread is already too long...

Re: GPL V3 and Linux - Dead Copyright Holders

[Luke-Jr]

On Thursday 02 February 2006 19:11, Christopher Friesen wrote:
> Lee Revell wrote:
> > What nvidia is doing is already illegal under the GPLv2.
>
> I don't think that's been legally proven.
>
> The question is whether it is a derivative work. 

The LGPL deals with only derivative works. The GPL also deals with mere 
*linking*. If glibc were GPL'd, it would be illegal to make an OS based on it 
with a single C program incompatible with the GPL.

> If they ship the binary blob as well as code that interfaces the binary
> blob with the kernel, and the end-user compiles the code together and
> loads it into the kernel, does that necessarily violate the GPL?

The 'code that interfaces the binary blob with the kernel' would then be 
illegal, because the code cannot be both GPL and proprietary. If the code is 
GPL and acceptable for kernel-linking, then under the terms of the GPL, the 
code cannot link to a GPL-incompatible binary blob.

Re: GPL V3 and Linux - Dead Copyright Holders

[Zan Lynx]

Luke-Jr wrote:
> On Thursday 02 February 2006 19:11, Christopher Friesen wrote:
>   
>>  ship the binary blob as well as code that interfaces the binary
>> blob with the kernel, and the end-user compiles the code together and
>> loads it into the kernel, does that necessarily violate the GPL?
>>     
>
> The 'code that interfaces the binary blob with the kernel' would then be 
> illegal, because the code cannot be both GPL and proprietary. If the code is 
> GPL and acceptable for kernel-linking, then under the terms of the GPL, the 
> code cannot link to a GPL-incompatible binary blob.
>   
Not at all.  The linking interface code does not have to enforce GPL
restrictions until it is linked to GPL code.  It could be BSD, for
example, as much kernel code is.  BSD can link to binary blob without a
problem.  BSD can link to GPL without a problem.  The final result is a
undistributable bastard, but the end user will not be distributing.

Taken as separate pieces, the code cannot be illegal.  It is only by
arguing the intent to link to GPL code, and the distribution of the glue
and proprietary together that there would even be the possibility of
proving a license violation.  If that looked to be a possibility, all
nVidia, or anyone else would have to do is publish only their binary
blob along with instructions to third parties on how to create interface
glue.  Such a kernel module, written by a third party, to published
documentation, would clearly be an interface, not a link, and not
something the GPL could apply to.

So, you could _maybe_ stop people from directly providing proprietary
binaries plus glue modules in the same package, but you could not
prevent them from achieving the same result with a bit of extra effort.

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> The LGPL deals with only derivative works. The GPL also deals with mere
> *linking*. If glibc were GPL'd, it would be illegal to make an OS
> based on it
> with a single C program incompatible with the GPL.

	The GPL also only deals with derivative works. If linking does not create a
derivative work, then the GPL cannot affect it. The GPL cannot define its
own scope, only copyright law does that. GPL section zero states this, but
even if it didn't it would still be true.

	The GPL could say that it affected every work every created by any human
being. It could say that it affected a work created by any person who ever
used a GPL'd work. But these things would have no force because copyright
law provides no such mechanism.

	The only way the GPL can control work Y because it affects work Z is
because Y is a derivative work of work Z. If it's not, then the works are
legally unrelated, and no matter what the GPL says, it can't affect work Y.

	If work Z is a "mere aggregate" containin all or part of work Y, then the
GPL would still apply to work Y, but not to any part of work Z not from work
Y.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Kyle Moffett]

On Feb 06, 2006, at 16:07, David Schwartz wrote:
>> The LGPL deals with only derivative works. The GPL also deals with  
>> mere *linking*. If glibc were GPL'd, it would be illegal to make  
>> an OS based on it with a single C program incompatible with the GPL.
>
> The only way the GPL can control work Y because it affects work Z  
> is because Y is a derivative work of work Z. If it's not, then the  
> works are legally unrelated, and no matter what the GPL says, it  
> can't affect work Y.

To say this more simplistically, the LGPL essentially says "Even if  
dynamic linking constitutes making a derivative work, we allow you to  
dynamically link, so long as the rules are followed for the LGPL code  
to which you link".  The GPL essentially says "If dynamic linking is  
making a derivative work, then these rules apply to the whole  
derivative work and all of its constituent parts".

Whether or not an NVidia binary module is a derivative work is left  
up to the courts to decide.  It _may_ be legal (don't trust me,  
consult your lawyer) to have a very simple cross-platform interface  
and some BSD-licensed glue.  On the other hand, if your interface  
derives from or exposes any kind of kernel-internals, then it is most  
certainly a derivative work (because you can't argue that the binary  
interface was written to be independent of Linux, and it therefore  
falls under the GPL.

Cheers,
Kyle Moffett

--
Unix was not designed to stop people from doing stupid things,  
because that would also stop them from doing clever things.
   -- Doug Gwyn

Re: GPL V3 and Linux - Dead Copyright Holders

[Luke-Jr]

On Monday 06 February 2006 22:38, Kyle Moffett wrote:
> On Feb 06, 2006, at 16:07, David Schwartz wrote:
> >> The LGPL deals with only derivative works. The GPL also deals with
> >> mere *linking*. If glibc were GPL'd, it would be illegal to make
> >> an OS based on it with a single C program incompatible with the GPL.
> >
> > The only way the GPL can control work Y because it affects work Z
> > is because Y is a derivative work of work Z. If it's not, then the
> > works are legally unrelated, and no matter what the GPL says, it
> > can't affect work Y.
>
> To say this more simplistically, the LGPL essentially says "Even if
> dynamic linking constitutes making a derivative work, we allow you to
> dynamically link, so long as the rules are followed for the LGPL code
> to which you link".  The GPL essentially says "If dynamic linking is
> making a derivative work, then these rules apply to the whole
> derivative work and all of its constituent parts".

Does the "if" clause exist? If not, then by distributing the original 
software, the distributor is agreeing to abide by the GPL terms for anything 
linking with it. If the linking code is not GPL-compatible, the distributor 
cannot legally distribute the GPL'd component.

> Whether or not an NVidia binary module is a derivative work is left
> up to the courts to decide.  It _may_ be legal (don't trust me,
> consult your lawyer) to have a very simple cross-platform interface
> and some BSD-licensed glue. 

They might have a legal loophole if dynamic linking doesn't constitute a 
derivative work *and* nVidia doesn't distribute Linux (thus does not need to 
agree to GPL terms).

Re: GPL V3 and Linux - Dead Copyright Holders

[Luke-Jr]

On Monday 06 February 2006 21:07, David Schwartz wrote:
> > The LGPL deals with only derivative works. The GPL also deals with mere
> > *linking*. If glibc were GPL'd, it would be illegal to make an OS
> > based on it
> > with a single C program incompatible with the GPL.
>
> 	The GPL also only deals with derivative works. If linking does not create
> a derivative work, then the GPL cannot affect it. The GPL cannot define its
> own scope, only copyright law does that. GPL section zero states this, but
> even if it didn't it would still be true.

GPL can establish terms for distribution of the original software, even if it 
goes unmodified. In other words, by distributing the GPL software, one must 
agree that all software being distributed at the same time linking to it are 
GPL-compatible. The GPL cannot prohibit you from distributing the linking 
software in this case, but it can prohibit distribution of the original 
software.

> 	The GPL could say that it affected every work every created by any human
> being. It could say that it affected a work created by any person who ever
> used a GPL'd work. But these things would have no force because copyright
> law provides no such mechanism.

They would have force as conditions of distributing GPL software. If someone 
distributed software under the GPL, they would be accepting its terms and 
thus accepting that the GPL applies to all work they created. This example 
could be a bit extreme, however, and prohibited just on the grounds of being 
an invalidly-broad contract.

> 	The only way the GPL can control work Y because it affects work Z is
> because Y is a derivative work of work Z. If it's not, then the works are
> legally unrelated, and no matter what the GPL says, it can't affect work Y.

But it can prohibit distribution of work Z when work Y does not comply with 
the terms.

> 	If work Z is a "mere aggregate" containin all or part of work Y, then the
> GPL would still apply to work Y, but not to any part of work Z not from
> work Y.

Linking is not mere aggregation.

Re: GPL V3 and Linux - Dead Copyright Holders

[Ingo Molnar]

* Linus Torvalds <torvalds@osdl.org> wrote:

> It boils down to this: we wrote the software. That's the only part _I_ 
> care about, and perhaps (at least to me) more importantly, because 
> it's the only part we created, it's the only part that I feel we have 
> a moral right to control.

yes, that's how i feel too. Quid pro Quo. But we are in the minority.  
The majority of Linux users, and the vast majority of commercial Linux 
players doesnt give a rat's a** about giving back a Quid, as a fair 
compensation for our Quo.

(i dont see a problem with that by the way - i consider it a basic moral 
right to have different moral rules. The moment i'd require others to 
share exactly the same morals i'd become just another crusader. But i 
digress.)

what the others see is the black and white letter of the GPL [or just 
some free software they can download and install], not a moral situation 
with us, which they would have to understand and meet. They are and will 
increasingly try to give back as little as they can get away with - 
depending on their own moral rules. Some will also try to employ tactics 
that we see as "immoral", and they wont see it as immoral, under their 
own rules.

(one extreme example for a community with very strong moral rules is the 
Sicilian Mafia. They had (and still have) extremely strong inner rules, 
and everyone within a given family is loved and is being taken care of 
for a lifetime - in the positive sense of the word. Still the external 
effect on the rest of society is perverse and disastrous. But i digress 
again.)

so the only solution is to convert our moral rules to an objective set 
of rules, as accurately as possible. This in our society would be the 
(neutral) letter of the law, and a copyright license in particular.  

after that act of conversion we can only hope for the best that no 
detail important to us gets lost in the process. Our (your) initial 
approach to that was the GPLv2, and it was a pretty accurate (and lucky) 
first approximation.

what do we have today? We've got bin-only kernel modules, much of which 
are clearly immoral, they are clearly hurting us and still we do things 
to keep them going - e.g. the refusal to remove 8K stacks from the 
.config. We are increasingly getting into a situation where loopholes 
are found and utilized to give back as little as possible, upsetting the 
balance.

so i believe _something_ should be done to tip the balance, because the 
negative effects are already hurting us. I'd support the move to the 
GPLv3 only as a tool to move the balance back into a fairer situation, 
not as some new moral mechanism. The GPLv3 might be overboard for that, 
but still the situation does exist undeniably.

> I _literally_ feel that we do not - as software developers - have the 
> moral right to enforce our rules on hardware manufacturers. [...]

yes, and i do share your morals on that. OTOH i do think there are a 
few more aspects:

- most of the known violations of the Quid pro Quo comes from the space 
  of closed hardware that is pushing for DRM best. So there's some 
  itching in me to just make things more strict in the area that is 
  causing us the most problems.

- can they give us the source code for the modifications on Linux on a 
  DRM-ed medium that we cannot read on any open hardware? From a moral 
  POV they cannot, but can they do it legally? Could they argue in 5 
  years that SHD-DVD version 10 is a 'widely used' medium, and that they 
  met the letter of the GPLv2? If we never enforce that the source be 
  actually compilable and usable on real hardware, how can we suddenly 
  claim to have a right to run it on open hardware? We might create a
  legal waiver or estoppel situation if we dont enforce the usefulness 
  of the source code given back to us. Even cockroaches are surprisingly
  creative - after all there is a business entity on this planet that 
  thought it to be fair to produce source code in discovery by printing 
  it out to a ton of paper and then scanning it back in ;-) And they 
  even got away with it!

- we really grew up on the supposition that there is a fundamental 
  ability and right to tinker. Business entitities were simply not able 
  back then to restrict that, technologically. Today there are business 
  models that seem to be working just fine with closed hardware. Content 
  and programmability restrictions via crytography are more and more
  practical, and the day will come when the Xbox will be truly 
  cryptographically safe and totally closed.

  NOTE: i do know that the elimination of tinkering is bad for society 
  down the road - so in theory we should win in the long run. But 
  society might not care! Maybe only 1 civilization out of a 100 get 
  past this stage of development - the rest destroy themselves and 
  create a burned out shell of a planet with some proto-civilization
  and no resources left. Nature is really, really cruel.

  Do we have the moral right to restrict (in the worst case, eliminate) 
  our children's ability to tinker _at all_? Doesnt our software become 
  totally useless if the possibility to tinker gets eliminated? Only 
  1-5% of all people have the brain structure and desire to tinker (and 
  to think creatively), so if it were up for a vote today we'd lose in 
  the polls, badly ...

  Dont you think we have the moral obligation to support all the 
  "infrastructure" that gave us this ideal paradise of tinkering with 
  Linux in the first place? Dont you sense we (programmers, tinkerers) 
  are a minority that could _easily_ be opressed by society at large, 
  without them even noticing? Supporting fact: society is heading 
  towards a nice big greenhouse effect right now, with 99% of the 
  scientists crying bloody murder already - and basically nothing is
  done. Crushing these 'geek dudes' which would indeed result in lost 
  productivity a few decades down the line, but right now it wouldnt 
  even be a blip on the policy radar i'm afraid, as long as it results 
  in blockbuster movies getting on to the TV screen faster, and as long 
  as it results in a 5% cheaper HD-DVD player ...

  Thinking about those issues and current trends, i'm really getting an 
  urge to grab some bigger protective gear, like the GPLv3!

	Ingo

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Glauber de Oliveira Costa wrote:
> I may be missing the point here, (In case you're more than welcome to
> correct me), but ... Why? Can't a software license restrict the usage
> of the software? In which ways do you think the sentence "Don't use in
> DRM'ed hardware" differs from sentences like  "Not allowed in country
> X",  "Don't use for commercial purposes",  and other alikes ? I think
> that saying in which hardware your software can or cannot run is a
> pretty valid license term (without messing with the question about it
> being the right thing to do here).

You're right, I didn't think it through properly ... I guess what I'm
trying to say is that the more restrictions a licensed work imposes
on its runtime environment, the less free it is. In my view of freedom,
a "free" license should indeed protect the software licensed under it,
but it should be as unintrussive with regards to runtime as possible
-- simply because the original author may not have anticipated all the
possible runtime scenarios. As a user, I should be free to decide what
hardware I'd like to see this software run on. In that regard, I think
GPLv2 strikes the right balance, while GPLv3 attempts to solve one
issue by introducing a lot of other problems.

... and bearing in mind that there are legitimate non-DRM embedded and
security applications where runtime software restrictions are required/
inherent ... think:
- software controlling consumer appliances such as cars, etc.
- masked-ROM software (non-flashable)
- access-control hardware
- high-end IT security
- etc.

FWIW, private discussions with RMS a few years back showed a clear
misunderstanding of how important embedded devices are. It really
seems to me that the FSF's newly found urgency for solving an
increasingly disturbing problem (DRM) is unfortunately not based on
internal familiarity with the embedded world. Contrary to the FSF's
long-standing experience of having its software used in workstations
and servers, it's only very recently that its software, and software
licensed under licenses it publishes, has been found in embedded
devices -- Though GCC & co have been used to cross-build for embedded
devices for a very long time, this is different from having the
actual software run on the gizmo.

This argument, in itself, doesn't diminish the value of the FSF's
position. They are indeed intent on defending software freedom and
in that I cannot condem them. However, I really think that those
championing this new wording should think through all the
possibilities. As a user, I clearly hate DRM and would indeed like
to see it disappear. As a developer, and an active participant in
the development of many kinds of embedded/customized systems, I
also see that the new wording imposes unrealistic limitations to
legitimate designs. IOW: right cause, wrong venue.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Pavel Machek]

Hi!

> ... and bearing in mind that there are legitimate non-DRM embedded and
> security applications where runtime software restrictions are required/
> inherent ... think:
> - software controlling consumer appliances such as cars, etc.

What is wrong with changing software in my car? I'm allowed to cut my
own brakes, why should I be disallowed in software in my car.

> - masked-ROM software (non-flashable)

That's fair; I don't think GPLv3 forbids ROMs, through.

							Pavel
-- 
Thanks, Sharp!

Re: GPL V3 and Linux - Dead Copyright Holders

[Lee Revell]

On Thu, 2006-02-02 at 23:02 +0100, Pavel Machek wrote:
> What is wrong with changing software in my car? I'm allowed to cut my
> own brakes, why should I be disallowed in software in my car. 

Not if you want to drive on a public street...

Lee

Re: GPL V3 and Linux - Dead Copyright Holders

[Karim Yaghmour]

Pavel Machek wrote:
> What is wrong with changing software in my car? I'm allowed to cut my
> own brakes, why should I be disallowed in software in my car.

Lee aptly answered this one.

> That's fair; I don't think GPLv3 forbids ROMs, through.

If that's all it takes to not fall under the DRM requirements of the
GPLv3 then certainly no consumer electronics manufacturer that I know
of will hesitate to ship Linux (or any other piece of software I
should add) on masked ROMs should it ever convert to GPLv3. It's not
like they haven't been doing that for the past 30+ years. Any good
that will have done to any of those championing GPLv3 to help stop
DRM ...

Seriously, though, my earlier suggestion is even simpler. Just let
both signed and unsigned kernels run, just don't enable key user-
space-application-controlled hardware components if not signed and
do so in such a way requiring a power-up/hard-reset for another
pass at enable/disable. Perfectly permissible in the draft.

And I dare say that I hardly see how the FSF could try to cover this
one in any future draft since doing so would require having to place
restrictions on what independently-developed software/hardware
combinations are allowed in any system containing the covered
"work".

Like I said before: right cause, wrong venue.

Karim
-- 
President  / Opersys Inc.
Embedded Linux Training and Expertise
www.opersys.com  /  1.866.677.4546

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Fri, 27 Jan 2006, Alan Cox wrote:

> On Mer, 2006-01-25 at 17:39 -0500, Linus Torvalds wrote:
> > In other words: the _default_ license strategy is always just the 
> > particular version of the GPL that accompanies a project. If you want to 
> > license a program under _any_ later version of the GPL, you have to state 
> > so explicitly. Linux never did.
> 
> Not correct. See section 9.

Sorry, I think you're wrong.

We've _always_ said which license explicitly. It's in the COPYING file.

Even before the clarification, the COPYING file has always said 


                    GNU GENERAL PUBLIC LICENSE
                       Version 2, June 1991

 Copyright (C) 1989, 1991 Free Software Foundation, Inc.
                       51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 Everyone is permitted to copy and distribute verbatim copies
	...

at the very top.

How can you say that we didn't specify a version?

If you distribute a program, and you just say "I license this under the 
GPL", THEN you don't specify a verion.

But if you distribute a program, and the ONLY license that is associated 
with it is a specific version of a license file, then that's what you 
have, UNLESS SOMETHING SAYS OTHERWISE.

This is basic copyright law, btw, and has nothing to do with the GPL per 
se. If you don't have a license, you don't have any copyright AT ALL.

Linux kernel files don't say "This is licensed under the GPL". Not mine, 
at least. I don't see the point, and I never have. There's a COPYING file 
that specifies what the license is, and that COPYING file very much 
specifies a very _specific_ version of the GPL. Always has.

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Fri, 27 Jan 2006, Linus Torvalds wrote:
> 
> This is basic copyright law, btw, and has nothing to do with the GPL per 
> se. If you don't have a license, you don't have any copyright AT ALL.

This is really important, btw.

Yes, when we speak colloquially we talk about the fact that Linux is 
licensed "under the GPL", but that is _not_ how anybody actually has ever 
gotten a license legally. The ONLY way anybody has ever legally licensed 
Linux is either with the original very strict copyright _or_ thanks to the 
COPYING file. Nothing else really matters. 

So the version of the GPL has always been explicit. At no point has the 
kernel been distributed without a specific version being clearly mentioned 
in the ONLY PLACE that gave you rights to copy the kernel in the first 
place. So either you knew it was GPLv2, or you didn't have the right to 
copy it in the first place.

In other words, Linux has _never_ been licensed under anything but the GPL 
v2, and nobody has _ever_ gotten a legal Linux source distribution with 
anything but a complete copy of GPLv2 license file.

So when I say that the additions at the top of the COPYING file are 
nothing but clarifications, I'm not just making that up. Anybody who 
claims that any Linux kernel I've ever made has ever been licensed under 
anything else than those exact two licenses is just not correct.

And Alan, I know we've had this discussion before. You've claimed before 
that my clarifications are somehow "changing" the license, and I've told 
you before that no, they don't change the license, they just clarify 
things that people keep on gettign wrong, or keep on being nervous about.

So people can argue all they want about this. But unless you get a real 
legal opinion (not just any random shyster - a real judge making a 
statement, or a respected professional who states his firm legal opinion 
in no uncertain terms), I don't think you have a legal leg to stand on.

But no, IANAL. I'd be willing to bet real money that a real lawyer would 
back me up on this, though.

			Linus

---

PS. Just out of historical interest, the only other copyright license ever 
distributed with the kernel was this one:

 "This kernel is (C) 1991 Linus Torvalds, but all or part of it may be
  redistributed provided you do the following:

	- Full source must be available (and free), if not with the
	  distribution then at least on asking for it.

	- Copyright notices must be intact. (In fact, if you distribute
	  only parts of it you may have to add copyrights, as there aren't
	  (C)'s in all files.) Small partial excerpts may be copied
	  without bothering with copyrights.

	- You may not distibute this for a fee, not even "handling"
	  costs.

  Mail me at "torvalds@kruuna.helsinki.fi" if you have any questions."

and that one was only valid between kernel versions 0.01 and 0.12 or 
something like that.

Re: GPL V3 and Linux - Dead Copyright Holders

[Al Viro]

On Sat, Jan 28, 2006 at 12:22:58AM -0500, Linus Torvalds wrote:
> PS. Just out of historical interest, the only other copyright license ever 
> distributed with the kernel was this one:
> 
>  "This kernel is (C) 1991 Linus Torvalds, but all or part of it may be
>   redistributed provided you do the following:
> 
> 	- Full source must be available (and free), if not with the
> 	  distribution then at least on asking for it.
> 
> 	- Copyright notices must be intact. (In fact, if you distribute
> 	  only parts of it you may have to add copyrights, as there aren't
> 	  (C)'s in all files.) Small partial excerpts may be copied
> 	  without bothering with copyrights.
> 
> 	- You may not distibute this for a fee, not even "handling"
> 	  costs.
> 
>   Mail me at "torvalds@kruuna.helsinki.fi" if you have any questions."
> 
> and that one was only valid between kernel versions 0.01 and 0.12 or 
> something like that.

Interesting...  What does that do to e.g. DVD with full (OK, modulo missing
early versions) kernel history all way back to 0.01?

Even funnier question is what does that do to full CVS including the
early versions.  Can that be distributed at all and what license would
fit it?  Arguing that it's mere aggregation is possible, but it's a
bit of a stretch...

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Sat, 28 Jan 2006, Al Viro wrote:
> > 
> > 	- You may not distibute this for a fee, not even "handling"
> > 	  costs.
> > 
> >   Mail me at "torvalds@kruuna.helsinki.fi" if you have any questions."
> > 
> > and that one was only valid between kernel versions 0.01 and 0.12 or 
> > something like that.
> 
> Interesting...  What does that do to e.g. DVD with full (OK, modulo missing
> early versions) kernel history all way back to 0.01?

Well, the good news is that I was the only real copyright holder back then 
(there's a couple of other people who contributed to 0.11 and/or 0.12, 
mainly Ted T'so with the BSD terminal control stuff - ^Z and friends).

I used to even re-write patches to suit my style (this was back then, the 
patches were smaller, and I was younger and had more energy). So some 
things that others sent in patches for (I think Peter McDonald did pty's) 
I ended up re-writing myself (and in the process I mixed up the master and 
slave pty major number, iirc ;)

> Even funnier question is what does that do to full CVS including the
> early versions.  Can that be distributed at all and what license would
> fit it?  Arguing that it's mere aggregation is possible, but it's a
> bit of a stretch...

I think you can take it for granted that the GPL re-licensing was 
retro-active. I'm the sole copyright holder for 99% of it, and there were 
no objections to the relicensing even back when it happened, so I can 
pretty much guarantee that there would be none now ;)

It was a kind of strange license. I didn't spend a whole lot of time 
thinking about it ;)

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Gwe, 2006-01-27 at 21:06 -0500, Linus Torvalds wrote:
> > Not correct. See section 9.
> 
> Sorry, I think you're wrong.
> 
> We've _always_ said which license explicitly. It's in the COPYING file.
> 
> Even before the clarification, the COPYING file has always said 
> 
> 
>                     GNU GENERAL PUBLIC LICENSE
>                        Version 2, June 1991


--
If the Program specifies a version number of this License which applies
to it and "any later version", you have the option of following the
terms and conditions either of that version or of any later version
published by the Free Software Foundation.  If the Program does not
specify a version number of this License, you may choose any version
ever published by the Free Software Foundation.
--

"COPYING" is the license not the program. If it were the program you
would be violating the GPL as the GPL license text is not itself GPL
compatible. It is also not a statement of the version of the software
but a statement of the version of the document itself. Both seem quite
self evident and clear. Since the document itself provides choices for
versions to be applied it cannot be said to imply a version either.

Nowdays you specifically state version 2 then reproduce the license
document. That likewise seems self evident and clear. Section 9 probably
gives you the right to do that for any code that did not specify a
version rule already.

You might also want to ask "if the FSF COPYING text specified the
program version as you claim then how would you specify versions
differently". And likewise "How come the FSF itself, author of the
license, distributes its default COPYING file with code clearly intended
and marked to be GPL v2 or later".

In short your interpretation of the past state of affairs would not
stand up to scrutiny.

> If you distribute a program, and you just say "I license this under the 
> GPL", THEN you don't specify a verion.

And merely adding the copying file likewise is still not specifying a
version. You may be *implying* one but that is not specifying.

> Linux kernel files don't say "This is licensed under the GPL". Not mine, 
> at least. I don't see the point, and I never have. There's a COPYING file 

The point is to avoid ambiguity. Consider the statement in clause 2

--
In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
--

The COPYING file is mere aggregation - it is a seperately licensed and
independant work to the program with incompatible conditions. In the
kernel case this does not matter as you included text with COPYING to
make the intent clear, and there is no doubt, nor alternate licenses.

Consider however taking BSD 2 clause licensed code and relicensing it
GPL with changes. If the GPL was merely placed with the code it would
not be clear that the GPL now applied to it, especially if there are
other independant GPL programs in the same archive. The advice text with
the GPL itself thus provides for the 'fail safe' worst case scenario
rather than neccessarily addressing all cases in the minimal and neatest
fashion. Lawyers dislike amibguity because it causes expensive problems
later on.

(Also historically an assertion of copyright was neccessary avoid being
public domain in the USA and some other countries.)

Alan

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Mon, 30 Jan 2006, Alan Cox wrote:
>
> You might also want to ask "if the FSF COPYING text specified the
> program version as you claim then how would you specify versions
> differently". And likewise "How come the FSF itself, author of the
> license, distributes its default COPYING file with code clearly intended
> and marked to be GPL v2 or later".

Alan, you're weasel-wording, and making up arguments that aren't valid.

You can avoid specifying a particular version of the GPL by just saying 
"This project is distributed under the GPL" and pointing to the FSF. Not 
everybody necessarily even includes the LICENSE file.

Please realize that the kernel is one of the largest open-source projects 
in the world, if not _the_ largest. Most projects are much much smaller. 
Many projects are very ad-hoc, and not all of them have necessarily been 
all that careful with adding LICENSE files etc.

As to your "FSF itself" argument, the code the FSF distributes obviously 
_has_ to state "GPL v2 or later" exactly because they want the "or later" 
thing to be effective.

Their suggestions very much is to include the GPL license in its entirety 
_and_ to say "GPL v2 or later" in the files, exactly because they need to 
_expand_ the license from just the one in the license file (ie the "or 
later" part is essentially an open-ended dual license).

None of your arguments in any way argue that Linux wouldn't be GPLv2.

> In short your interpretation of the past state of affairs would not
> stand up to scrutiny.

You can claim anything you like. I think you're wrong. But in the 
meantime, that doesn't matter. If it ever goes to court, you'll see what a 
real judge will claim.

My bet is that my interpretation is the only sane one. 

> The COPYING file is mere aggregation - it is a seperately licensed and
> independant work to the program with incompatible conditions.

The fact that the COPYING file has a different copyright really doesn't 
matter. It's still part of the release. 

It's absolutely not different from having a separate "Release notes" file 
which specifies the copyright conditions. That's how Linux-0.01 did it: 
the thing was outside the actual main tar-ball, and sent out both as part 
of the announcemnt and as a separate file in the same directory on the 
ftp-site.

Yes, it may be "mere aggregation", but that has absolutely _zero_ impact 
on my argument. It's the only license you have to copy Linux, and it very 
much has an EXPLICIT VERSION. Namely version 2.

At no time has Linux ever been distributed without the version of the GPL 
that it is distributed under being in any question at all. THAT is my 
argument. 

And as said - you can argue against it as much as you damn well please. I 
simply don't care. I think you are very obviously wrong, but hey, in the 
end that doesn't matter either. Take it to a judge. Arguing it to me or to 
the public has absolutely zero relevance.

And regardless of what you argue, for the last 5 years there has been the 
explicit explanatory note. And you can't claim that people didn't know 
about it: if I remember correctly, you yourself sent updates to some of 
the files you felt you had copyright on to add the ".. or any later 
version" verbiage when I suggested people do so.

So why are you even arguing? It is an UNDENIABLE FACT that a noticeable 
portion of the Linux kernel is version-2 only. You'd have to do a lot of 
work if you wanted to re-license it - and the burden of proof is on _you_ 
to do so, not on me. Keeping the old license is the _only_ case that 
obviously needs no proof at all, since regardless of circumstances, it's 
always safe.

The fact is, the kernel is not licenseable under GPLv3 without tons of 
work. Work that I'm not in the least interested in doing, or even helping 
with. If you want to start such an effort, I'd suggest:

 - spend hours and hours of your time talking to your lawyers, trying to 
   convince them that your argument has any merit at all. I doubt you'll 
   be able to do that.

 - than start from the state 5 years ago.

Btw, at least in the US, intent actually matters. The fact that I've made 
it clear that my _intent_ was always GPLv2 (and I've been very consistent 
on this) together with the fact that people have accepted the addendum to 
the COPYING file actually _does_ have legal meaning. 

Weasel-wording and trying to work around the fact that the version has 
always been explicitly mentioned is not a way to make a legal argument. 

Yet that's really all your argument boils down to.

I can make very specific arguments for why version 2 ONLY is the specific 
license that covers Linux. In contrast you can only make weasel-wording 
"but you _could_ misunderstand it to mean xyzzy" kind of noises. That 
should tell you something.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

>>In short your interpretation of the past state of affairs would not
>>stand up to scrutiny.
>>    
>>
>
>You can claim anything you like. I think you're wrong. But in the 
>meantime, that doesn't matter. If it ever goes to court, you'll see what a 
>real judge will claim.
>
>My bet is that my interpretation is the only sane one. 
>
>  
>
And how many times have you actually stood in front of a Judge over IP 
and contract issues?
The language "GPLv2 or any later version" is what it is. You can change 
it moving forward, but
you cannot undo the past. You put this language in there and IT WAS WHAT 
YOU MEANT AT THE
TIME. Trying to alter that would most likely result in a finding you are 
acting in bad faith.

Anyway, I am not religious on it, v2 or v3 I don't care, but I like v3 
and I am moving my stuff to it. It's a
good idea.

Anyway, good luck figuring it out Linus. You should listen to Alan -- 
he's right and is trying to help you.

:-)

Jeff

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Tue, 31 Jan 2006, Jeff V. Merkey wrote:
>
> And how many times have you actually stood in front of a Judge over IP and
> contract issues?

Well, at least I know what I'm talking about.

> The language "GPLv2 or any later version" is what it is. You can change 
> it moving forward, but you cannot undo the past. You put this language 
> in there and IT WAS WHAT YOU MEANT AT THE TIME. Trying to alter that 
> would most likely result in a finding you are acting in bad faith.

I did _not_ put that language in, which is the whole point.

That language is in almost all GPL-licensed projects _except_ for Linux. 
It's in the FSF guidelines for what they _suggest_ people will do when 
they license something under the GPL. It's in all the FSF projects, 
obviously, and a lot of other GPL'd projects have also just mindlessly 
copied the FSF-suggested boilerplate.

Linux never did. Linux has _never_ had the "v2 or later" license wording 
in general. Go take a look. The kernel on the whole tends to not mention 
licenses in the individual files, preferring to instead rely on the 
external COPYING file that it is distributed with. That's very much on 
purpose: I personally _hate_ seeing a screenful of crapola that adds 
nothing over and over again.

In short, apart from the very early code in 1991 and early -92 (versions 
0.01 through 0.12), Linux has been licensed with _only_ the GPLv2 license 
file, and normally no mention of "v2 or later" in the actual sources.

And the way copyrights work, you have only as many rights as explicitly 
granted to you, so nobody has _ever_ had rights to re-license Linux under 
any other license than the one it came with: namely the GPLv2. Alan is 
trying to argue that the fact that it has been licensed under the GPLv2 
would somehow "magically" mean that it has been licensed under any version 
of GPL that you can pick, BUT THAT IS AN OBVIOUSLY LEGALLY FLAWED 
ARGUMENT.

It is so obviously flawed that I'm surprised he continues to argue it. 
There has _never_ been anything that says "any version of the GPL", or 
indeed just "the GPL" without any version. The version has _always_ been 
explicit: the kernel comes with the GPLv2 and no other version. If you 
don't accept the COPYING file as the license, then you had no license AT 
ALL to distribute Linux under.

So you have one choice, and once choice ONLY: accept the GPLv2 (as 
reproduced in COPYING) or don't accept the license at all. The option 
that Alan seems to want to do is "I'll take just the word 'GPL' from the 
COPYING file, and then stick to that" has simply never been an option.

Now, I can't stop Alan making stupid arguments. People can argue anything 
they damn well please, whether it makes sense or not. As SCO has shows us, 
people can argue crap for years, even in front of a judge, without any 
actual fact or paper to stand on.

And that is what Alan does. He tries to argue that the kernel has somehow 
magically been released under "the GPL" (without version specifier), even 
though the only license that it was ever released under (apart from the 
original non-GPL made-up-by-yours-truly license) very explicitly says 
which version it is, in big letters at the very top.

The fact that I made it even _more_ obvious five years ago by adding a 
further explanatory notice doesn't change anything at all, except make it 
more obvious.

Alan - talk to a lawyer. Really. Show him this email thread and my 
arguments, and ask him what he believes. I bet you can get a lawyer to 
argue your case if you _pay_ him (lawyers are whores - they are paid to 
argue for their client, not for the law), but ask him what he honestly 
thinks a judge would rule. THEN come back to me.

Because let's face it, the burden on proof on changing the kernel license 
is on _Alan_, not me. Alan is the one arguing for change. 

Now, some individual files in the kernel are dual-licensed. Some of them 
are dual-licensed with a BSD-license, others are "v2 or later version". 
The latter is by no means uncommon, but it's definitely in the minority. 
Just to give you an idea:

    [torvalds@g5 linux]$ git-ls-files '*.c' | wc -l
    7978
    [torvalds@g5 linux]$ git grep -l "any later version" '*.c' | wc -l
    2720
    [torvalds@g5 linux]$ git grep -l "Redistributions in binary form must" '*.c' | wc -l
    230

ie of the C files, only about a third have the "any later version" 
verbiage needed to be able to convert GPL v2 to v3 (and a small minority 
look like they are dual-BSD licensed - I didn't know exactly what to grep 
for, so I just picked a part of the normal BSD license, but they can 
probably also be converted to GPLv3 thanks to the BSD license being a 
strictly less restrictive license).

(I picked just the '*.c' files because that seemed fairer. If you could 
_all_ files, the "any later version" percentage drops even further).

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Linus Torvalds wrote:

>On Tue, 31 Jan 2006, Jeff V. Merkey wrote:
>  
>
>>And how many times have you actually stood in front of a Judge over IP and
>>contract issues?
>>    
>>
>
>Well, at least I know what I'm talking about.
>  
>
Linus, it's awfully complicated when a judge has to look at this stuff. 
 From my experience they have
zero understanding of high tech so they always end up looking at 
agreements and wording of contracts
and licenses, since they can understand this.

>  
>
>>The language "GPLv2 or any later version" is what it is. You can change 
>>it moving forward, but you cannot undo the past. You put this language 
>>in there and IT WAS WHAT YOU MEANT AT THE TIME. Trying to alter that 
>>would most likely result in a finding you are acting in bad faith.
>>    
>>
>
>I did _not_ put that language in, which is the whole point.
>  
>

If you can provide this beyond all doubt, then I agree you have a solid 
basis to object.
However, it being there does make the whole arguments nebulous. I would 
suggest
removing any such language from kernel.org and state GPLv2 ONLY.

>That language is in almost all GPL-licensed projects _except_ for Linux. 
>It's in the FSF guidelines for what they _suggest_ people will do when 
>they license something under the GPL. It's in all the FSF projects, 
>obviously, and a lot of other GPL'd projects have also just mindlessly 
>copied the FSF-suggested boilerplate.
>
>Linux never did. Linux has _never_ had the "v2 or later" license wording 
>in general. Go take a look. The kernel on the whole tends to not mention 
>licenses in the individual files, preferring to instead rely on the 
>external COPYING file that it is distributed with. That's very much on 
>purpose: I personally _hate_ seeing a screenful of crapola that adds 
>nothing over and over again.
>
>In short, apart from the very early code in 1991 and early -92 (versions 
>0.01 through 0.12), Linux has been licensed with _only_ the GPLv2 license 
>file, and normally no mention of "v2 or later" in the actual sources.
>  
>

Again, get rid of this language completely about later versions.

>And the way copyrights work, you have only as many rights as explicitly 
>granted to you, so nobody has _ever_ had rights to re-license Linux under 
>any other license than the one it came with: namely the GPLv2. Alan is 
>trying to argue that the fact that it has been licensed under the GPLv2 
>would somehow "magically" mean that it has been licensed under any version 
>of GPL that you can pick, BUT THAT IS AN OBVIOUSLY LEGALLY FLAWED 
>ARGUMENT.
>  
>

It's not. I was also under the impression based on the language "any 
later license"
and I am a very bright chap. So if I got it wrong, then imagine how many 
other
folks are likely to be confused.

>It is so obviously flawed that I'm surprised he continues to argue it. 
>There has _never_ been anything that says "any version of the GPL", or 
>indeed just "the GPL" without any version. The version has _always_ been 
>explicit: the kernel comes with the GPLv2 and no other version. If you 
>don't accept the COPYING file as the license, then you had no license AT 
>ALL to distribute Linux under.
>
>So you have one choice, and once choice ONLY: accept the GPLv2 (as 
>reproduced in COPYING) or don't accept the license at all. The option 
>that Alan seems to want to do is "I'll take just the word 'GPL' from the 
>COPYING file, and then stick to that" has simply never been an option.
>
>Now, I can't stop Alan making stupid arguments. People can argue anything 
>they damn well please, whether it makes sense or not. As SCO has shows us, 
>people can argue crap for years, even in front of a judge, without any 
>actual fact or paper to stand on.
>  
>

Alan is trying to help you. I have never seen him do anything other than 
support you
to the hilt. Sure, disagreements happen, but he is there for you and 
Linux and has
been from day one.

>And that is what Alan does. He tries to argue that the kernel has somehow 
>magically been released under "the GPL" (without version specifier), even 
>though the only license that it was ever released under (apart from the 
>original non-GPL made-up-by-yours-truly license) very explicitly says 
>which version it is, in big letters at the very top.
>
>The fact that I made it even _more_ obvious five years ago by adding a 
>further explanatory notice doesn't change anything at all, except make it 
>more obvious.
>
>Alan - talk to a lawyer. Really. Show him this email thread and my 
>arguments, and ask him what he believes. I bet you can get a lawyer to 
>argue your case if you _pay_ him (lawyers are whores - they are paid to 
>argue for their client, not for the law), but ask him what he honestly 
>thinks a judge would rule. THEN come back to me.
>  
>

Linus, remove all nebulous language and post a notice on kernel.org 
clarifying your
position on this code, and I think the issue becomes closed. There's 
still the possiblity
that under the doctrine of esstopel, someone can claim or will try to 
claim conversion
to GPLv3. You will have to address this when and if it happens.

>Because let's face it, the burden on proof on changing the kernel license 
>is on _Alan_, not me. Alan is the one arguing for change. 
>  
>

A change to GPLv3 would be a good thing for you.

>Now, some individual files in the kernel are dual-licensed. Some of them 
>are dual-licensed with a BSD-license, others are "v2 or later version". 
>The latter is by no means uncommon, but it's definitely in the minority. 
>Just to give you an idea:
>
>    [torvalds@g5 linux]$ git-ls-files '*.c' | wc -l
>    7978
>    [torvalds@g5 linux]$ git grep -l "any later version" '*.c' | wc -l
>    2720
>    [torvalds@g5 linux]$ git grep -l "Redistributions in binary form must" '*.c' | wc -l
>    230
>
>ie of the C files, only about a third have the "any later version" 
>verbiage needed to be able to convert GPL v2 to v3 (and a small minority 
>look like they are dual-BSD licensed - I didn't know exactly what to grep 
>for, so I just picked a part of the normal BSD license, but they can 
>probably also be converted to GPLv3 thanks to the BSD license being a 
>strictly less restrictive license).
>
>(I picked just the '*.c' files because that seemed fairer. If you could 
>_all_ files, the "any later version" percentage drops even further).
>  
>

Given, the whole kernel cannot claim multiple licensing -- you have 
convinced me
on this point.

Jeff

>			Linus
>
>  
>

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Tue, 31 Jan 2006, Jeff V. Merkey wrote:
>
> > I did _not_ put that language in, which is the whole point.
> 
> If you can provide this beyond all doubt, then I agree you have a solid 
> basis to object.

It's really easy to prove.

Look at core kernel source code today, and look at it 10 years ago. Look 
at it 15 years ago. Nothing has changed.

The really core files have copyright notices like this:

	/*
	 *  kernel/sched.c
	 *
	 *  Kernel scheduler and related syscalls
	 *
	 *  Copyright (C) 1991-2002  Linus Torvalds
	...

with absolutely no mention of any license rights at all. Not "this is 
under the GPL", not "GPLv2 or later". The _only_ license rights anybody 
ever had to those files come from the COPYING file, which very clearly 
states that it's "version 2, 1991"

(And yes, I'm a lazy bastard. I don't update the years. Some of the files 
I wrote still say "1991, 1992" even though they've obviously been edited 
since by me. If they fall into the public domain a couple of years 
earlier, I really don't see myself caring, since I will have been dead for 
a long while by that time _anyway_, judging by the current copyright 
nonsense).

[ Side note: the _core_ kernel files are more universally GPL v2-only than 
  the rest of the kernel. So for example, while almost a third of all C 
  files have the "any later version" notice in them, when you look at just 
  the core files in kernel/ mm/ fs/, it's a _lot_ less rare. For example, 
  in fs/*.c, it's only two files out of 57, and those aren't even the most 
  core files.

  So _qualitatively_, a lot more than "just" two thirds of the kernel are 
  based on my core files, and are GPLv2 _only_. The "..any later version" 
  stuff tends to exist mostly in drivers (and some filesystems: 9pfs, 
  afs, autofs, cifs, jfs, ntfs, ocfs2 have the "any later version" in 
  them, but the most common ones do not, and are often derived 
  (admittedly very indirectly, by now) from my original code. ]

> However, it being there does make the whole arguments 
> nebulous. I would suggest removing any such language from kernel.org and 
> state GPLv2 ONLY.

The COPYING file was edited (over _five_ years ago) to clarify the issue, 
exactly because some people were confused. So the COPYING file now 
explicitly says:

 Also note that the only valid version of the GPL as far as the kernel
 is concerned is _this_ particular version of the license (ie v2, not
 v2.2 or v3.x or whatever), unless explicitly otherwise stated.

and that has been the case for the last 5+ years.

(Another clarification is even older: the clarification that using "normal 
system calls" is _not_ considered linking, and thus the GPL doesn't infect 
any normal user-level programs. That one is over ten years old, since some 
people seriously worried about it. Again, it was really pretty obvious 
from the license itself, but the clarification made the question stop and 
made some people stop worrying unnecessarily).

Alan argues that that extra notice "changed" the license (and that any 
code that is older than 5 years would somehow not be GPLv2). I argue 
otherwise. I argue that for the whole history, Linux has been v2-only 
unless otherwise explicitly specified.

And I don't think even Alan will argue that the "v2 only" thing hasn't 
been true for the last five years.

> I was also under the impression based on the language "any later license"
> and I am a very bright chap. So if I got it wrong, then imagine how many other
> folks are likely to be confused.

Exactly. That's why I added the clarification on top of the COPYING file: 
people _have_ been confused.

That confusion doesn't stem from Linux, btw, but from the FSF distribution 
of the GPLv2 license itself. The license is distributed as one single 
file, which actually contains three parts: (1) the "preamble", (2) the 
actual license itself and (3) the "How to Apply These Terms to Your New 
Programs" mini-FAQ.

And that third part actually contains the wording "(at your option) any 
later version.", but a lot of people seemed to not realize that this was 
just part of a FSF-suggested boiler-plate on what to put in your source 
files.  In other words, that was never actually part of the license 
itself, but just a "btw, here's how you should use it" post-script.

A lot of people seemed to be confused by that, and this is exactly why the 
Linux COPYING file got the additional explanation.

(Side note: from a legal standpoint, "intent" does actually matter in the 
US legal system. So the FSF can actually argue that their pre-amble and 
their post-script to the license carry legal weight, because it shows 
their _intent_. However, they can only argue that for programs that they 
own copyright to, or when the license itself might be unclear - they can't 
argue that it shows _my_ intent. I've made my intent very clear over the 
years, and I've been consistent on this matter, so nobody can claim that 
I've "changed the rules").

> Alan is trying to help you. I have never seen him do anything other than 
> support you to the hilt. Sure, disagreements happen, but he is there for 
> you and Linux and has been from day one.

Absolutely. And I actually try to be very open to changing my mind if 
somebody has a valid point. Open source is absolutely not about just the 
source code - it's very much about the process, and about (mostly the lack 
of) control.

And hey, Alan tends to be mostly right in his concerns. Which is why he's 
so respected in the community. I just think that he is off the deep end on 
this one, and I have yet to see any actual convincing arguments for his 
standpoint. 

> Linus, remove all nebulous language and post a notice on kernel.org 
> clarifying your position on this code, and I think the issue becomes 
> closed.

The thing is, even the _clarification_ HAS BEEN THERE FOR 5 YEARS. At the 
very top of the COPYING file.

This really is nothing new. How much more prominent can it be than be in 
the top-level COPYING file that gets distributed with every single kernel 
version?

> > Because let's face it, the burden on proof on changing the kernel license is
> > on _Alan_, not me. Alan is the one arguing for change.  
> 
> A change to GPLv3 would be a good thing for you.

A lot of people like the GPLv3. I personally don't _dislike_ the current 
draft, but I don't think it's appropriate for the kernel. Part of why I 
liked the GPL in the first place (v2 at that point, obviously - v3 hadn't 
even been thought about) was that it put no restrictions at all on the 
_use_ of binaries. 

So I actually prefer the GPLv2. I don't think the current GPLv3 draft is 
"evil" or "bad", or anything like that, but it's not the license I would 
have selected when I started, and I don't see any reason to change to it 
for the kernel.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Linus Torvalds wrote:

>On Tue, 31 Jan 2006, Jeff V. Merkey wrote:
>  
>
>>>I did _not_ put that language in, which is the whole point.
>>>      
>>>
>>If you can provide this beyond all doubt, then I agree you have a solid 
>>basis to object.
>>    
>>
>
>It's really easy to prove.
>
>Look at core kernel source code today, and look at it 10 years ago. Look 
>at it 15 years ago. Nothing has changed.
>
>The really core files have copyright notices like this:
>
>	/*
>	 *  kernel/sched.c
>	 *
>	 *  Kernel scheduler and related syscalls
>	 *
>	 *  Copyright (C) 1991-2002  Linus Torvalds
>	...
>
>with absolutely no mention of any license rights at all. Not "this is 
>under the GPL", not "GPLv2 or later". The _only_ license rights anybody 
>ever had to those files come from the COPYING file, which very clearly 
>states that it's "version 2, 1991"
>
>(And yes, I'm a lazy bastard. I don't update the years. Some of the files 
>I wrote still say "1991, 1992" even though they've obviously been edited 
>since by me. If they fall into the public domain a couple of years 
>earlier, I really don't see myself caring, since I will have been dead for 
>a long while by that time _anyway_, judging by the current copyright 
>nonsense).
>
>[ Side note: the _core_ kernel files are more universally GPL v2-only than 
>  the rest of the kernel. So for example, while almost a third of all C 
>  files have the "any later version" notice in them, when you look at just 
>  the core files in kernel/ mm/ fs/, it's a _lot_ less rare. For example, 
>  in fs/*.c, it's only two files out of 57, and those aren't even the most 
>  core files.
>
>  So _qualitatively_, a lot more than "just" two thirds of the kernel are 
>  based on my core files, and are GPLv2 _only_. The "..any later version" 
>  stuff tends to exist mostly in drivers (and some filesystems: 9pfs, 
>  afs, autofs, cifs, jfs, ntfs, ocfs2 have the "any later version" in 
>  them, but the most common ones do not, and are often derived 
>  (admittedly very indirectly, by now) from my original code. ]
>
>  
>
OK. This one might make it.

>>However, it being there does make the whole arguments 
>>nebulous. I would suggest removing any such language from kernel.org and 
>>state GPLv2 ONLY.
>>    
>>
>
>The COPYING file was edited (over _five_ years ago) to clarify the issue, 
>exactly because some people were confused. So the COPYING file now 
>explicitly says:
>
> Also note that the only valid version of the GPL as far as the kernel
> is concerned is _this_ particular version of the license (ie v2, not
> v2.2 or v3.x or whatever), unless explicitly otherwise stated.
>
>and that has been the case for the last 5+ years.
>
>(Another clarification is even older: the clarification that using "normal 
>system calls" is _not_ considered linking, and thus the GPL doesn't infect 
>any normal user-level programs. That one is over ten years old, since some 
>people seriously worried about it. Again, it was really pretty obvious 
>from the license itself, but the clarification made the question stop and 
>made some people stop worrying unnecessarily).
>
>Alan argues that that extra notice "changed" the license (and that any 
>code that is older than 5 years would somehow not be GPLv2). I argue 
>otherwise. I argue that for the whole history, Linux has been v2-only 
>unless otherwise explicitly specified.
>
>And I don't think even Alan will argue that the "v2 only" thing hasn't 
>been true for the last five years.
>  
>
Got it. Agreed. There was a COPYING file inside the tar.gz (you need to 
check this one)
I recall seeing but I think it was 2.4 and some header files that still 
said this -- better clean those up.

>  
>
>>I was also under the impression based on the language "any later license"
>>and I am a very bright chap. So if I got it wrong, then imagine how many other
>>folks are likely to be confused.
>>    
>>
>
>Exactly. That's why I added the clarification on top of the COPYING file: 
>people _have_ been confused.
>
>That confusion doesn't stem from Linux, btw, but from the FSF distribution 
>of the GPLv2 license itself. The license is distributed as one single 
>file, which actually contains three parts: (1) the "preamble", (2) the 
>actual license itself and (3) the "How to Apply These Terms to Your New 
>Programs" mini-FAQ.
>
>And that third part actually contains the wording "(at your option) any 
>later version.", 
>

Take this out of the GPL license. It's granting as a term of the license 
the ability to
adopt later versions. An obvious loophole -- modify the license text and 
state
LINE NUMBER AND PARAGRAPH that this language DOES NOT APPLY
and BE SPECIFIC in your explanation. If it's not numbered in the GPL 
text, number
it so you can make it CLEAR.

>but a lot of people seemed to not realize that this was 
>just part of a FSF-suggested boiler-plate on what to put in your source 
>files.  In other words, that was never actually part of the license 
>itself, but just a "btw, here's how you should use it" post-script.
>
>A lot of people seemed to be confused by that, and this is exactly why the 
>Linux COPYING file got the additional explanation.
>
>(Side note: from a legal standpoint, "intent" does actually matter in the 
>US legal system. So the FSF can actually argue that their pre-amble and 
>their post-script to the license carry legal weight, because it shows 
>their _intent_. However, they can only argue that for programs that they 
>own copyright to, or when the license itself might be unclear - they can't 
>argue that it shows _my_ intent. I've made my intent very clear over the 
>years, and I've been consistent on this matter, so nobody can claim that 
>I've "changed the rules").
>
>  
>
>>Alan is trying to help you. I have never seen him do anything other than 
>>support you to the hilt. Sure, disagreements happen, but he is there for 
>>you and Linux and has been from day one.
>>    
>>
>
>Absolutely. And I actually try to be very open to changing my mind if 
>somebody has a valid point. Open source is absolutely not about just the 
>source code - it's very much about the process, and about (mostly the lack 
>of) control.
>
>And hey, Alan tends to be mostly right in his concerns. Which is why he's 
>so respected in the community. I just think that he is off the deep end on 
>this one, and I have yet to see any actual convincing arguments for his 
>standpoint. 
>  
>

Alan is awesome -- period.

>  
>
>>Linus, remove all nebulous language and post a notice on kernel.org 
>>clarifying your position on this code, and I think the issue becomes 
>>closed.
>>    
>>
>
>The thing is, even the _clarification_ HAS BEEN THERE FOR 5 YEARS. At the 
>very top of the COPYING file.
>  
>
CITE REFERENCES. Number the GPL and Cite where it is invalid.

>This really is nothing new. How much more prominent can it be than be in 
>the top-level COPYING file that gets distributed with every single kernel 
>version?
>
>  
>
>>>Because let's face it, the burden on proof on changing the kernel license is
>>>on _Alan_, not me. Alan is the one arguing for change.  
>>>      
>>>
>>A change to GPLv3 would be a good thing for you.
>>    
>>
>
>A lot of people like the GPLv3. I personally don't _dislike_ the current 
>draft, but I don't think it's appropriate for the kernel. Part of why I 
>liked the GPL in the first place (v2 at that point, obviously - v3 hadn't 
>even been thought about) was that it put no restrictions at all on the 
>_use_ of binaries. 
>
>So I actually prefer the GPLv2. I don't think the current GPLv3 draft is 
>"evil" or "bad", or anything like that, but it's not the license I would 
>have selected when I started, and I don't see any reason to change to it 
>for the kernel.
>  
>

Lets see where it goes. It protects the end users and gives them the 
freedom to combine
GPL and non-GPL stuff, which is great.

Jeff


>			Linus
>
>  
>

Re: GPL V3 and Linux - Dead Copyright Holders

[Matan Peled]

Jeff V. Merkey wrote:
> Got it. Agreed. There was a COPYING file inside the tar.gz (you need to 
> check this one)
> I recall seeing but I think it was 2.4 and some header files that still 
> said this -- better clean those up.

You can't really do this part. If you make changes to the GPL, you're not 
allowed to call it the GPL anymore... (See the FAQ)

-- 
[Name      ]   ::  [Matan I. Peled    ]
[Location  ]   ::  [Israel            ]
[Public Key]   ::  [0xD6F42CA5        ]
[Keyserver ]   ::  [keyserver.kjsl.com]
encrypted/signed  plain text  preferred

Re: GPL V3 and Linux - Dead Copyright Holders

[Matan Peled]

David Leimbach wrote:
> But I've seen this done in GNU projects many times.  There are "special 
> exceptions" for things like the Objective-C runtime.  These things seem 
> strange to me as well but the name of the license is still GPL.
> 
> See the following link:
> http://darwinsource.opendarwin.org/August2002GCCUpdate/gcc3-1161/libobjc/objects.c
> 
> Dave

Maybe, but I think adding a "special exception" is entirely different than just 
outright removing a part of the license.

But IANAL...


-- 
[Name      ]   ::  [Matan I. Peled    ]
[Location  ]   ::  [Israel            ]
[Public Key]   ::  [0xD6F42CA5        ]
[Keyserver ]   ::  [keyserver.kjsl.com]
encrypted/signed  plain text  preferred

Re: GPL V3 and Linux - Dead Copyright Holders

[Paul Jakma]

On Tue, 31 Jan 2006, Linus Torvalds wrote:

> under the GPL", not "GPLv2 or later". The _only_ license rights anybody
> ever had to those files come from the COPYING file, which very clearly
> states that it's "version 2, 1991"

No one has disputed that I think. My understanding of that text, as 
in:


                     GNU GENERAL PUBLIC LICENSE
                        Version 2, June 1991

Is that it refers as the version of the license itself. The 'revision 
ID' of the published license. And that very version of the GPL has a 
clause to allow version 'conversion' in section 9 to other versions.

> The COPYING file was edited (over _five_ years ago) to clarify the issue,

That wasn't clarifying the issue, that was to make sure section 9's 
"any version" could no longer apply. You even seem to have 
acknowledged that very text of section 9 in your announcement of 
2.4.0-test8's COPYING change:

`There's been some discussions of a GPL v3 which would limit 
licensing to certain "well-behaved" parties, and I'm not sure I'd 
agree with such restrictions - and the GPL itself allows for "any 
version" so I wanted to make this part unambigious as far as my 
personal code is concerned."'

From:

http://www.uwsg.iu.edu/hypermail/linux/kernel/0009.1/0096.html

Until you made that change, a reasonable person might have presumed 
that lack of version statement (no, the actual GPL license version at 
the top of the GPL itself does not count ;) ) would mean section 9 
would have applied.

I'm not familiar enough with Linux kernel history to know whether the 
implicit "only version 2" view was widely known amongst kernel 
developers back then, prior to the edit.

> and that has been the case for the last 5+ years.

Right. The "version 2 only" only has been the status quo for 5+ years 
now, so most people must be content with it, grudgingly or not.

> Exactly. That's why I added the clarification on top of the COPYING 
> file: people _have_ been confused.

Read your own email again. :)

> That confusion doesn't stem from Linux, btw, but from the FSF distribution
> of the GPLv2 license itself. The license is distributed as one single
> file, which actually contains three parts: (1) the "preamble", (2) the
> actual license itself and (3) the "How to Apply These Terms to Your New
> Programs" mini-FAQ.
>
> And that third part actually contains the wording "(at your option) any
> later version.",

Right.

But the actual /license/, the second part, says none specified -> 
"any version ever published" in section 9. The "any later version" 
text is in the 3rd part, the suggested preamble.

> files.  In other words, that was never actually part of the license
> itself, but just a "btw, here's how you should use it" post-script.

Right. But there are *two* (any.*version):

1. "Any version", which is:

 	a) In the license itself, to cover case where programme
 	   doesn't specify which version(s) apply, as section 9.
 	   (ie in the "second part").

 	b) the text you refered to in the 2.4.0-test8 above (ie
  	   you were looking at section 9 back then, not the preamble ;) )

2. "Any later version", which is what is mentioned in the preamble
    (the third part).

Alan's reply to you was on the basis of 1. You're arguing based on 2, 
having apparently completely overlooked 1 (though you apparently had 
not back when you composed that 2.4.0-test8 email).

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
God is real, unless declared integer.

Re: GPL V3 and Linux - Dead Copyright Holders

[Stephen Hemminger]

Please take me off this thread.
If I want a sophomoric soporific, then I'll read it on LKML.

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Tue, 31 Jan 2006, Paul Jakma wrote:
>
> http://www.uwsg.iu.edu/hypermail/linux/kernel/0009.1/0096.html
> 
> Until you made that change, a reasonable person might have presumed that lack
> of version statement (no, the actual GPL license version at the top of the GPL
> itself does not count ;) ) would mean section 9 would have applied.

Actually, I don't think it was really debatable even before that, but yes, 
I wanted to cover my ass. _Exactly_ because of that confusion over "any 
version" in section 9.

IOW, it isn't a new confusion. It's an old confusion that I've always felt 
was silly - and incorrect - and that I wanted to address explicitly, 
exactly so that there wouldn't be any ambiguoity.

So I maintain that "version 2" has always been the version of the GPL as 
it pertains to the kernel as far as I'm concerned, but exactly because 
some other people have been confused, I wanted to make it not only 
explicit, but also mention it publicly when the clarification was done, so 
that people who _had_ been confused could just add in the necessary 
verbiage so that code that they owned would fall under the license they 
had _thought_ was the right one.

Note how I called it a "clarification" even back then. It wasn't a change 
to make things GPLv2 - it was _clarifying_ that the kernel had always been 
GPLv2, and acknowledging that there had indeed been confusion.

And hey, if people want to put the confusion at my door, go right ahead. 
I'm just happy I did it five years ago, so that by _now_, things are 
hopefully pretty damn clear-cut.

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Gene Heskett]

On Tuesday 31 January 2006 16:45, Paul Jakma wrote:
>On Tue, 31 Jan 2006, Linus Torvalds wrote:
>> under the GPL", not "GPLv2 or later". The _only_ license rights
>> anybody ever had to those files come from the COPYING file, which
>> very clearly states that it's "version 2, 1991"
>
>No one has disputed that I think. My understanding of that text, as
>in:
>
>
>                     GNU GENERAL PUBLIC LICENSE
>                        Version 2, June 1991
>
>Is that it refers as the version of the license itself. The 'revision
>ID' of the published license. And that very version of the GPL has a
>clause to allow version 'conversion' in section 9 to other versions.
>
>> The COPYING file was edited (over _five_ years ago) to clarify the
>> issue,
>
>That wasn't clarifying the issue, that was to make sure section 9's
>"any version" could no longer apply. You even seem to have
>acknowledged that very text of section 9 in your announcement of
>2.4.0-test8's COPYING change:
>
>`There's been some discussions of a GPL v3 which would limit
>licensing to certain "well-behaved" parties, and I'm not sure I'd
>agree with such restrictions - and the GPL itself allows for "any
>version" so I wanted to make this part unambigious as far as my
>personal code is concerned."'
>
>From:
>
>http://www.uwsg.iu.edu/hypermail/linux/kernel/0009.1/0096.html
>
>Until you made that change, a reasonable person might have presumed
>that lack of version statement (no, the actual GPL license version at
>the top of the GPL itself does not count ;) ) would mean section 9
>would have applied.
>
>I'm not familiar enough with Linux kernel history to know whether the
>implicit "only version 2" view was widely known amongst kernel
>developers back then, prior to the edit.
>
>> and that has been the case for the last 5+ years.
>
>Right. The "version 2 only" only has been the status quo for 5+ years
>now, so most people must be content with it, grudgingly or not.
>
>> Exactly. That's why I added the clarification on top of the COPYING
>> file: people _have_ been confused.
>
>Read your own email again. :)
>
>> That confusion doesn't stem from Linux, btw, but from the FSF
>> distribution of the GPLv2 license itself. The license is distributed
>> as one single file, which actually contains three parts: (1) the
>> "preamble", (2) the actual license itself and (3) the "How to Apply
>> These Terms to Your New Programs" mini-FAQ.
>>
>> And that third part actually contains the wording "(at your option)
>> any later version.",
>
>Right.
>
>But the actual /license/, the second part, says none specified ->
>"any version ever published" in section 9. The "any later version"
>text is in the 3rd part, the suggested preamble.
>
>> files.  In other words, that was never actually part of the license
>> itself, but just a "btw, here's how you should use it" post-script.
>
>Right. But there are *two* (any.*version):
>
>1. "Any version", which is:
>
>  a) In the license itself, to cover case where programme
>     doesn't specify which version(s) apply, as section 9.
>     (ie in the "second part").
>
>  b) the text you refered to in the 2.4.0-test8 above (ie
>      you were looking at section 9 back then, not the preamble ;) )
>
>2. "Any later version", which is what is mentioned in the preamble
>    (the third part).
>
>Alan's reply to you was on the basis of 1. You're arguing based on 2,
>having apparently completely overlooked 1 (though you apparently had
>not back when you composed that 2.4.0-test8 email).
>
>regards,

Well, IANAL either, but section 9 isn't the least bit ambiguous in my 
opinion.  To quote what everyone with a functioning eye can read:

  9. The Free Software Foundation may publish revised and/or new 
versions of the General Public License from time to time.  Such new 
versions will be similar in spirit to the present version, but may 
differ in detail to address new problems or concerns.

Each version is given a distinguishing version number.  If the Program
specifies a version number of this License which applies to it and "any
later version", you have the option of following the terms and 
conditions either of that version or of any later version published by 
the Free Software Foundation.  If the Program does not specify a 
version number of this License, you may choose any version ever 
published by the Free Software Foundation.

End of snip.  The point is that version 2 is spelled out at the top of 
the COPYING file and the word "and" preceeds the "any later version" in 
the above paragraph, which it does not in the top of the COPYING file 
specifying the license, means that its version 2 and thats the end of 
it.  There is no "any later version" in the statement specifying the 
license version at the top of the COPYING file.

Why is that so hard to grasp?

-- 
Cheers, Gene
People having trouble with vz bouncing email to me should add the word
'online' between the 'verizon', and the dot which bypasses vz's
stupid bounce rules.  I do use spamassassin too. :-)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2006 by Maurice Eugene Heskett, all rights reserved.

Re: GPL V3 and Linux - Dead Copyright Holders

[Rene Herman]

Linus Torvalds wrote:

> Alan argues that that extra notice "changed" the license (and that any 
> code that is older than 5 years would somehow not be GPLv2). I argue 
> otherwise. I argue that for the whole history, Linux has been v2-only 
> unless otherwise explicitly specified.

Hope you don't mind an opinion from a bystander...

I actually believe that Alan Cox is making a fair point, when viewed 
from the perspective of the strict language of the license.

The license as distributed with the kernel itself states that unless the 
program specifies the version of the GPL that it's under, any version 
will do. Alan makes the point that at least upto the 2.4.0-test8 
addition, the program never specified the version, as the "v2" was only 
in the _license_, which is not the _program_.

And this is not a bad point -- the license and the program are indeed 
not the same; they are not even copyrighted by the same people. With the 
addition to the COPYING file, there's something added which clearly is 
yours, but before that it was just the generic GPL, copyrighted by the FSF.

As additional "proof" of the fact that the license can not be considered 
part of "the program" he pointed out that the GPL document in itself is 
not GPL compatible, meaning that under this strict interpretation it 
could even be argued that there would be a legal problem in combining 
this document with the rest of the program.

Sure, I noted all the "intent" stuff, and you may feel an interpretation 
this strict is not sensible, but I really do believe he's making a fair 
point if you do not want to trust the law, and all judges, to always be 
as sensible as you want them to be.

> And I don't think even Alan will argue that the "v2 only" thing
> hasn't been true for the last five years.

I would not at least. The added bit needs to be considered part of the 
program itself, solving the issue.

Rene.

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Tue, 31 Jan 2006, Rene Herman wrote:
> 
> And this is not a bad point -- the license and the program are indeed not the
> same; they are not even copyrighted by the same people.

Umm.. I really think that is a total strawman.

_Most_ of the kernel isn't copyrighted by "the same people". Many parts of 
the kernel are available from tons of different people (and are sometimes 
available under different licenses). That doesn't make them less part of 
the kernel program.

And btw, the GPL (the license text) is not incompatible with itself. Yes, 
the license says

 "Everyone is permitted to copy and distribute verbatim copies
 of this license document, but changing it is not allowed."

but the fact is, the GPL also says that any license notices must be kept 
intact, and that a copy of the GPL itself must be given along with the 
program (in section 1).

The requirement that you be able to modify and distribute the modified 
(section 2) is only _provided_ you also honor section 1. So the fact that 
you're not allowed to change the license text is _not_ against the GPL 
itself, and including the license as part of the program is in no way 
against the GPL.

So the fact that the license has a different copyright and is written by 
different people is in _no_ way different from the fact that other parts 
of the kernel were perhaps originally BSD-licensed. GPLv2 can happily 
cover those cases.

So go back instead to "section 0". Now THAT is I think the relevant one:

  "0. This License applies to any program or other work which contains
  a notice placed by the copyright holder saying it may be distributed
  under the terms of this General Public License. ..."

Notice how the GPLv2 text says that it applies to any program that just 
says it is licensed under the General Public License.

I'm convinced _that_ is how you get "no version specified" in section 9. 
You have a program that just says "This is licensed under the GPL", 
instead of doing the full thing.

And I say that the Linux kernel has contained a notice placed by the 
copyright holder (the "COPYING" file) that says that it's to be 
distributed under (and I quote from the top):

                    GNU GENERAL PUBLIC LICENSE
                       Version 2, June 1991

and that's it.

Ok. Back to work. This thread has been _waa-aayy_ too long.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Paul Jakma]

On Tue, 31 Jan 2006, Linus Torvalds wrote:

> And btw, the GPL (the license text) is not incompatible with itself. Yes,
> the license says
>
> "Everyone is permitted to copy and distribute verbatim copies
> of this license document, but changing it is not allowed."

Right, that's to discourage any hasty and likely ill-advised 'local' 
changes to the GPL. The FSF do grant permission to modify the GPL 
though:

 	http://www.gnu.org/licenses/gpl-faq.html#ModifyGPL

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
If there is no wind, row.
 		-- Polish proverb

Re: GPL V3 and Linux - Dead Copyright Holders

[Paul Jakma]

<trimming the wide Cc list due to complaints>

On Tue, 31 Jan 2006, Linus Torvalds wrote:

> Notice how the GPLv2 text says that it applies to any program that 
> just says it is licensed under the General Public License.
>
> I'm convinced _that_ is how you get "no version specified" in section 9.
> You have a program that just says "This is licensed under the GPL",
> instead of doing the full thing.

> And I say that the Linux kernel has contained a notice placed by the
> copyright holder (the "COPYING" file) that says that it's to be
> distributed under (and I quote from the top):
>
>                    GNU GENERAL PUBLIC LICENSE
>                       Version 2, June 1991
>
> and that's it.

Well, most people would recognise this to be exactly the same text of 
the GPLv2 as published by the FSF, as used by many software 
programmes:

 	http://www.gnu.org/licenses/gpl.txt

and hence would have trouble recognising that you intended this to 
also be the text governing exactly which version(s) of the GPL apply 
linux specifically.

Note that if what you say is correct, that the immutable header of 
the GPLv2 acts so as to apply a version restriction on any specific 
covered programme per default, then that would mean:

1a. The "If the Program does not specify a version" text of Section 9
     of the GPLv2 is utterly without effect and meaningless.

1b. Indeed, the whole of section 9 might be without effect (the
     version is already specified). (It'd depend on the exact
     text of the preamble maybe, but it'd be quite ambigious).

2. Hence all GPLv2 programmes which do not have include some
    additional "or any later version" preamble text definitely will
    not automatically upgrade to GPLv3 when it is finally published.

You've mentioned intent a few times before as being a strong factor 
in interpretation[1], you have to therefore ask what RMS and the FSF 
intended when they included section 9: did they really intend that 
the GPLv2's 'Version 2, June 1991' would completely override the text 
regarding cases where "the Program does not specify a version number 
of this License"? The intent surely of section 9 was to allow for the 
GPLv2 to be upgraded smoothly, even where a programme applied it 
imprecisely.

Hence, logically, the intent of the framers must have been that the 
'Version 2, June 1991' was to act as a version identifier for the GPL 
text only, rather than to be interpreted as the version applying to 
the whole programme.

That the licence then goes on to discuss how to apply the GPL (the 
third part you mentioned before), and mention how to do so (including 
the "any later version" text) further suggests that simply including 
a copy of the the text of the license itself does not constitute 
tying down the version.

Note that I'm only arguing with you on your interpretation of the GPL 
- which I believe is flawed and deserves to be contradicted lest 
others assume it - not on which version of the GPL applies to Linux 
in its aggregate (which definitely is "v2 only", and has been for 
ages).

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
Ernest asks Frank how long he has been working for the company.
 	"Ever since they threatened to fire me."

Re: GPL V3 and Linux - Dead Copyright Holders

[Rene Herman]

Linus Torvalds wrote:

> _Most_ of the kernel isn't copyrighted by "the same people". Many 
> parts of the kernel are available from tons of different people (and 
> are sometimes available under different licenses). That doesn't make 
> them less part of the kernel program.

Granted, I could have probably done without the "copyrighted by the same 
people" argument. What I attempted to emphasize is that the license text 
itself is an independent work, licensed under a different license than 
the program.

> And btw, the GPL (the license text) is not incompatible with itself.
> Yes, the license says
> 
> "Everyone is permitted to copy and distribute verbatim copies of this
> license document, but changing it is not allowed."
> 
> but the fact is, the GPL also says that any license notices must be
> kept intact, and that a copy of the GPL itself must be given along
> with the program (in section 1).

The GPL text itself really is not under a GPL compatible license. The 
GPL says that, under certain provisions, "You may modify your copy or 
copies of the Program or any portion of it" (section 2) while the GPL 
document does not allow any modification, under any provision: "but 
changing it is not allowed".

The bits above (also note how you said "along _with_ the program" and 
how the GPL itself says "applies _to_ the program) only further enhance 
the point with which I by now fully agree that the GPL text itself is 
not part of the program. As such, something other than the GPL text will 
have to specify a version to have the _program_ specify a version per 
section 9.

Your addition to the COPYING file certainly does just that, meaning any 
possible problem has been long solved.

> The requirement that you be able to modify and distribute the
> modified (section 2) is only _provided_ you also honor section 1. So
> the fact that you're not allowed to change the license text is _not_
> against the GPL itself, and including the license as part of the
> program is in no way against the GPL.

To be GPL compatible, you need to be compatible with all requirements of 
  the GPL, one of which is section 2. The licence the GPL text is under 
is not compatible with this section.

No, it's not against the program's license (the program being Linux and 
the license the GPL v2) to include the license with the program. There 
would only be a potential problem if the GPL text were to be considered 
_part of_ the program since then combining the GPL licensed program and 
the license text would require the license text to be under a GPL 
compatible license. As it is, it is included as a mere aggregation, and 
this is certainly fine.

Hope I'm not annoying you -- these are obviously points normally only a 
lawyer would enjoy arguing. If I am, maybe pointing out the thoroughly 
amusing recursion of having to distribute the GPL alongside the GPL if 
the GPL text itself would be licensed under the GPL itself (not even 
just a compatible license) makes it better? (<-- joke...).

> I'm convinced _that_ is how you get "no version specified" in section
> 9. You have a program that just says "This is licensed under the
> GPL", instead of doing the full thing.
> 
> And I say that the Linux kernel has contained a notice placed by the 
> copyright holder (the "COPYING" file) that says that it's to be 
> distributed under (and I quote from the top):
> 
> GNU GENERAL PUBLIC LICENSE Version 2, June 1991
> 
> and that's it.

I just saw Paul Jakma make a very strong point here. Section 9 would be 
useless if having the version in the header, of the same document, would 
be all that is required to have the program specify a version.

Having said all that... everybody agrees that the bit you added to the 
COPYING file is enough to have the kernel under the GPL v2 only. So, 
there's no need for this, but do you think it could be a good idea to 
add a file LICENSE, copyrighted by you, to the kernel that says that the 
kernel is licensed under the GPL v2, as supplied in the file COPYING 
(and moving the syscall bit as well, and restore COPYING to pristine)?

No, this would not change anything other than provide for an even _more_ 
unambiguous situation. In that case, noone would have to argue whether 
or not anything added to the COPYING file was part of the program 
proper. On the other hand, if doing so would make some people believe 
that anything changed, it could be counter-productive as well.

Wanted to suggest it though...

Rene.

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Wed, 1 Feb 2006, Rene Herman wrote:
> > 
> > but the fact is, the GPL also says that any license notices must be
> > kept intact, and that a copy of the GPL itself must be given along
> > with the program (in section 1).
> 
> The GPL text itself really is not under a GPL compatible license. The GPL says
> that, under certain provisions, "You may modify your copy or copies of the
> Program or any portion of it" (section 2) while the GPL document does not
> allow any modification, under any provision: "but changing it is not allowed".

You continue to ignore the big important fact that:

 - section 2 is _conditional_ on section 1
 - section 1 (and FSF guidelines) _requires_ you to leave copyright 
   notices intact and give the license out along with the program.

IOW, your claim that the GPL requires you to be able to make changes is 
INCORRECT.

The GPL requires you to be able to make a certain _class_ of changes, but 
other changes (like modifying the copyright license of notices or removing 
the license) are expressly _forbidden_ both by the GPL and by copyright 
law.

Think about it. The GPL is _not_ incompatible with including the GNU 
General Public License as part of the program. 

This can, btw, also be shown independently by the fact that the FSF 
clearly _intended_ the license to be actively linked into the program: 
they ask you (in the "How to Apply These Terms to Your New Programs") to 
have commands to view parts of the license if your program is interactive.

In other words, your claim that the license file itself is somehow 
"incompatible" with being included as part of the the GPL'd Program is 
simply not possible.

Btw, try this:

	shell$ gdb

	(gdb) show copying

and if you strace it (or do "strings" on the binary), you'll even notice 
that the license is very much part of the binary image itself. How much 
more "part of the program" can it be?

So if you seriously suggest that the GNU General Public License file is 
incompatible with the GPL, then that ignores the fact that

 - the license file itself assks you to combine it with the program
   (much more tightly than "mere aggregation")

 - the GPL license itself has the requirement that copyright licenses be 
   left intact (actually, the license itself seems to say just keep intact 
   all notices that "refer to this License", but the FSF has made it clear 
   that it covers _any_ copyright license notes, and that the GPL is 
   compatible with the BSD license which requires the same retention of 
   copyright notices).

or alternatively:

 - the FSF actively _encourages_ people to violate their own copyright 
   (which means that they can't sue, under the doctine of estoppel), and 
   your point is moot _anyway_.

So. Claiming that the GPL license text itself cannot be part of the 
program is disingenious. According to your reading, the modified BSD 
license wouldn't be compatible with the GPL either, because it requires:

 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions, and the following disclaimer,
 *    without modification.

yet the FSF has clearly stated that this is perfectly fine, even though it 
also disallows modifications to the license text.

Quod erat demonstrandum.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Rene Herman]

Linus Torvalds wrote:

> You continue to ignore the big important fact that:
> 
>  - section 2 is _conditional_ on section 1

I'm not ignoring this -- I really just don't see it to be relevant. All 
sections have to be obeyed the same, and in this sense all sections are 
conditional on all others. Violate one, and the license is not available 
to you. This goes without saying.

>  - section 1 (and FSF guidelines) _requires_ you to leave copyright 
>    notices intact and give the license out along with the program.
> 
> IOW, your claim that the GPL requires you to be able to make changes is 
> INCORRECT.

There are two things here that need seperating. On the one hand, we have 
"the program". On the other we have "the license" (and license notices). 
The GPL requires you to be able to make changes to the program, not the 
license. In the context of the Linux kernel, Linux is the program, and 
the GPLv2 is the license. In the context of the GPLv2, the GPLv2 is the 
program, and:

  Everyone is permitted to copy and distribute verbatim copies
  of this license document, but changing it is not allowed.

is the license. If you couldn't change just this bit (and the FSF 
copyright) then things would not conflict. But this very license says 
you can't change _the program_ (the license document). This conflicts.

> This can, btw, also be shown independently by the fact that the FSF 
> clearly _intended_ the license to be actively linked into the
> program: they ask you (in the "How to Apply These Terms to Your New
> Programs") to have commands to view parts of the license if your
> program is interactive.

This, in fact, seems to be a good point. This one wants an FSF lawyer.

> So. Claiming that the GPL license text itself cannot be part of the 
> program is disingenious. According to your reading, the modified BSD 
> license wouldn't be compatible with the GPL either, because it requires:
> 
>  * 1. Redistributions of source code must retain the above copyright
>  *    notice, this list of conditions, and the following disclaimer,
>  *    without modification.
> 
> yet the FSF has clearly stated that this is perfectly fine, even though it 
> also disallows modifications to the license text.

But please note that this is indeed perfectly fine, and does not 
contradict anything I say. This requirement only disallows modifications 
  to the copright and license notices, not modifications of the 
_program_ and the GPL is perfectly fine with that.

This is the point I've tried to make a number of times now. The program 
is not the same thing as the license and allowing or disallowing 
something being done to the license is not the same thing as allowing or 
disallowing something to be done to the program. And in the case of the 
GPL license text, the license text _is_ the program. Not the license.

Note, we are arguing here over whether or not the GPL document itself is 
GPL compatible. I feel it obvious that it is not, you do not agree. What 
we _do_ agree on though is that it's wholy irrelevant for the kernel at 
least since 2.4.0-test8.

Ever since then, the kernel as a whole has been V2, not a doubt about 
it. You feel that even before that, it was V2 only but then you are 
ignoring the very strong point Paul Jakma made and which I repeated: if 
the "Version 2" in the GPL header were enough to have the program 
specify a version, section 9 would be utterly useless, and as such it's 
obvious that at very least the intent of the GPL authors here was that 
it was _not_ enough.

Hey, you still replied, so you probably don't think I'm completely full 
of it. Did you think there was any merit in the suggestion of seperating 
the GPL and the NOTE into separate files, so as to leave even fewer room 
for people trying to milk the argument?

Rene.

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Thu, 2 Feb 2006, Rene Herman wrote:
> > So. Claiming that the GPL license text itself cannot be part of the program
> > is disingenious. According to your reading, the modified BSD license
> > wouldn't be compatible with the GPL either, because it requires:
> > 
> >  * 1. Redistributions of source code must retain the above copyright
> >  *    notice, this list of conditions, and the following disclaimer,
> >  *    without modification.
> > 
> > yet the FSF has clearly stated that this is perfectly fine, even though it
> > also disallows modifications to the license text.
> 
> But please note that this is indeed perfectly fine, and does not contradict
> anything I say. This requirement only disallows modifications  to the copright
> and license notices, not modifications of the _program_ and the GPL is
> perfectly fine with that.

You continue to make this totally arbitrary distinction between "copyright 
and license notices" and the "program".

That distinction does not make any sense.

The "license notices" _are_ an integral part of the program. Nothing else 
makes sense. They are often in the same files (that's the most common case 
by far), and as I showed you, they are often (as in the case of the GPL 
license notice itself) actually _linked_in_ into the actual binary itself.

I find absolutely zero legal ground for your distinction that they are 
somehow a totally independent entity from "the program", and it is in 
fact in total disagreement with GPLv2 section 9 where they talk about the 
"Program" specifying a license version.

You seem to be saying that no copyright license or notice is ever part of 
the "Program", but that's totally idiotic. It would mean that you can 
_never_ specify a "license version" in the Program, because by your 
totally inconsistent rules, that copyright license note would 
automatically somehow magically not be part of "Program".

So your reading simply DOES NOT MAKE SENSE.

I claim that Copyright notices and licenses very much _are_ part of the 
program. I claim that because (a) the license says nothing to the contrary 
and (b) the license clearly _does_ talk of copyright notices within the 
Program (since the "Program" can specify a license version as per section 
9: how the hell could you specify a license version "in" the Program, if 
all copyright license notices were somehow magically "outside" the 
Program?)

You cannot have it both ways. You can't say that copyright license notes 
are sometimes outside the Program, and sometimes part of the Program.

The _only_ consistent reading is that copyright notices are _always_ part 
of the Program, and that you must leave them intact as per section 1. This 
also makes it clear that the "you must be able to modify" (section 2) does 
_not_ cover copyright notices and licenses, even though they _are_ part of 
the Program.

See? My reading is perfectly self-consistent, and doesn't need any 
artificial separation of "copyright notice" vs "Program". Yours is not. 

Your notion of copyright notes and licenses somehow being separate from 
"Program" simply cannot be reconciled with reality, common sense _or_ the 
wording in section 9.

The fact is, the copyright notices and licenses really _are_ a very 
integral part of "Program". You cannot separate one from the other - not 
conceptually, and not physically - and trying to do so would in fact be a 
clear violation of section 1.

			Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

> The "license notices" _are_ an integral part of the program. Nothing else 
> makes sense. They are often in the same files (that's the most common case 
> by far), and as I showed you, they are often (as in the case of the GPL 
> license notice itself) actually _linked_in_ into the actual binary itself.

The COPYING file is not part of the program. I'm sorry you can't get
that into your brain but continually bleating that it doesn't make sense
won't actually get you anywhere. Please go and talk to a law student or
just let the subject drop.

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> The fact is, the copyright notices and licenses really _are_ a very
> integral part of "Program". You cannot separate one from the other - not
> conceptually, and not physically - and trying to do so would in fact be a
> clear violation of section 1.

	Is it your position that a program distribution that includes a copy of
GPLv2 and contains no other information about licensing is licensed *only*
under GPLv2 and you can not substitute a later version at your option?

	It may help to review section 9:

"Each version is given a distinguishing version number.  If the Program
specifies a version number of this License which applies to it and "any
later version", you have the option of following the terms and conditions
either of that version or of any later version published by the Free
Software Foundation.  If the Program does not specify a version number of
this License, you may choose any version ever published by the Free Software
Foundation."

	Do you see in here *any* reference to a way to limit a program so that only
one particular version of the GPL applies to it?

	In fact, the kernel's notice that you may only use v2 and no later version
of the GPL appears, to me, to be an additional restriction that is not
permitted by the GPL. That is, unless you can find a GPL section that
permits that type of restriction. I was unable to find one. None of the
examples included with the GPL include any such restriction of that type.

	I previously wrote that I thought that you could do so by including a
notice to that effect in every file. I now do not believe that is the case.
You cannot include any other type of restriction in a file under the GPL,
and I don't see why that type of restriction would be an exception.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Rene Herman]

Linus Torvalds wrote:

> That distinction does not make any sense.

I'm afraid I have by now run out of steam. I you still feel the subject 
interesting, please consult a lawyer.

Rene.

Re: GPL V3 and Linux - Dead Copyright Holders

[Andrew James Wade]

On Wednesday 01 February 2006 18:29, Rene Herman wrote:
> There are two things here that need seperating. On the one hand, we have 
> "the program". On the other we have "the license" (and license notices). 
> The GPL requires you to be able to make changes to the program,

... Subject to the terms of Section 1. It says so explicitly: 
"You may ... under the terms of Section 1 above, provided that you also ..."
The GPL requires you to be able to make changes to the program _except_
for changes that would not "keep intact all the notices that refer to this
License" (+ a couple of other exceptions). Editing the text of the GPL in
the COPYING file would certainly violate this. However, copying COPYING
to a new file (license_for_foo, say) and editing that copy wouldn't. The
GPL allows this, the GPL's license does not, ergo, in the final analysis
I think you are right that the GPL's license is not literally
GPL-compatible.

> > This can, btw, also be shown independently by the fact that the FSF 
> > clearly _intended_ the license to be actively linked into the
> > program: they ask you (in the "How to Apply These Terms to Your New
> > Programs") to have commands to view parts of the license if your
> > program is interactive.
> 
> This, in fact, seems to be a good point. This one wants an FSF lawyer.

There is a very strong implication there that the GPL can be incorporated
into GPLed code. I'd read it as allowing the GPL v 2 to be included in
GPLed source code notwithstanding Section 2. It'd be nice if that were
explicit and part of the Terms and Conditions, but the verbiage is part of
the License, and I suspect counts for something. But IANAL.

As a practical matter, even if the GPL is technically GPL-incompatible,
the chances of anyone objecting to their GPLed code rubbing shoulders
with the GPL is remote.

Andrew Wade

Re: GPL V3 and Linux - Dead Copyright Holders

[Kyle Moffett]

On Feb 02, 2006, at 01:39, Andrew James Wade wrote:
> ... Subject to the terms of Section 1. It says so explicitly:
> "You may ... under the terms of Section 1 above, provided that you  
> also ..." The GPL requires you to be able to make changes to the  
> program _except_ for changes that would not "keep intact all the  
> notices that refer to this License" (+ a couple of other exceptions).

But see, even assuming the really odd case of a project consisting of  
one file, the GPL, that project would be completely GPL compatible.   
As the license specifies the licensing terms for the project (IE: the  
GPL), it may not legally be modified _even_ _under_ _copyright_ _law_  
(because it's the project license).  As a result, it is that the GPL  
document may also be GPL licensed (because the only restrictions  
therein are automatically implied by copyright law in the first  
place).  The FSF also mentions in their "deriving from the GPL" page  
that you may create any derivation you want, but you may not  
reference the GPL or FSF in such a revised license.  That doesn't  
affect or infringe on the GPL either, because such misuse would be  
trademark infringement (outside the GPL) and not copyright  
infringement (covered by the GPL), and therefore still doesn't affect  
the terms of the GPL itself.

Cheers,
Kyle Moffett

--
Somone asked me why I work on this free (http://www.fsf.org/ 
philosophy/) software stuff and not get a real job. Charles Schulz  
had the best answer:

"Why do musicians compose symphonies and poets write poems? They do  
it because life wouldn't have any meaning for them if they didn't.  
That's why I draw cartoons. It's my life."
   -- Charles Schulz

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

	Kyle Moffet wrote:

> But see, even assuming the really odd case of a project consisting of
> one file, the GPL, that project would be completely GPL compatible.
> As the license specifies the licensing terms for the project (IE: the
> GPL), it may not legally be modified _even_ _under_ _copyright_ _law_
> (because it's the project license).  As a result, it is that the GPL
> document may also be GPL licensed (because the only restrictions
> therein are automatically implied by copyright law in the first
> place).

	That's just not true. There is no reason under copyright law why the author
of a program could not modify its license. However, the GPL explicitly
prohibits *anyone* from modifying it.

	The reason you can't modify the GPL, even we assume the GPL is licensed
under the GPL, is because the GPL says you can't modify the GPL.

	Andrew Wade wrote:

> As a practical matter, even if the GPL is technically GPL-incompatible,
> the chances of anyone objecting to their GPLed code rubbing shoulders
> with the GPL is remote.

	It is logically impossible for the GPL to be GPL-incompatible. To be
GPL-incomptabile, a license would have to contain requirements or
restrictions not found in the GPL. How could the GPL possibly do that?

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Andrew James Wade]

On Thursday 02 February 2006 02:11, David Schwartz wrote:
> 	The reason you can't modify the GPL, even we assume the GPL is licensed
> under the GPL, is because the GPL says you can't modify the GPL.
     So you're saying that the two lines that are the GPL's copyright
license, are also part of the terms and conditions of the GPL, and act in
that capacity as a restriction of what you can do under Section 2? That's
an interesting argument, I might agree.

> 	It is logically impossible for the GPL to be GPL-incompatible. To be
> GPL-incomptabile, a license would have to contain requirements or
> restrictions not found in the GPL. How could the GPL possibly do that?
     It's not the GPL doing that, it's the GPL's license. But perhaps the
restrictions of the license are also restrictions of the GPL, they are part
of the text of that document after all.

Well enough pedantism for me.

Andrew Wade

     

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> On Thursday 02 February 2006 02:11, David Schwartz wrote:

> > 	The reason you can't modify the GPL, even we assume the GPL
> > is licensed
> > under the GPL, is because the GPL says you can't modify the GPL.

>      So you're saying that the two lines that are the GPL's copyright
> license, are also part of the terms and conditions of the GPL, and act in
> that capacity as a restriction of what you can do under Section 2? That's
> an interesting argument, I might agree.

	Yes, the GPL is the license agreement for both the GPL and anything
licensed under the GPL.

> > 	It is logically impossible for the GPL to be GPL-incompatible. To be
> > GPL-incomptabile, a license would have to contain requirements or
> > restrictions not found in the GPL. How could the GPL possibly do that?

>      It's not the GPL doing that, it's the GPL's license. But perhaps the
> restrictions of the license are also restrictions of the GPL,
> they are part
> of the text of that document after all.

	I have no idea what you mean by the "GPL's license". If it's contained in
the GPL, it's part of the GPL. The GPL is all of a piece.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Andrew James Wade]

On Friday 03 February 2006 00:35, David Schwartz wrote:
> 	I have no idea what you mean by the "GPL's license". If it's contained in
> the GPL, it's part of the GPL. The GPL is all of a piece.

 "Everyone is permitted to copy and distribute verbatim copies
 of this license document, but changing it is not allowed."

Those two lines look like a copyright license for the GPL. It makes sense
for the GPL to be licensed under the GPL, but it's less than obvious to
me that that is actually the case.

Andrew Wade

Re: GPL V3 and Linux - Dead Copyright Holders

[Andrew James Wade]

On Thursday 02 February 2006 01:57, Kyle Moffett wrote:
> But see, even assuming the really odd case of a project consisting of  
> one file, the GPL, that project would be completely GPL compatible.   
> As the license specifies the licensing terms for the project (IE: the  
> GPL), it may not legally be modified _even_ _under_ _copyright_ _law_  
> (because it's the project license).

Well ok, but modified copies of it (that aren't the license) could be
made. For the sake of argument, imaging copying the preamble into
another file and adding a few paragraphs there, with a prominent notice of
your modifications. This is allowed by the GPL (the License is the same,
and you're not violating the other terms either as far as I can see). But
it is not allowed by the license for the text of the GPL. Hence the
problem. It's an odd squirrelly corner case of the sort that cause licenses
to bloat up with verbiage. 

The out is if the license file is not a portion of the program, in which
case section 2 of the GPL doesn't apply. I think this is what Alan Cox
is arguing. However, while it would be convenient if the License text
wasn't considered part of the program, I'm not sure that reasoning
would hold up in court.

It's also a complete non-issue: both because no-one is going to actually
object to the text of the GPL being there, and because it can be replaced
with a simple link in the extremely unlikely event that the need arises.

Andrew Wade

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

> Well ok, but modified copies of it (that aren't the license) could be
> made. For the sake of argument, imaging copying the preamble into
> another file and adding a few paragraphs there, with a prominent notice of
> your modifications. This is allowed by the GPL (the License is the same,
> and you're not violating the other terms either as far as I can see). But
> it is not allowed by the license for the text of the GPL. Hence the
> problem. It's an odd squirrelly corner case of the sort that
> cause licenses
> to bloat up with verbiage.

	I totally don't understand why anyone thinks this is a problem. The GPL
restricts copying and modification in various ways. One of them is that you
can't modify the GPL itself. Another is that you can't remove license
notices or disclaimers of liability.

	There is no problem and there is no conflict. The GPL is licensed under the
GPL, it's self-licensing. The GPL specifies the terms of use for the works
covered by it, including the GPL itself.

	The GPL is not GPL-incompatible because something can only be
GPL-incompatible if it contains restrictions not found in the GPL. It is
logically impossible for the GPL to do so.

	DS

RE: GPL V3 and Linux - Dead Copyright Holders

[David Schwartz]

Linus Torvalds wrote:

> but the fact is, the GPL also says that any license notices must be kept
> intact, and that a copy of the GPL itself must be given along with the
> program (in section 1).

	The GPL says you must keep intact notices in the Program that refer to the
license. That logically cannot include the license itself. For one thing,
the license is not in the program. For another thing, the license does not
refer to itself.

> I'm convinced _that_ is how you get "no version specified" in section 9.
> You have a program that just says "This is licensed under the GPL",
> instead of doing the full thing.

	You are essentially arguing here that the FSF screwed up. They intended the
normal case to be that you got a copy of the GPL with the program but could
substitute a later version at your option. Why would they have made it
almost impossible (or even extra-difficult) to get the result that they
preferred?

	A much more logical reading is that the License and the Program are two
distinct things. When it says "notices in the Program", that's precisely
what it means, notices that are in the Program. When it says "the Program
specifies a version number", that's also what it means, a version number
specified by the Program.

	Not only does this make it easy to get the result the FSF preferred, but ot
is supported by the License's internal definition of the term "Program".

	I don't think you can lock a person down to a single version of the GPL
unless you explicitly say so in each individual file. Otherwise, nothing
prevents someone from separating out one file and distributing it, kicking
in clause 9's rule about a Program that does not specify a version number.
The requirement to keep license notices intact does not mean you must add
them or maintain them. Just that you can't remove or modify them.

	It has always been well-understood on this mailing list that a file can be
excerpted from a GPL project and inserted into another one. Surely the FSF
didn't intend to make the default position be that GPL'd projects would be
mutually incompatible.

	DS

Re: GPL V3 and Linux - Dead Copyright Holders

[Krzysztof Halasa]

"David Schwartz" <davids@webmaster.com> writes:

> 	I don't think you can lock a person down to a single version of the GPL
> unless you explicitly say so in each individual file. Otherwise, nothing
> prevents someone from separating out one file and distributing it, kicking
> in clause 9's rule about a Program that does not specify a version number.

Nope. If you don't have a licence to use that single file, you can't
use it (distribute etc.) at all. So you basically choose between that
single version of GPL or nothing.
-- 
Krzysztof Halasa

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Maw, 2006-01-31 at 12:38 -0800, Linus Torvalds wrote:
> Alan argues that that extra notice "changed" the license (and that any 
> code that is older than 5 years would somehow not be GPLv2). I argue 
> otherwise. I argue that for the whole history, Linux has been v2-only 
> unless otherwise explicitly specified.
> 
> And I don't think even Alan will argue that the "v2 only" thing hasn't 
> been true for the last five years.

I would argue its irrelevance

Two cases (lets call them a and b)

	a) The GPLv2 only was always the case
	b) There was no version so it was open to choice

Which ultimately means either

	a) Linus changed nothing
	b) Linus chose a version as the License allowed him to in accordance
with section 9.

So we have two legal outcomes both of which produce the right answer for
any vaguely recent source tree. At which point does it matter ?

My point was to make clear that assuming the GPL original text implies
the version of the code is wrong, and explain why the FSF recommend the
long text.

Is there doubt about the license status of the current code - not in
this area, no. The COPYING file is extremely clear on this, and more
importantly in other possible unclear and problematic areas. For example
the statement that the system calls are not derivative statement which
resolves the biggest interpretation concern of all.

Alan

Re: GPL V3 and Linux - Dead Copyright Holders

[Linus Torvalds]

On Tue, 31 Jan 2006, Alan Cox wrote:
> 
> I would argue its irrelevance

Well, I can absolutely agree with that.

It really wasn't ever meant to _change_ anything, so it definitely should 
be irrelevant in that sense. 

That said, I think it actually hass one big result: we may be discussing 
this for a couple of weeks, but I'm pretty sure we won't be discussing it 
for months and having a huge split over it in the kernel community. 

Which to me makes it less-than-irrelevant ;)

I actually tend to have two very different kinds of worries:

 - the technical problem of the month. Something doesn't work, and we 
   don't know why, and it's been broken for long enough that it seems to 
   be pretty fundamental.

   These things worry me, but they don't keep me up at night. 

 - the worry that somebody who submitted code to the kernel feels that his 
   code is mis-used and feels let down by the process.

   This thing _worries_ me. This is something I end up losing sleep over.

The second case is unusual, but it does happen. Things like the 
MMX-optimized AES routines, where Jari Ruusu ended up objecting to his 
code getting used. I really don't think he had any _legal_ reasons to feel 
that way, but _that_ is what really really tends to make me unhappy: 
regardless of legal correctness, I want developers to feel _proud_ of what 
they did in Linux, not feel like their code is being trampled on.

Do unto others.. and so on.

This is also why I don't much like BSD->GPL conversion, and try to ask the 
original author to OK it (the way we did with the original AES assembly 
code authored by Dr Brian Gladman - just to make sure). And regardless, I 
want to keep dual licenses _actively_ dual-licensed in the kernel, just to 
respect the original wishes instead of just converting them to the 
(stricter) GPLv2.

So the reason I've spent a lot of time on this thread is basically that I 
worry about people who would _like_ to upgrade (and incorrectly _expected_ 
to upgrade) the whole kernel to GPLv3 being unhappy.

So I'm spending time on this thread trying to make sure that everybody 
realizes that GPLv2 was always the license of choice - people may still 
have wished for something else, but at least I can do my damned best to 
explain why things are how they are, and explaining that any expectations 
to the contrary really were misguided.

> Two cases (lets call them a and b)
> 
> 	a) The GPLv2 only was always the case
> 	b) There was no version so it was open to choice
> 
> Which ultimately means either
> 
> 	a) Linus changed nothing
> 	b) Linus chose a version as the License allowed him to in accordance
> with section 9.
> 
> So we have two legal outcomes both of which produce the right answer for
> any vaguely recent source tree. At which point does it matter ?

Well, it does matter if somebody wants to start anew at the state we were 
at five years ago in the belief that that old state is GPLv3-compatible, 
and then add in the more modern part of the files that are explicitly 
compatible with GPLv3...

And if somebody feels _that_ strongly about it, and feels that five years 
ago there really was a real ambiguity, I'd be interested to hear his or 
her arguments. And I don't see myself likely objecting to it, if only 
because I'd be intrigued to see how far they get ;)

> Is there doubt about the license status of the current code - not in
> this area, no. The COPYING file is extremely clear on this, and more
> importantly in other possible unclear and problematic areas. For example
> the statement that the system calls are not derivative statement which
> resolves the biggest interpretation concern of all.

I was always surprised by how anybody could _possibly_ worry about that 
one, but it did come up very early. I forget exactly when, but I think 
that clarification was added way before Linux-1.0.

In general, people worry too much. And I worry about other people 
worrying.

Hopefully I also worry without any real cause.

		Linus

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Maw, 2006-01-31 at 16:18 -0800, Linus Torvalds wrote:
> That said, I think it actually hass one big result: we may be discussing 
> this for a couple of weeks, but I'm pretty sure we won't be discussing it 
> for months and having a huge split over it in the kernel community. 

Agreed

> Do unto others.. and so on.

This is how you spot a lawyer from a normal human. The human agrees but
the lawyer says "but what about my dentist, will he really want me to
drill his teeth" ;)

> So I'm spending time on this thread trying to make sure that everybody 
> realizes that GPLv2 was always the license of choice - people may still 

The intent is clear by any view for current code, and the past is for
argument but equally its unfixable whatever the result. 

> I was always surprised by how anybody could _possibly_ worry about that 
> one, but it did come up very early. I forget exactly when, but I think 
> that clarification was added way before Linux-1.0.
> 
> In general, people worry too much. 

Its the job of some of them to worry. After all many of them have
shareholders to answer for, and whether they worry about misusing your
pension fund or whether they'll still get a new yacht if there is a
lawsuit, they worry.

I've actually talked to a number of people for whom the syscall
clarification was critical to their choice to produce software for Linux
so don't underestimate your smart planning in advance ;)

Alan

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Linus,

For what it's worth, maybe you should run some of these issues past the 
Electronic Frontier Foundation (eff.org) or Robin Gross at IPJustice.org 
who might be able to review these issues and help make sure you're doing 
it right.

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Marc Perkel wrote:

> Linus,
>
> For what it's worth, maybe you should run some of these issues past 
> the Electronic Frontier Foundation (eff.org) or Robin Gross at 
> IPJustice.org who might be able to review these issues and help make 
> sure you're doing it right.


Great advice, however, the EFF are not experts in IP law.  I would 
suggest a good firm in Palo Alto, like Rosati, Goodrich and Sonsini, 
they are experts
in corporate IP law and Corporate law in general.  They would know all 
the loopholes people and companies are likely to use.  Stay away from these
freebie Law Groups who are more likely to kiss your butt and tell you 
waht you want to hear rather than what you NEED to hear. 

OSDL should pick up the tab.  It will cost around 30K for a good 
analysis from them.

Jeff

>
> -
> To unsubscribe from this list: send the line "unsubscribe 
> linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Jeff V. Merkey wrote:
> Marc Perkel wrote:
>
>> Linus,
>>
>> For what it's worth, maybe you should run some of these issues past 
>> the Electronic Frontier Foundation (eff.org) or Robin Gross at 
>> IPJustice.org who might be able to review these issues and help make 
>> sure you're doing it right.
>
>
> Great advice, however, the EFF are not experts in IP law.  I would 
> suggest a good firm in Palo Alto, like Rosati, Goodrich and Sonsini, 
> they are experts
> in corporate IP law and Corporate law in general.  They would know all 
> the loopholes people and companies are likely to use.  Stay away from 
> these
> freebie Law Groups who are more likely to kiss your butt and tell you 
> waht you want to hear rather than what you NEED to hear.
> OSDL should pick up the tab.  It will cost around 30K for a good 
> analysis from them.
>
> Jeff
>

Actually EFF does have some good IP lawyers on staff and Robin Gross - a 
former EFF employee (robin@ipjustice.org) is also very up on this. And 
the prices is right too if you don't want to spend the big bucks. I 
would rather get legal advice from lawyer who care about GPL than 
lawyers who just want to run up a big fee and screw you in the end.

It might be worth looking into doing it right before some other SCO type 
decides they can get rich by busting Linux.

Re: GPL V3 and Linux - Dead Copyright Holders

[Jeff V. Merkey]

Marc Perkel wrote:

>
>
> Jeff V. Merkey wrote:
>
>> Marc Perkel wrote:
>>
>>> Linus,
>>>
>>> For what it's worth, maybe you should run some of these issues past 
>>> the Electronic Frontier Foundation (eff.org) or Robin Gross at 
>>> IPJustice.org who might be able to review these issues and help make 
>>> sure you're doing it right.
>>
>>
>>
>> Great advice, however, the EFF are not experts in IP law.  I would 
>> suggest a good firm in Palo Alto, like Rosati, Goodrich and Sonsini, 
>> they are experts
>> in corporate IP law and Corporate law in general.  They would know 
>> all the loopholes people and companies are likely to use.  Stay away 
>> from these
>> freebie Law Groups who are more likely to kiss your butt and tell you 
>> waht you want to hear rather than what you NEED to hear.
>> OSDL should pick up the tab.  It will cost around 30K for a good 
>> analysis from them.
>>
>> Jeff
>>
>
> Actually EFF does have some good IP lawyers on staff and Robin Gross - 
> a former EFF employee (robin@ipjustice.org) is also very up on this. 
> And the prices is right too if you don't want to spend the big bucks. 
> I would rather get legal advice from lawyer who care about GPL than 
> lawyers who just want to run up a big fee and screw you in the end.

No.  I think tey would salivate to be able to work on Linux stuff with 
Linus.  Wilson Sonsini, who was on the Novell Board at the time, directed
the lawsuit against me at TRG -- he nailed me no matter what squirming I 
did.  I have had other dealings with them.  They would do a great job
for Linus.

>
> It might be worth looking into doing it right before some other SCO 
> type decides they can get rich by busting Linux.
>
>
Not likely.  Our building is right next to them and over the past few 
months, there's fewer and fewer cars in the parking lot and more lights
turned off in offices in the evening.  I doubt they are going to do 
anything to Linux.  IBM is going to get nailed on some stuff, but I think
Linux is unaffected, at least those parts RedHat, Suse, and the rest of 
us need to keep going.

Jeff

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Maw, 2006-01-31 at 11:07 -0800, Linus Torvalds wrote: 
> granted to you, so nobody has _ever_ had rights to re-license Linux under 
> any other license than the one it came with: namely the GPLv2. Alan is 
> trying to argue that the fact that it has been licensed under the GPLv2 
> would somehow "magically" mean that it has been licensed under any version 
> of GPL that you can pick, BUT THAT IS AN OBVIOUSLY LEGALLY FLAWED 
> ARGUMENT.

Clause 9 is clear and is part of the GPL v2. The GPL v2 text gives
people (including you) that right itself if no version is specified in
the program. You and the other authors granted that right if you didn't
include a statement about version. Now as it happens various
contributors specified versions long ago and you clarified it too.

> Alan - talk to a lawyer. 

Actually I did, long ago before this argument even appeared, because it
was important for another situation.

Also please get one thing straight. I'm not arguing for a license
change, I'm pointing out misunderstandings that might lead people,
particularly other GPL projects to make mistakes.

The big problem with any license change is actually the moral one, as
I'm sure you'd agree: Do you have the moral right to change the rules
(eg on DRM) when many have contributed with many differing views and in
many ways not all of them leading to them being copyright holders ?

Alan

Re: GPL V3 and Linux - Dead Copyright Holders

[Rogier Wolff]

On Tue, Jan 31, 2006 at 11:57:31PM +0000, Alan Cox wrote:
> Clause 9 is clear and is part of the GPL v2. The GPL v2 text gives
> people (including you) that right itself if no version is specified in
> the program. You and the other authors granted that right if you didn't
> include a statement about version. Now as it happens various
> contributors specified versions long ago and you clarified it too.

Alan, Linus, 

I have indicated my wish to have my code GPL-2 only in 2000. Linus, 
declined to change the files. My intention has been for the
past 5 and a half years to have my code GPL-2  only, only the 
files don't reflect that change yet. 

REW wrote on Fri, 8 Sep 2000 23:17:09 +0200 (MEST):
> Linus, 

> I  trust YOU on this. Could you change 

>  *      modify it under the terms of the GNU General Public License as
>  *      published by the Free Software Foundation; either version 2 of
>  *      the License, or (at your option) any later version.

> into 

>  *      modify it under the terms of the GNU General Public License 
>  *      version 2 as published by the Free Software Foundation.

> for all the files that I own copyright on?


Linus answered on: Sat, 9 Sep 2000 09:29:01 +0200 (MEST)
> Thanks. For the moment, I'd rather like to not do any extra changes, and
> see what the reaction from the FSF and other people is - let's not hurry
> into this any faster than we have to. I wanted to make a public statement
> because I thought it needed to be done, but at the same time I do not want
> to make a rift between Linux and the FSF bigger than necessary.
> 
> So let this lie for a while, and see what happens, ok?

	Roger. 

-- 
** R.E.Wolff@BitWizard.nl ** http://www.BitWizard.nl/ ** +31-15-2600998 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
Q: It doesn't work. A: Look buddy, doesn't work is an ambiguous statement. 
Does it sit on the couch all day? Is it unemployed? Please be specific! 
Define 'it' and what it isn't doing. --------- Adapted from lxrbot FAQ

Re: GPL V3 -- PLEA FOR SANITY

[Theodore Ts'o]

Let's give this thread a rest, OK?

Whether or not Linux is licensed under the GPLv2 only or not is
ultimately a matter for the lawyers.

The one big problem I see with the GPLv3 effort is given the
additional restrictions regarding DRM, it doesn't seem to clear to me
whether a project which has even a single line of GPLv2-only code can
accept GPLv3 code. That is, GPLv3 is designed to be compatible with
more licenses, but that doesn't matter of GPLv2 isn't compatible with
GPLv3.  If that is the case, if only a _single_ person (like Rogier)
has I want GPLv2-only, the whole project is can't use any GPLv3 code
unless they are willing to track down and rewrite all of the code
written by that person or persons.

If that is true (and again, at the end of the day lawyers or more
importantly, a judge is going to have to make that call, not debating
technologists) it's hard to see the GPLv3 making any headway.

						- Ted

Re: GPL V3 and Linux - Dead Copyright Holders

[Diego Calleja]

El Tue, 31 Jan 2006 09:57:52 -0800 (PST),
Linus Torvalds <torvalds@osdl.org> escribió:

> Please realize that the kernel is one of the largest open-source projects 
> in the world, if not _the_ largest. Most projects are much much smaller. 

Obviously you've never downloaded openoffice source code :)

My kernel sources weight 259 MB after cleaning it, but according to:
http://download.openoffice.org/2.0.0/source.html openoffice 2.0
weights 260.

Openoffice is much better at bloating code, you can't win.

Re: GPL V3 and Linux - Dead Copyright Holders

[Josh Boyer]

On 1/31/06, Diego Calleja <diegocg@gmail.com> wrote:
> El Tue, 31 Jan 2006 09:57:52 -0800 (PST),
> Linus Torvalds <torvalds@osdl.org> escribió:
>
> > Please realize that the kernel is one of the largest open-source projects
> > in the world, if not _the_ largest. Most projects are much much smaller.
>
> Obviously you've never downloaded openoffice source code :)
>
> My kernel sources weight 259 MB after cleaning it, but according to:
> http://download.openoffice.org/2.0.0/source.html openoffice 2.0
> weights 260.
>
> Openoffice is much better at bloating code, you can't win.

It was more a comment on the number of contributors.  Not download size.

josh

Re: GPL V3 and Linux - Dead Copyright Holders

[Alan Cox]

On Maw, 2006-01-31 at 09:57 -0800, Linus Torvalds wrote:
> The fact that the COPYING file has a different copyright really doesn't 
> matter. It's still part of the release. 

Law is about precision and exact wording as well as intent. The exact
wording is "the Program" not "the release". And Program is capitalised
to indicate the use of the definition made earlier. That is: "The
"Program", below, refers to any such program or work, and a "work based
on the Program" means either the Program or any derivative work under
copyright law"

The COPYING file is not derivative and is not by any usual
interpretation part of the program. As I said and showed earlier the
COPYING file cannot be part of the program without contradiction.

Ask a law student if you don't believe me, or do it in equations not
words... 

> It's absolutely not different from having a separate "Release notes" file 
> which specifies the copyright conditions. That's how Linux-0.01 did it: 
> the thing was outside the actual main tar-ball, and sent out both as part 
> of the announcemnt and as a separate file in the same directory on the 
> ftp-site.

The release notes are part of the program. The section starting "NOTE!"
is maybe part of the program (thats a real lawyer question). However
whether it is or not is such clear intent that nobody would win an
argument that you had not said nowdays that it is V2

> I can make very specific arguments for why version 2 ONLY is the specific 
> license that covers Linux. In contrast you can only make weasel-wording 

Well if you consider accurate interpretation of language 'weasel
wording' I can only feel sorry for you and be glad you write code not
English. But since I don't actually want to change the license on the
kernel and I already can if I wanted for the small percentage I wrote I
don't see the need to continue the debate.

Alan

Re: GPL V3 and Linux - Dead Copyright Holders

[Marc Perkel]

Alan Cox wrote:
> On Maw, 2006-01-31 at 09:57 -0800, Linus Torvalds wrote:
>   
>> The fact that the COPYING file has a different copyright really doesn't 
>> matter. It's still part of the release. 
>>     
>
> Law is about precision and exact wording as well as intent. The exact
> wording is "the Program" not "the release". And Program is capitalised
> to indicate the use of the definition made earlier. That is: "The
> "Program", below, refers to any such program or work, and a "work based
> on the Program" means either the Program or any derivative work under
> copyright law"
>
>   

Actually the law is more precisely about the intent of the people making 
the license/contract as opposed to the wording. Especially when there is 
fast moving technology involved. If in 1991 Linus released his software 
under GPL v2 and the wold changes in 15 years so that words mean things 
differently than they did then, or newer licenses come out to deal with 
new issues not contemplated back then, the meaning of the agreement is 
interpreted on the basis of the intent and not on the exact wording of 
the contract.

For example. I am a painter and I contract to paint your house white at 
123 main street and I show up on the job and your house is really 125 
main street and you change your mind and decide to have me paint it 
yellow and I do that, and then you say you don't owe me because that 
contract says you agreed to paint 123 main st white and you painted 125 
main street yellow - you lose. The intent of the contract was that I 
paint your house.

The meaning of the Linux license is what Linus intended it to be in 1991 
and the way it is used in the following years. People who contribute to 
the Kernel are presumed to understand the spirit of the project and if 
their license says something else but they contributed it to the Kernel 
then they are deamed to have waived their rights by contributing it to 
the Kernel and accepting the Kernel's license.

I have way too much legal experience for someone who isn't a lawyer, but 
if you don't believe me print this out and run it past a real lawyer and 
see if they don't agree. The reality is that the Kernel license it what 
Linus says it is and we all trust Linus as a condition for contributing 
to the Kernel.

Having said that, lawyers can argue anything so if the wording and 
reality were the same it makes it easier. So it is worth it to some 
extent to clear up issues if we can.

Re: GPL V3 and Linux - Dead Copyright Holders

[Kurt Wall]

On Wed, Jan 25, 2006 at 03:44:13PM -0500, linux-os (Dick Johnson) took 74 lines to write:
> 
> All of Linux has undergone such a conversion and it is
> effectively owned by the "Free Software Foundation, Inc."
> Of course RMS didn't tell you this when he appropriated
> it, but it's done.

WTF are you babbling about? Better yet, WTF are you smoking and may I
have some?

Kurt
-- 
If you give Congress a chance to vote on both sides of an issue, it
will always do it.
		-- Les Aspin, D., Wisconsin

Re: GPL V3 and Linux

[Alexander Shishckin]

On 1/20/06, Jeff V. Merkey <jmerkey@wolfmountaingroup.com> wrote:
> Cudos to Stallman, The patent retaliation clause is exactly what has
> been missing.  The inclusion of custom binaries was a little vague, but
> the net of it is that the end user can combine the separate parts, and
> have the freedom to do so given the GPL3 terms.  Any concensus on
> whether Linux will move to GPL3?   I support adoption and congrats to
> Stallman -- A++++.
GPLv3 tends to get on top of the most braindead things ever known to
software development. It is, in fact, a one-too-many example of how a
person who cannot be seriously considered to be a computer programmer
tries to have his one-too-many revenge on companies which employ real
software developers and produce real world software. Someone should
probably put an end to these miserable efforts.

--
I am free of all prejudices. I hate every one equally.

Re: GPL V3 and Linux

[Chase Venters]

On Friday 20 January 2006 20:27, Alexander Shishckin wrote:
> GPLv3 tends to get on top of the most braindead things ever known to
> software development. It is, in fact, a one-too-many example of how a
> person who cannot be seriously considered to be a computer programmer
> tries to have his one-too-many revenge on companies which employ real
> software developers and produce real world software. Someone should
> probably put an end to these miserable efforts.

Why does everyone assume that Stallman is out to 'get revenge' on companies? 
Is his desire for freedom so hard to grasp and believe that all you can do is 
spin it into silly conspiracies?

Why do people not recognize that his GNU project has built significant things? 
Do you not realize that Linux is licensed GPLv2, which is also Stallman's 
license?

I'm not going to trumpet around in 'patriotic' support of Stallman for too 
long, but if you're going to go on a Stallman/GPL bashing tirade, try having 
some real reasons instead of moaning like a rock in the wind.

As for the implicit allegation that he's wrong for not accepting the 
"company's" way of doing thing, last I checked, most of this 'free software' 
stuff was started and written by people as a hobby, for themselves and their 
users -- not for companies. It just happens that Stallman's license allows 
business and industry to harmonize.

> --
> I am free of all prejudices. I hate every one equally.

It would seem...

Re: GPL V3 and Linux

[Alexander Shishckin]

On 1/21/06, Chase Venters <chase.venters@clientec.com> wrote:
> On Friday 20 January 2006 20:27, Alexander Shishckin wrote:
> > GPLv3 tends to get on top of the most braindead things ever known to
> > software development. It is, in fact, a one-too-many example of how a
> > person who cannot be seriously considered to be a computer programmer
> > tries to have his one-too-many revenge on companies which employ real
> > software developers and produce real world software. Someone should
> > probably put an end to these miserable efforts.
>
> Why does everyone assume that Stallman is out to 'get revenge' on companies?
> Is his desire for freedom so hard to grasp and believe that all you can do is
> spin it into silly conspiracies?
Ain't that obvoius? Every second word that you read in GPLs is either
'freedom' or 'share' and the rest of the document has absolutely
nothing to do with both, just restricting our *freedom* to *share*.
This is not exactly the way these words are meant to be used.

> Why do people not recognize that his GNU project has built significant things?
> Do you not realize that Linux is licensed GPLv2, which is also Stallman's
> license?
Surely significant things/projects do have their origin there. This is
yet another sad consequence. I hope we can someday get rid of all this
'GNU is ...' crap and build a better world that understands 'freedom'
for what it really is, not just
I'll-sue-you-should-you-use-my-code-in-your-proprietary-product stinky
sort of thing.

> I'm not going to trumpet around in 'patriotic' support of Stallman for too
> long, but if you're going to go on a Stallman/GPL bashing tirade, try having
> some real reasons instead of moaning like a rock in the wind.
He's not a god, nor a prophet, just a human, face it.

> As for the implicit allegation that he's wrong for not accepting the
> "company's" way of doing thing, last I checked, most of this 'free software'
> stuff was started and written by people as a hobby, for themselves and their
> users -- not for companies. It just happens that Stallman's license allows
> business and industry to harmonize.
Not even close. Too much of RMS's bloody 'lectures/prophecies/you-name-it'
can do this to a person. So please be careful.

Re: GPL V3 and Linux

[Chase Venters]

On Friday 20 January 2006 23:56, Alexander Shishckin wrote:
> > Why does everyone assume that Stallman is out to 'get revenge' on
> > companies? Is his desire for freedom so hard to grasp and believe that
> > all you can do is spin it into silly conspiracies?
>
> Ain't that obvoius? Every second word that you read in GPLs is either
> 'freedom' or 'share' and the rest of the document has absolutely
> nothing to do with both, just restricting our *freedom* to *share*.
> This is not exactly the way these words are meant to be used.

Perhaps you are looking for BSD? This is Linux, and the GPL, and we use 
something called 'copyleft' that is designed to ensure everyone has freedom 
and no one can take it away.

> > Why do people not recognize that his GNU project has built significant
> > things? Do you not realize that Linux is licensed GPLv2, which is also
> > Stallman's license?
>
> Surely significant things/projects do have their origin there. This is
> yet another sad consequence. I hope we can someday get rid of all this
> 'GNU is ...' crap and build a better world that understands 'freedom'
> for what it really is, not just
> I'll-sue-you-should-you-use-my-code-in-your-proprietary-product stinky
> sort of thing.

Heh... I suspect you won't be spending much time on or around LKML if that's 
your attitude. There are a fair number of kernel developers that believe 
strongly enough in all this stuff that they would make legal threats / sue 
(rightfully so) over someone writing proprietary drivers.

> > I'm not going to trumpet around in 'patriotic' support of Stallman for
> > too long, but if you're going to go on a Stallman/GPL bashing tirade, try
> > having some real reasons instead of moaning like a rock in the wind.
>
> He's not a god, nor a prophet, just a human, face it.

Indeed, but at least he's provided the world with something of value.

> > As for the implicit allegation that he's wrong for not accepting the
> > "company's" way of doing thing, last I checked, most of this 'free
> > software' stuff was started and written by people as a hobby, for
> > themselves and their users -- not for companies. It just happens that
> > Stallman's license allows business and industry to harmonize.
>
> Not even close. Too much of RMS's bloody 'lectures/prophecies/you-name-it'
> can do this to a person. So please be careful.

What exactly were you saying "not even close" to? That free software is 
started and written by most as a hobby? That GPL is business friendly? 
Because if you deny it, you're simply WRONG.

I have an important question for you. If you don't like the license / 
philosophy / players behind this free software / open source thing, what the 
fuck are you doing in LKML? It's really obnoxious to come waltzing into a 
community and start telling people "Someone should
probably put an end to these miserable efforts."

In truth, I have some co-workers that feel like you do about copyleft and the 
GPL - moaning, groaning and bitching about it. The irony is that they would 
be in NO WAY affected by the GPL (it's a *license to distribute*, not a 
*contract for use*) unless they decided to make copies. More specifically, 
the reason they're always bitching is because they _very specifically_ want 
to TAKE and not GIVE BACK.

If you recall, Eric Raymond once said we don't need the GPL anymore. If I 
recall correctly, that was the last time Eric Raymond made the headlines. He 
rode out on a wave of vast disagreement from the community, and last I 
checked, no one has gone from the GPL back to public domain.

Re: GPL V3 and Linux

[Geert Uytterhoeven]

On Sat, 21 Jan 2006, Alexander Shishckin wrote:
> On 1/21/06, Chase Venters <chase.venters@clientec.com> wrote:
> > On Friday 20 January 2006 20:27, Alexander Shishckin wrote:
> > > GPLv3 tends to get on top of the most braindead things ever known to
> > > software development. It is, in fact, a one-too-many example of how a
> > > person who cannot be seriously considered to be a computer programmer
> > > tries to have his one-too-many revenge on companies which employ real
> > > software developers and produce real world software. Someone should
> > > probably put an end to these miserable efforts.
> >
> > Why does everyone assume that Stallman is out to 'get revenge' on companies?
> > Is his desire for freedom so hard to grasp and believe that all you can do is
> > spin it into silly conspiracies?
> Ain't that obvoius? Every second word that you read in GPLs is either
> 'freedom' or 'share' and the rest of the document has absolutely
> nothing to do with both, just restricting our *freedom* to *share*.

The problem with `just sharing' is that some people are only interested in the
`receiving' part of sharing, so unfortunately you need some extra legal stuff
in the license to make sure everyone plays fair.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: GPL V3 and Linux

[Horst von Brand]

Alexander Shishckin <alexander.shishckin@gmail.com> wrote:

[...]

> Ain't that obvoius? Every second word that you read in GPLs is either
> 'freedom' or 'share' and the rest of the document has absolutely nothing
> to do with both, just restricting our *freedom* to *share*.

How so? The existence of GNU doesn't restrict *my* right to share as *I*
wish. If I, freely, place my stuff under GPL it /does/ restrict other
people in just "sharing" (i.e., taking without giving in return). And that
is fine with me. Not with them, I presume...
-- 
Dr. Horst H. von Brand                   User #22616 counter.li.org
Departamento de Informatica                     Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria              +56 32 654239
Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513

Re: GPL V3 and Linux

[linux-os (Dick Johnson)]

On Sat, 21 Jan 2006, Horst von Brand wrote:

> Alexander Shishckin <alexander.shishckin@gmail.com> wrote:
>
> [...]
>
>> Ain't that obvoius? Every second word that you read in GPLs is either
>> 'freedom' or 'share' and the rest of the document has absolutely nothing
>> to do with both, just restricting our *freedom* to *share*.
>
> How so? The existence of GNU doesn't restrict *my* right to share as *I*
> wish. If I, freely, place my stuff under GPL it /does/ restrict other
> people in just "sharing" (i.e., taking without giving in return). And that
> is fine with me. Not with them, I presume...
> --
> Dr. Horst H. von Brand                   User #22616 counter.li.org
> Departamento de Informatica                     Fono: +56 32 654431
> Universidad Tecnica Federico Santa Maria              +56 32 654239
> Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513

The problem is that every rule and every law takes
away rights. Laws do not give rights. Rules do not
give rights. Amendments to existing laws sometimes
prevent the restriction of rights (like the first 10
amendments of the US Constitution), however there
are no rules or laws that ever, anywhere, provided
any rights whatsoever. Rules, regulations, and laws
are all about restricting rights.

Sometimes the restrictions are necessary. For instance,
except in very special circumstances, governments usually
take away the inherent rights to kill, etc.

The initial writer was correct. The GPL was supposed
to be all about freedom. Then, there are hundreds of
words that have nothing to do with freedom. They
establish rules. The crybaby says; "You will play
by my rules or..." Rules restrict freedom.

Perhaps these rules are necessary. However, for 20
years before the Internet even existed, people were
sharing source-code without rules. This was the
principle behind the PROGRAM EXCHANGE and other
obsolete BBS systems. At that time the ground-
work of most all the file-compression routines,
file-transmission routines, file-types, flight-
simulators, etc., the stuff now claimed by others,
was freely given away. Some expected their names
to remain in the source, but eventually their
names were changed to "Microsoft" or GPL. For
example, Phil Katz. He invented "zip" and gunzip
and all that stuff. He's now dead. His lifetime
of work has been stolen by others and claimed
as their own.

The Internet gets established and somebody who's
claim-to-fame was the development of the world's
most complicated word-processor, establishes some
legalese and a lot of well intentioned persons
fall into his trap as he claims that he developed
GNU/Linux as well. Wake up.


Cheers,
Dick Johnson
Penguin : Linux version 2.6.13.4 on an i686 machine (5589.54 BogoMips).
Warning : 98.36% of all statistics are fiction.
.

****************************************************************
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 DeliveryErrors@analogic.com - and destroy all copies of this information, including any attachments, without reading or disclosing them.

Thank you.

Re: GPL V3 and Linux

[Chase Venters]

On Monday 23 January 2006 16:10, linux-os (Dick Johnson) wrote:
> The problem is that every rule and every law takes
> away rights. Laws do not give rights. Rules do not
> give rights. Amendments to existing laws sometimes
> prevent the restriction of rights (like the first 10
> amendments of the US Constitution), however there
> are no rules or laws that ever, anywhere, provided
> any rights whatsoever. Rules, regulations, and laws
> are all about restricting rights.
>
> Sometimes the restrictions are necessary. For instance,
> except in very special circumstances, governments usually
> take away the inherent rights to kill, etc.
>
> The initial writer was correct. The GPL was supposed
> to be all about freedom. Then, there are hundreds of
> words that have nothing to do with freedom. They
> establish rules. The crybaby says; "You will play
> by my rules or..." Rules restrict freedom.

There's nothing crybaby about it! Copyleft is, as Stallman puts it, to flip 
copyright on its head. If there was no such thing as copyright or other forms 
of laws / restrictions on sharing ideas and intangible implementations, there 
would be zero reason for the GPL because all the GPL aims to do is to 
preserve the freedom to share.

So in that sense the GPL is about *enforcing* freedom.

I must admit that I'm terribly confused - I thought all of this was well 
understood and accepted. Why must everyone attempt to spin the GPL into 
something it is not? The only way the terms of this license would be in any 
way restrictive on anyone is if they decided to *exploit* a given GPL work in 
order to *restrict* the rights of future users of the work. In that case, the 
GPL would intervene and say 'No'.

> Perhaps these rules are necessary. However, for 20
> years before the Internet even existed, people were
> sharing source-code without rules. This was the
> principle behind the PROGRAM EXCHANGE and other
> obsolete BBS systems. At that time the ground-
> work of most all the file-compression routines,
> file-transmission routines, file-types, flight-
> simulators, etc., the stuff now claimed by others,
> was freely given away. Some expected their names
> to remain in the source, but eventually their
> names were changed to "Microsoft" or GPL. For
> example, Phil Katz. He invented "zip" and gunzip
> and all that stuff. He's now dead. His lifetime
> of work has been stolen by others and claimed
> as their own.

The GPL isn't about anyone's "work" or "credit". The licenses that ARE 
concerned with "credit" are licenses like the original BSD license with its 
attribution clause - you must stamp your product and documentation with 
notices that say your stuff was done by the University of Berkeley.

> The Internet gets established and somebody who's
> claim-to-fame was the development of the world's
> most complicated word-processor, establishes some
> legalese and a lot of well intentioned persons
> fall into his trap as he claims that he developed
> GNU/Linux as well. Wake up.
>

Are you suggesting that Stallman claims he developed GNU/Linux? Perhaps I'm 
just really misunderstanding what you just said, but if you really did mean 
it that way, perhaps you ought to point out where. If you're referring to the 
fact that Stallman doesn't like people calling the combination of the kernel 
and GNU tools "Linux", then it's only crybaby if you refuse to believe his 
published intentions (that "we" care more about open source than free 
software, and he's only trying to make sure that "free software" isn't 
forgotten). That last part I don't agree with (I think plenty of kernel 
people care greatly about free software), but I don't think it's silly to 
call the full OS GNU/Linux either. 

In any case, I don't think Stallman is concerned about fame so much as he is 
concerned with his movement. If it were most other people, I'd say it's all 
about fame, but Stallman has at least in my eyes proven himself to 
consistently concern himself only with his philosophy.

I don't really blame Stallman for anything he's said that you might take as an 
attempt to take credit. I get the impression from seeing the interplay 
between "open source" and "free software", ESR and RMS, that ESR would be 
happy to write free software and Richard Stallman right out of history. And 
since I value "free software" just as much as I value "open source", I don't 
want to see that happen.

The bottom line of all my rambling here is that these licenses aren't about 
maintaining credit at all, which I think you were implying in your message. 
Sans these "You will play by my rules or..." clauses, there would be no huge 
open source community because the proprietary software vendors would vacuum 
up anything of value and use it as leverage to lock people in. (And as for 
your comment about BBSes and the days without licenses, a zillion dollar 
proprietary software business had not yet been invented).

If you want open source licenses with no restrictions at all, that's "public 
domain". And "public domain" won't ever be what it could be until you abolish 
copyright.

Cheers,
Chase

Re: GPL V3 and Linux

[Harald Arnesen]

"linux-os \(Dick Johnson\)" <linux-os@analogic.com> writes:

> The problem is that every rule and every law takes
> away rights. Laws do not give rights. Rules do not
> give rights. Amendments to existing laws sometimes
> prevent the restriction of rights (like the first 10
> amendments of the US Constitution), however there
> are no rules or laws that ever, anywhere, provided
> any rights whatsoever. Rules, regulations, and laws
> are all about restricting rights.

True. Without laws, everything would be chaos (some would say "anarchy",
but every anarchist I know have their own rules of conduct and customs).

It's a bit like when the laws of physics take away the universe's right
to to as it pleases (except that the Universe is pretty much the boss,
and if it does something not allowed by the laws, the laws will better
change).

> Sometimes the restrictions are necessary. For instance,
> except in very special circumstances, governments usually
> take away the inherent rights to kill, etc.

But on the other hand, the (sane) governments give me the right to avoid
being killed by a madman with a gun, because they control who is allowed
to own a gun.

> The initial writer was correct. The GPL was supposed
> to be all about freedom. Then, there are hundreds of
> words that have nothing to do with freedom. They
> establish rules. The crybaby says; "You will play
> by my rules or..." Rules restrict freedom.

And most other "open source" licences take away the programmers freedom
to *keep* derived works free for all to use.

> Perhaps these rules are necessary. However, for 20
> years before the Internet even existed, people were
> sharing source-code without rules. This was the
> principle behind the PROGRAM EXCHANGE and other
> obsolete BBS systems. At that time the ground-
> work of most all the file-compression routines,
> file-transmission routines, file-types, flight-
> simulators, etc., the stuff now claimed by others,
> was freely given away. Some expected their names
> to remain in the source, but eventually their
> names were changed to "Microsoft" or GPL. For
> example, Phil Katz. He invented "zip" and gunzip
> and all that stuff. He's now dead. His lifetime
> of work has been stolen by others and claimed
> as their own.
>
> The Internet gets established and somebody who's
> claim-to-fame was the development of the world's
> most complicated word-processor, establishes some
> legalese and a lot of well intentioned persons
> fall into his trap as he claims that he developed
> GNU/Linux as well. Wake up.

There (probably) wouldn't be a GNU/Linux without the man who developed
the worlds best word-processor (and the worlds best programming editor,
and the framework for the worlds best e-mail client).
-- 
Hilsen Harald.

Re: GPL V3 and Linux

[Ian Kester-Haney]

Linux shouldn;t move to the GPL3 for the very reason that the DRM
restrictions would make linux incompatible with soon to be released
displays.  Also Nvidia and such would not be  able to make binary
drivers available.
Copyright for one work is set forward in law.  My view is that Artists
and their sponsors deserve
the right to prevent piracy.  In my view the Open Source Community
have an incompatible attitude.  In my mind the buying of a DVD means
that I watch it on DVD players be it on my computer or on the TV. 
while I beleive that I should be able to watch my DVDs on a linux
based system, it behooves the open source community to support it in a
legal way.  Cracking Access Control Sytems might be fun, but it only
generates huge controversy in concerned industries.  An Open Source
Access Control System that is respected by the FOSS community would be
a great diplomatic way to allow for more access to content.  My
personal view is that copying for my own personal use is ok, however
the converting of such material in a way not granted to me by the
Creator is not ethical.  Richard Stallman is painting himself into a
corner in this way.

Hence:
     The GNU/Linux community needs to work with the MPAA, RIAA and
other DRM players and work to support basic restrictions on copying
content while preserving the Creator/Companies right to sustain their
works.

Thankyou

Re: GPL V3 and Linux

[Kyle Moffett]

On Jan 23, 2006, at 20:55, Ian Kester-Haney wrote:
> Linux shouldn;t move to the GPL3 for the very reason that the DRM  
> restrictions would make linux incompatible with soon to be released  
> displays.

I'm sorry, what are you saying here?  I've not heard about these soon- 
to-be-released displays, could you elucidate, possibly summarizing or  
linking to references?

> Also Nvidia and such would not be able to make binary drivers  
> available.

In many peoples eyes, this would be a _good_ thing, besides, it's not  
clear whether or not their binary drivers are legal _now_, without  
concern for their status under the GPLv3

> Copyright for one work is set forward in law.  My view is that  
> Artists and their sponsors deserve the right to prevent piracy.

Have you tried describing music "piracy" to a small child (say, age  
6) in a way that distinguishes it from "sharing your things with  
other people"?  It's rather difficult, possibly even impossible.   
Many people argue that with modern technology, this distinction has  
become artificial and an artifact of an aging business model.  A  
number of artists who promote some music sharing have been doing very  
well.

> In my view the Open Source Community have an incompatible attitude.

I will ignore the second bit, since it's mostly a matter of opinion,  
but the Open Source Community (especially the Linux Community) values  
copyright extremely highly.  In fact, it is this copyright that makes  
the Linux Kernel sources a true democracy.  You cannot relicense the  
whole without agreement from all (or an extremely large majority  
bordering on "all") of the developers consent.

> In my mind the buying of a DVD means that I watch it on DVD players  
> be it on my computer or on the TV.

Or, according to Fair Use as set down in a number of court cases,  
make a single copy for backup, show privately to friends, convert to  
an alternative format for viewing in a different way or with other  
equipment.

> While I beleive that I should be able to watch my DVDs on a linux  
> based system, it behooves the open source community to support it  
> in a legal way.

I go to the store, I find a black box on the shelf, I pick it up, go  
to the cashier, pay for it, and leave the store.  At that point, I've  
purchased an object and may do whatever I like with said object.  I  
never signed any license or filled out any forms prior to paying  
money for it, and there was no condition that I do so, therefore no  
_license_ conditions apply.

On the other hand, copying the DVD and giving a copy to all your  
friends is _distributing_ it and therefore covered by _copyright_ law  
(not license/contract law), which makes it illegal for me to do so  
(although my personal opinion is that needs careful review and  
possible revision).

> Cracking Access Control Sytems might be fun, but it only generates  
> huge controversy in concerned industries.

<Biased Personal Opinion>
The industries do not matter.  The point of the government is to help  
and protect the _people_.  This means that to a limited extent, the  
government protects individuals copyrights, and allows corporations  
some rights (because they provide jobs, services, etc).  This does  
not mean that the government should do anything the industry wants  
even though hundreds of millions of people are breaking that law on a  
daily basis.  Something that widespread (especially given the lack of  
issues arising thereof) indicates that the _law_ is wrong, not the  
many millions doing the breaking.  This bends more towards the  
copyright issues I talk about above, but applies here too.
</Biased Personal Opinion>

Besides, a DRM system is pointless and futile; it's trying to protect  
data from people by giving them the encrypted data, the algorithm,  
and the key.  Any cryptographer will tell you that you are bound to  
lose from the start.

> An Open Source Access Control System that is respected by the FOSS  
> community would be a great diplomatic way to allow for more access  
> to content.

Open Source Access Control System:

if (access_allowed(media_descriptor, user_data)) {
	provide_data(media_descriptor);
}

My 3-line patch to "fix" it to let me watch my German DVD in the US:

--- oldfile
+++ newfile
-if (access_allowed(media_descriptor, user_data)) {
  	provide_data(media_descriptor);
-}

This is a _fundamentally_ _flawed_ idea.

> My personal view is that copying for my own personal use is ok

Good

> however the converting of such material in a way not granted to me  
> by the Creator is not ethical.

Why does the creator have any say in what you can do with it  
personally?  I'm legally allowed to buy a copy of MS Windows and burn  
it for symbolic value; why the hell would we want to allow a  
*CORPORATION* (Read: bunch of greedy rich executives) control what  
you can do with stuff.  Heck, we don't trust the _government_ (Read:  
bunch of greedy rich lawyers), or sometimes even one's personal  
_church_ to control.

> The GNU/Linux community needs to work with the MPAA, RIAA and other  
> DRM players and work to support basic restrictions on copying  
> content while preserving the Creator/Companies right to sustain  
> their works.

We don't _need_ to work with anybody, we're just converting abstract  
mathematical algorithms to a more practically useable form and  
publishing the result.  The fact that somebody with lawlessness in  
mind could do something illegal with our formalized codified  
published math files is totally irrelevant.  In the US (where I live,  
can't speak for other countries) we don't blame the gun manufacturers  
for what people do with submachine guns, so why should we blame the  
software developers (Read: practical mathematicians) for what people  
do with their programs?

Cheers,
Kyle Moffett

--
Simple things should be simple and complex things should be possible
   -- Alan Kay

Re: GPL V3 and Linux

[Bernd Petrovitsch]

On Mon, 2006-01-23 at 19:55 -0600, Ian Kester-Haney wrote:
[...]
> Copyright for one work is set forward in law.  My view is that Artists
> and their sponsors deserve
> the right to prevent piracy.  In my view the Open Source Community

Of course and ATM there are - at least in the free world - more than
enough possibilities to punish "piracy" of copied copyrighted work.
The whole DRM (which actually shopuld be read as "Digital Restrictions
Management") and "against copyright piracy" campaign is to take away
legal rights like "playing a legallay produced and bought DVD as often
was you wish" and to limit you to "view it at most 3 times and disallow
any copy - expecially legal ones".

> have an incompatible attitude.  In my mind the buying of a DVD means
> that I watch it on DVD players be it on my computer or on the TV. 
> while I beleive that I should be able to watch my DVDs on a linux
> based system, it behooves the open source community to support it in a
> legal way.  Cracking Access Control Sytems might be fun, but it only

The copyright-industry to-be-implemented access control is in fact
illegal.

> generates huge controversy in concerned industries.  An Open Source

Yes, because the concerned industries ignored the develoment in last 20
years. And they don't like certain aspects and rights of e.g.
continental European laws (yes, copying music CDs privately and giving
them away as a birthday present is completely *legal* hereover. We
actually *pay* for this right with ~40 eurocent per writable medium
since decades, i.e. since music tapes were young and paper copying
machines were very expensive).

> Access Control System that is respected by the FOSS community would be
> a great diplomatic way to allow for more access to content.  My

You are listening and beliving to the propaganda too much.

> personal view is that copying for my own personal use is ok, however
> the converting of such material in a way not granted to me by the
> Creator is not ethical.  Richard Stallman is painting himself into a

The creator (if you mean the artist) and the above mentioned "concerned
industries" are two different things.
And you probably have at the moment far more rights to copy for your
private use than you know - and this is at stake.

The strategic problem, that DRM has is: It doesn't hurt (or even apply
to) the big commercial (and thus illegal) copying organizations. And -
at least as far I the propaganda read - are the big evils.
The only target is the private consumer.
Do you want to pay 1$ for every time you hear a song?
Think of DVDs/CDs which *require* Internet access (next generation game
console will deliver that probably) and a (properly filled) PayPal
account.

> corner in this way.

	Bernd
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services

Re: GPL V3 and Linux

[Alan Cox]

On Llu, 2006-01-23 at 19:55 -0600, Ian Kester-Haney wrote:
> Linux shouldn;t move to the GPL3 for the very reason that the DRM
> restrictions would make linux incompatible with soon to be released

The DRM restrictions mostly only restate some of the less clear effects
of the GPL. The GPL v2 already requires all the keys etc since it
requires the scripts to build. It just makes it clearer.

> displays.  Also Nvidia and such would not be  able to make binary
> drivers available.

The GPL doesn't permit them to anyway. If they are legal then it is
because they are not derivative works which forms an implicit barrier in
copyright law. Essentially copyright law limits itself to works based on
other works. So as an author of a book I can say "You may not copy this
book" but I cannot (in copyright enforced agreement) say "You may not
write a book on this subject if you read mine", only to control works
based upon mine in a material way (which is what "derivative work" is
all about).

That also means for example that a GPL OS running a non-derivative
application has no power to forbid that application from using DRM
itself. 

Now there is a case where it get much messier - GPL with exceptions/LGPL
being the clear one. A glibc that prohibited linking with DRM using code
would raise much more complicated problems, but that is for the
glibc/FSF list to argue over and does need addressing sensibly.


>      The GNU/Linux community needs to work with the MPAA, RIAA and
> other DRM players and work to support basic restrictions on copying
> content while preserving the Creator/Companies right to sustain their
> works.

What on earth makes you think those bodies want creators to have any
rights. You've obviously never watched creators and the "industry"
fighting each other. Anyway fighting DRM is actually helping even the
music "industry"  if H. Valarian's analysis is correct.

Free Software is about -freedom- that means the freedom to do things
like play music you own the rights to play and the freedom to distribute
music you have the right to distribute. The dream of big music industry
is mandatory DRM where there is no way for an artist to escape their
clutches and publish music without them getting a very large cut of, if
not all the profits.

Much of the big music industry today is like vanity publishing, they
sign you up, bill you for the costs of making the recordings, screw you
on royalty deals then charge you the remainder the royalties didn't
cover.

So why should we work with the RIAA members (especially as there is good
evidence to suggest that one of them stole GPL code for a proprietary
DRM system) ?

Free means that everyone must be able to use Linux

Alan

Re: GPL V3 and Linux

[Wartan Hachaturow]

On 1/24/06, Harald Arnesen <harald@skogtun.org> wrote:

> There (probably) wouldn't be a GNU/Linux without the man who developed
> the worlds best word-processor (and the worlds best programming editor,
> and the framework for the worlds best e-mail client).

Last I checked, vim and mutt were not written by RMS.

--
Regards, Wartan.

RE: GPL V3 and Linux

[David Schwartz]

> Sometimes the restrictions are necessary. For instance,
> except in very special circumstances, governments usually
> take away the inherent rights to kill, etc.

	I guess I can't figure out what you could possibly mean by the word "right"
such that the phrase "inherent rights to kill" is meaningful. Perhaps you
could clarify.

	DS

RE: GPL V3 and Linux

[Lee Revell]

On Tue, 2006-01-24 at 02:23 -0800, David Schwartz wrote:
> > Sometimes the restrictions are necessary. For instance,
> > except in very special circumstances, governments usually
> > take away the inherent rights to kill, etc.
> 
> 	I guess I can't figure out what you could possibly mean by the word "right"
> such that the phrase "inherent rights to kill" is meaningful. Perhaps you
> could clarify.

This discussion could not have less to do with kernel development,
PLEASE take it elsewhere.

Lee

RE: GPL V3 and Linux

[linux-os (Dick Johnson)]

On Tue, 24 Jan 2006, David Schwartz wrote:

>
>> Sometimes the restrictions are necessary. For instance,
>> except in very special circumstances, governments usually
>> take away the inherent rights to kill, etc.
>
> 	I guess I can't figure out what you could possibly mean by the word "right"
> such that the phrase "inherent rights to kill" is meaningful. Perhaps you
> could clarify.
>
> 	DS

Simple, from Government 101. Suppose you start a new country unencumbered
with rules and laws. You have "total" freedom, therefore all rights.
Because you believe that everybody is "good" (you decide what that means)
and would, therefore, never do anything "bad", you don't need any
laws.

Sooner or later somebody does something "bad" (like kills somebody).
So, you make a law against killing. As you make that first law, you
have restricted rights. That's what laws do, they restrict rights.

Unfortunately, it never stops with the "obviously necessary"
laws. Eventually, every time somebody believes he or she has been
harmed somehow, the cry goes out; "There ought to be a law....".
Some  goody-twoshoes in the government makes a new law. Eventually,
there are so many laws that there is no freedom whatsoever.

Most laws, designed to protect, have far-reaching consequences that
actually cause more problems than they are supposed to solve. That's
the nature of Law and Government in general. That's why it's important
to control (reduce) the number of laws that exist and control the
size of government. Of course, once the government controls the schools
all is lost.

Cheers,
Dick Johnson
Penguin : Linux version 2.6.13.4 on an i686 machine (5589.54 BogoMips).
Warning : 98.36% of all statistics are fiction.
.

****************************************************************
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 DeliveryErrors@analogic.com - and destroy all copies of this information, including any attachments, without reading or disclosing them.

Thank you.

Re: GPL V3 and Linux

[Jeff V. Merkey]

linux-os (Dick Johnson) wrote:

>On Tue, 24 Jan 2006, David Schwartz wrote:
>
>  
>
>>>Sometimes the restrictions are necessary. For instance,
>>>except in very special circumstances, governments usually
>>>take away the inherent rights to kill, etc.
>>>      
>>>
>>	I guess I can't figure out what you could possibly mean by the word "right"
>>such that the phrase "inherent rights to kill" is meaningful. Perhaps you
>>could clarify.
>>
>>	DS
>>    
>>
>
>Simple, from Government 101. Suppose you start a new country unencumbered
>with rules and laws. You have "total" freedom, therefore all rights.
>Because you believe that everybody is "good" (you decide what that means)
>and would, therefore, never do anything "bad", you don't need any
>laws.
>
>Sooner or later somebody does something "bad" (like kills somebody).
>So, you make a law against killing. As you make that first law, you
>have restricted rights. That's what laws do, they restrict rights.
>  
>

Say rather than restrict rights they define where your rights end and 
the rights
of another person begin. Rather they "balance" rights by drawing a line 
between
the rights of individuals and the rights of the state. The complex case 
is something
called a "compelling interest". i.e. The government has a compelling 
interest to
protect citizens from killing each other.

>Unfortunately, it never stops with the "obviously necessary"
>laws. Eventually, every time somebody believes he or she has been
>harmed somehow, the cry goes out; "There ought to be a law....".
>Some  goody-twoshoes in the government makes a new law. Eventually,
>there are so many laws that there is no freedom whatsoever.
>
>  
>
There are so many laws, you need to have courts in order to perform
"balancing tests" between the rights of individuals (courts of equity)
and the rights of the governement (compelling interests vs. the rights 
of Individuals)

>Most laws, designed to protect, have far-reaching consequences that
>actually cause more problems than they are supposed to solve. That's
>the nature of Law and Government in general. That's why it's important
>to control (reduce) the number of laws that exist and control the
>size of government. Of course, once the government controls the schools
>all is lost.
>
>  
>
Why Dick, we have something in common -- we are both libertarians (political
party that believes less laws and governemtn control is a very good thing).

Jeff

>Cheers,
>Dick Johnson
>Penguin : Linux version 2.6.13.4 on an i686 machine (5589.54 BogoMips).
>Warning : 98.36% of all statistics are fiction.
>.
>
>****************************************************************
>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 DeliveryErrors@analogic.com - and destroy all copies of this information, including any attachments, without reading or disclosing them.
>
>Thank you.
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/
>
>  
>

Re: GPL V3 and Linux

[Ian Kester-Haney]

Allowing for the playing of DRM protected content is accepted.  While
it is sad to see big business rule the day, it is still copyrighted
works we are talking about.  GPL v3 says that the use of DRM in or
around the GPLv3 software would be a direct violation and therfore
illegal under the GPL liscense.  Respecting the limits setup by the
DRM or ACS would be a proactive step in making linux more corporate
friendly.  Should the individuals producing, directing and starring in
movies and music be penalized for the abuses of the recording
industry.  Piracy does exist and supporting the real creative works of
others is important.

According to the Managed Copy Protection in upcoming HD-DVD and
Blu-Ray discs, if the channel between digital display devices is not
authenticated as direct to display, the resolution would be cut to
prevent electronic copies from being easily made.

I seruiosly doubt that anyone commited to the Open Source community
would condone piracy, just as I am sure that Protection schemes always
break down in the end.  It is sometimes irresponsible to circumvent
methods designed to protect peoples copyright.  The viral liscensing
in the GPL v 2 allows for the best integration with 3rd party
applications.
The GPLv3 would not allow any copyrighted materials under DRM to be
viewed, one could even argue that protected PDF files might constitute
DRM and not be allowed under the new GPL.  I think that subverting the
efforts of companies and artists to protect their works is
unnaceptable.  Public crap about hacking iTunes to get unprotected
files is detrimental to the Open Source Community.  It should be noted
that DRM is not inherently bad, implementations are currently pretty
crappy, but surely an open source DRM could be presented, and a step
to make that happen would be to keep the Kernel under GPLv2.

On 1/24/06, Jeff V. Merkey <jmerkey@wolfmountaingroup.com> wrote:
> linux-os (Dick Johnson) wrote:
>
> >On Tue, 24 Jan 2006, David Schwartz wrote:
> >
> >
> >
> >>>Sometimes the restrictions are necessary. For instance,
> >>>except in very special circumstances, governments usually
> >>>take away the inherent rights to kill, etc.
> >>>
> >>>
> >>      I guess I can't figure out what you could possibly mean by the word "right"
> >>such that the phrase "inherent rights to kill" is meaningful. Perhaps you
> >>could clarify.
> >>
> >>      DS
> >>
> >>
> >
> >Simple, from Government 101. Suppose you start a new country unencumbered
> >with rules and laws. You have "total" freedom, therefore all rights.
> >Because you believe that everybody is "good" (you decide what that means)
> >and would, therefore, never do anything "bad", you don't need any
> >laws.
> >
> >Sooner or later somebody does something "bad" (like kills somebody).
> >So, you make a law against killing. As you make that first law, you
> >have restricted rights. That's what laws do, they restrict rights.
> >
> >
>
> Say rather than restrict rights they define where your rights end and
> the rights
> of another person begin. Rather they "balance" rights by drawing a line
> between
> the rights of individuals and the rights of the state. The complex case
> is something
> called a "compelling interest". i.e. The government has a compelling
> interest to
> protect citizens from killing each other.
>
> >Unfortunately, it never stops with the "obviously necessary"
> >laws. Eventually, every time somebody believes he or she has been
> >harmed somehow, the cry goes out; "There ought to be a law....".
> >Some  goody-twoshoes in the government makes a new law. Eventually,
> >there are so many laws that there is no freedom whatsoever.
> >
> >
> >
> There are so many laws, you need to have courts in order to perform
> "balancing tests" between the rights of individuals (courts of equity)
> and the rights of the governement (compelling interests vs. the rights
> of Individuals)
>
> >Most laws, designed to protect, have far-reaching consequences that
> >actually cause more problems than they are supposed to solve. That's
> >the nature of Law and Government in general. That's why it's important
> >to control (reduce) the number of laws that exist and control the
> >size of government. Of course, once the government controls the schools
> >all is lost.
> >
> >
> >
> Why Dick, we have something in common -- we are both libertarians (political
> party that believes less laws and governemtn control is a very good thing).
>
> Jeff
>
> >Cheers,
> >Dick Johnson
> >Penguin : Linux version 2.6.13.4 on an i686 machine (5589.54 BogoMips).
> >Warning : 98.36% of all statistics are fiction.
> >.
> >
> >****************************************************************
> >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 DeliveryErrors@analogic.com - and destroy all copies of this information, including any attachments, without reading or disclosing them.
> >
> >Thank you.
> >-
> >To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> >the body of a message to majordomo@vger.kernel.org
> >More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >Please read the FAQ at  http://www.tux.org/lkml/
> >
> >
> >
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: GPL V3 and Linux

[Bernd Petrovitsch]

On Tue, 2006-01-24 at 19:21 -0600, Ian Kester-Haney wrote:
[...]
> friendly.  Should the individuals producing, directing and starring in
> movies and music be penalized for the abuses of the recording

Of course not. But I can't see any mentioning of these in the propaganda
of the big music industry.

> industry.  Piracy does exist and supporting the real creative works of
> others is important.

ACK.
But DRM will not stop (or even hinder seriously) the big commercial
copying organizations only the private copies (even if they are legal by
allmeans).
So this argument is plain simply moot.

> I seruiosly doubt that anyone commited to the Open Source community
> would condone piracy, just as I am sure that Protection schemes always
> break down in the end.  It is sometimes irresponsible to circumvent
> methods designed to protect peoples copyright.  The viral liscensing
> in the GPL v 2 allows for the best integration with 3rd party
> applications.
> The GPLv3 would not allow any copyrighted materials under DRM to be
> viewed, one could even argue that protected PDF files might constitute
> DRM and not be allowed under the new GPL.  I think that subverting the

With the exception that I *can* circumvent the protection on PDFs *if*
I'm legally allowed to copy the copyrighted work (with or without the
owner's permission - this is one reason for a legal copy. But there are
others which cannot be inhibited by the copyright holder - which is
usually not the artist).

> files is detrimental to the Open Source Community.  It should be noted
> that DRM is not inherently bad, implementations are currently pretty

No, but the motivation behind and the reasons for it existence is bad.

> crappy, but surely an open source DRM could be presented, and a step
> to make that happen would be to keep the Kernel under GPLv2.

Let's see if DRM actually get accepted or shares the success of micro
channel and other big business must-have "inventions".

[ Fullquote killed ]

	Bernd
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services

[OT] Re: GPL V3 and Linux

[Steven Rostedt]

On Wed, 2006-01-25 at 10:42 +0100, Bernd Petrovitsch wrote:
> On Tue, 2006-01-24 at 19:21 -0600, Ian Kester-Haney wrote:
> [...]
> > friendly.  Should the individuals producing, directing and starring in
> > movies and music be penalized for the abuses of the recording
> 
> Of course not. But I can't see any mentioning of these in the propaganda
> of the big music industry.
> 
> > industry.  Piracy does exist and supporting the real creative works of
> > others is important.
> 
> ACK.
> But DRM will not stop (or even hinder seriously) the big commercial
> copying organizations only the private copies (even if they are legal by
> allmeans).
> So this argument is plain simply moot.

I'll even go a step farther.  All these technical restrictions for
prevention of copying really backfires in the end.  For example, this
stupid Region coding of DVDs.  I go to Germany quite a lot, and to help
out my German, I buy DVDs there so when I'm home I can listen to movies
in German. The problem arises when I try to play these at home, since my
DVD player is coded for the US.  The only way I can watch movies that I
legally bought in Germany is to copy them (probably with illegal
software) and turn off the region code so I can watch them on my home
DVD players.

Now here's the kicker.  A colleague of mine asked why I even bother
buying the DVDs in the store, when I could buy them on the black market
for a much cheaper price and they will play on my home player.  This is
very tempting, but my conscience tells me to pay those that actually
make the films.  But it does beg the question of what the Region coding
is actually trying to stop?

Sorry, this is not about kernel programming, any replies should probably
be sent off list.

-- Steve

Re: GPL V3 and Linux

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 1156 bytes --]

On Wed, 25 Jan 2006 10:42:24 +0100, Bernd Petrovitsch said:

> With the exception that I *can* circumvent the protection on PDFs *if*
> I'm legally allowed to copy the copyrighted work (with or without the
> owner's permission - this is one reason for a legal copy. But there are
> others which cannot be inhibited by the copyright holder - which is
> usually not the artist).

Actually, in the US, it is in fact illegal to bypass a protection scheme
*even if the content is something you have legal rights to*.

http://www4.law.cornell.edu/uscode/html/uscode17/usc_sec_17_00001201----000-.html

17 USC 1201(a)(1)(A) says:

(A) No person shall circumvent a technological measure that effectively
controls access to a work protected under this title. The prohibition contained
in the preceding sentence shall take effect at the end of the 2-year period
beginning on the date of the enactment of this chapter. 

The rest of 1201(a)(1) talks about getting an exemption, good for 3 years only,
to allow you to bypass a control for a non-infringing use.

Got that?  You have to apply for special permission to bypass to get data that
you have rights to use....

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

Re: GPL V3 and Linux

[Bernd Petrovitsch]

On Thu, 2006-01-26 at 22:10 -0500, Valdis.Kletnieks@vt.edu wrote:
> On Wed, 25 Jan 2006 10:42:24 +0100, Bernd Petrovitsch said:
> 
> > With the exception that I *can* circumvent the protection on PDFs *if*
> > I'm legally allowed to copy the copyrighted work (with or without the
> > owner's permission - this is one reason for a legal copy. But there are
> > others which cannot be inhibited by the copyright holder - which is
> > usually not the artist).
> 
> Actually, in the US, it is in fact illegal to bypass a protection scheme
> *even if the content is something you have legal rights to*.

Well, the so-called "land of the free". SCNR .....

> http://www4.law.cornell.edu/uscode/html/uscode17/usc_sec_17_00001201----000-.html
> 
> 17 USC 1201(a)(1)(A) says:
> 
> (A) No person shall circumvent a technological measure that effectively
                                                              ^^^^^^^^^^
> controls access to a work protected under this title. The prohibition contained
  ^^^^^^^^^^^^^^^
Actually there is similar wording here (but of course in German) used
for the similar purpose. The problem with this kind of law is IMHO:
-) "effectively controls access": If I (or someone else) can circumvent
   it, it is obviously not "effective".
-) If I (and no one else) cannot circumvent it, the laws/court decisions
   as such is basically pointless because it doesn't limit or hinder
   anything.
And we have no definition (in the laws) hereover whatever "effective"
should mean and hoe *I* can determine (which a sufficient large chance
of getting it right) if a given protection scheme must be considered
"effective".

Don't get me wrong, I understand how it is meant what such rules should
achieve an, but I request from lawyers (as such) that they write
laws/court decisions down in an unambigous way (for a non-law person -
remember that laws affect *all* people so every law and court decision
should IMHO readable and understandable by the average citizen).

And if they can't write it down unambigously, I actually question if we
want to accept laws/court decisions about rules and concepts which
cannot be even written down in a simple enough and clear way.

[..]
> Got that?  You have to apply for special permission to bypass to get data that
> you have rights to use....

Yes, because that is the primary goal of all of the laws in that area in
last years: To effectively take away legal rights from you that you
actually legally have (or better: had).

	Bernd
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services

Re: GPL V3 and Linux

[Bas Westerbaan]

> should IMHO readable and understandable by the average citizen).
>
> And if they can't write it down unambigously, I actually question if
> we
> want to accept laws/court decisions about rules and concepts
> which
> cannot be even written down in a simple enough and clear way.

Ambiguous laws are compromizes. They are required sometimes. Who will
prevail depends on who can affort the best lawyer in court.

Don't get me wrong, I'd love readable laws, but with the current
politics it isn't possible.

--
Bas Westerbaan
http://blog.w-nz.com/
GPG Pub. Keys: http://w-nz.com/r/f

Re: GPL V3 and Linux

[Bernd Petrovitsch]

On Fri, 2006-01-27 at 11:13 +0100, Bas Westerbaan wrote:
> > should IMHO readable and understandable by the average citizen).
> >
> > And if they can't write it down unambigously, I actually question if
> > we
> > want to accept laws/court decisions about rules and concepts
> > which
> > cannot be even written down in a simple enough and clear way.
> 
> Ambiguous laws are compromizes. They are required sometimes. Who will

Technically, all laws are ambigous (since you do not have them based on
clear definitions as in mathemnatic and the like but only natural
language).
But I have a problem with that extreme type of ambigousness that
actually is IMHO already a contradiction.
And remember: Continental Europe doesn't have case law (except the mess
and intentions produced by the EPO). So the law system here *needs*
relatively clear laws which are of course interpreted by judges in court
and clarified with court decisions. But a court decision actually can
not make a law.
And now explain to me how a judge with a technical knowledge as a
typical 50 year old Win* user should decide if a given "access control"
is "effective" or not?
Of course they have external specialists (how are they callen in
English?) in court who try to explain and give the (fair) opinion on the
given case. But usually each party gets such a quote supporting their
side. So how should the judge actually decide correctly?
And IMHO it doesn't work out to define the judges decision as
"correct" (like usual in the law system as such) if it is technically
incorrect.

> prevail depends on who can affort the best lawyer in court.
> 
> Don't get me wrong, I'd love readable laws, but with the current
> politics it isn't possible.

It may be worse than with other politics, but layers are not interested
in readable laws.
You can make much more money out of a system where you need secret
languages (think of the catholic church in early medieval age with
everything in Latin) .....

	Bernd, fairly off-topic now. Sorry.
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services

Re: GPL V3 and Linux

[Olivier Galibert]

On Fri, Jan 27, 2006 at 10:54:07AM +0100, Bernd Petrovitsch wrote:
> > (A) No person shall circumvent a technological measure that effectively
>                                                               ^^^^^^^^^^
> > controls access to a work protected under this title. The prohibition contained
>   ^^^^^^^^^^^^^^^
> Actually there is similar wording here (but of course in German) used
> for the similar purpose. The problem with this kind of law is IMHO:
> -) "effectively controls access": If I (or someone else) can circumvent
>    it, it is obviously not "effective".

Wrong definition, try again.  "Effective" has multiple meanings in
english, one of them being "existing in fact, not theorical, real".
Check wordnet, it's getting impressively usable nowadays.

In other words, "effective"'s meaning can either move towards
"efficient" or "actual/existing", depending on the context.

  OG.

Re: GPL V3 and Linux

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 829 bytes --]

On Fri, 27 Jan 2006 10:54:07 +0100, Bernd Petrovitsch said:
> On Thu, 2006-01-26 at 22:10 -0500, Valdis.Kletnieks@vt.edu wrote:

> > 17 USC 1201(a)(1)(A) says:
> > 
> > (A) No person shall circumvent a technological measure that effectively
>                                                               ^^^^^^^^^^
> > controls access to a work protected under this title. The prohibition conta
ined
>   ^^^^^^^^^^^^^^^
> Actually there is similar wording here (but of course in German) used
> for the similar purpose. The problem with this kind of law is IMHO:
> -) "effectively controls access": If I (or someone else) can circumvent
>    it, it is obviously not "effective".

As Skylarov found out when he got into a pissing match with Adobe, ROT-13
qualifies as an "effective access control" as far as the law is concerned.

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

Re: GPL V3 and Linux

[Bernd Petrovitsch]

On Fri, 2006-01-27 at 12:51 -0500, Valdis.Kletnieks@vt.edu wrote:
> On Fri, 27 Jan 2006 10:54:07 +0100, Bernd Petrovitsch said:
> > On Thu, 2006-01-26 at 22:10 -0500, Valdis.Kletnieks@vt.edu wrote:
> 
> > > 17 USC 1201(a)(1)(A) says:
> > > 
> > > (A) No person shall circumvent a technological measure that effectively
> >                                                               ^^^^^^^^^^
> > > controls access to a work protected under this title. The prohibition conta
> ined
> >   ^^^^^^^^^^^^^^^
> > Actually there is similar wording here (but of course in German) used
> > for the similar purpose. The problem with this kind of law is IMHO:
> > -) "effectively controls access": If I (or someone else) can circumvent
> >    it, it is obviously not "effective".
> 
> As Skylarov found out when he got into a pissing match with Adobe, ROT-13

If even ROT-13 is "effective access control", anything which is titles
"Access Control" applies.

> qualifies as an "effective access control" as far as the law is concerned.

AFAICS several folks got my point ....

	Bernd
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services

Re: GPL V3 and Linux

[Rik van Riel]

On Tue, 24 Jan 2006, Ian Kester-Haney wrote:

> I think that subverting the efforts of companies and artists to protect 
> their works is unnaceptable.

That all depends on what they are protecting their works against.
If they were protecting their works against illegal copying, that
protection should indeed not be circumvented.

On the other hand, if all they are "protecting against" is lawful
and legitimate behaviour by joe six-pack consumers, I do not see
why those efforts should be protected or helped...

> It should be noted that DRM is not inherently bad,

It should also be noted that DRM does not protect works against
copying - the one thing that DRM was supposed to achieve.

-- 
All Rights Reversed

Re: GPL V3 and Linux

[Graham Murray]

Ian Kester-Haney <ikesterhaney@gmail.com> writes:

> The GPLv3 would not allow any copyrighted materials under DRM to be
> viewed, one could even argue that protected PDF files might constitute
> DRM and not be allowed under the new GPL.

That is not how I read the GPLv3. The way I read it is that a GPL'd
application cannot be used to *create* the protection, not that it
cannot be used to view or use the protected work. 

Re: GPL V3 and Linux

[Helge Hafting]

Alexander Shishckin wrote:

>On 1/21/06, Chase Venters <chase.venters@clientec.com> wrote:
>  
>
>>On Friday 20 January 2006 20:27, Alexander Shishckin wrote:
>>    
>>
>>>GPLv3 tends to get on top of the most braindead things ever known to
>>>software development. It is, in fact, a one-too-many example of how a
>>>person who cannot be seriously considered to be a computer programmer
>>>tries to have his one-too-many revenge on companies which employ real
>>>software developers and produce real world software. Someone should
>>>probably put an end to these miserable efforts.
>>>      
>>>
>>Why does everyone assume that Stallman is out to 'get revenge' on companies?
>>Is his desire for freedom so hard to grasp and believe that all you can do is
>>spin it into silly conspiracies?
>>    
>>
>Ain't that obvoius? Every second word that you read in GPLs is either
>'freedom' or 'share' and the rest of the document has absolutely
>nothing to do with both, just restricting our *freedom* to *share*.
>  
>
Wrong.  The GPL does not in any way take away your *freedom to share*.
It does take away your freedom to *restrict* though, it limits your freedom
to *prevent others from sharing*.

What problem could you possibly have with that?  Licences are simple -
don't like them, don't use the product.  Why complain about it?

I don't like licences where I have to pay *money* just to get some software.
Expensive, and a waste if I end up not using that sw much.
So I try to avoid that kind when possible.  Fortunately, that is 
possible almost
all the time for me.

Now, if you don't like the GPL - don't use any GPL-licenced products then!
Why nag about it?  I don't bitch about how the "pay" licences limits my
business opportunities. (I cannot possibly afford to buy
every payware program.) In particular, I cannot sell PCs containing lots
of pay-licences software without actually _paying_, thereby driving up
the price.  But complaining about that would be silly.

The same applies to you.  If you want to include GPL stuff in a product,
then your product goes GPL too.  That is the _price_ for GPL code.  If that
price is too high for you, don't use GPL code then!  That should
be fine with you, and fine with us.  But again - complaining about it is 
silly.


>This is not exactly the way these words are meant to be used.
>
>  
>
>>Why do people not recognize that his GNU project has built significant things?
>>Do you not realize that Linux is licensed GPLv2, which is also Stallman's
>>license?
>>    
>>
>Surely significant things/projects do have their origin there. This is
>yet another sad consequence. I hope we can someday get rid of all this
>'GNU is ...' crap and build a better world that understands 'freedom'
>for what it really is, not just
>I'll-sue-you-should-you-use-my-code-in-your-proprietary-product stinky
>sort of thing.
>  
>
Well, write your own better licence then, and use it!  Then you'll have the
satisfaction of seeing others use your code in their proprietary products.
Nobody stops you from doing that.

Helge Hafting

Re: GPL V3 and Linux

[Bernd Petrovitsch]

On Sat, 2006-01-21 at 08:56 +0300, Alexander Shishckin wrote:
> On 1/21/06, Chase Venters <chase.venters@clientec.com> wrote:
> > On Friday 20 January 2006 20:27, Alexander Shishckin wrote:
> > > GPLv3 tends to get on top of the most braindead things ever known to
> > > software development. It is, in fact, a one-too-many example of how a
> > > person who cannot be seriously considered to be a computer programmer
> > > tries to have his one-too-many revenge on companies which employ real
> > > software developers and produce real world software. Someone should
> > > probably put an end to these miserable efforts.
> >
> > Why does everyone assume that Stallman is out to 'get revenge' on companies?
> > Is his desire for freedom so hard to grasp and believe that all you can do is
> > spin it into silly conspiracies?
> Ain't that obvoius? Every second word that you read in GPLs is either

No, because MSFTs business model is not the only one in the software
world (though MSFT keeps teaching that).

	Bernd
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services