|
Home > Archive > Debian Developers > June 2004 > How long is it acceptable to leave *undistributable* files in the kernel package?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
How long is it acceptable to leave *undistributable* files in the kernel package?
|
|
| Nathanael Nerode 2004-06-15, 5:55 pm |
| Such inclusion exposes Debian to legal liability, right?
I ask because such a file is present in Debian's Linux kernel sources and
there seems to have been no attempt to remove it, despite the upload of
new versions since the bug report.
drivers/usb/misc/emi26_fw.h:
* The firmware contained herein is Copyright (c) 1999-2002 Emagic
* as an unpublished work. This notice does not imply unrestricted
* or public access to this firmware which is a trade secret of Emagic,
* and which may not be reproduced, used, sold or transferred to
* any third party without Emagic's written consent. All Rights Reserved.
This is the subject of bug #242895, which was (incorrectly) merged with
other bugs.
(There are a large number of other "firmware" files with no explicit license
from the copyright holder at all, which *also* means, under copyright law,
no permission to distribute. This is the most extreme case because it
explicitly states that there is no permission to distribute. :-P )
Forget, for now, the issues of DFSG-freenesss and GPL-compatibility. Is
it sane for Debian to be distributing copyrighted code without permission,
even if kernel.org does?
I request that the people planning to take over kernel maintenance comment
ASAP on their plans regarding this issue, and what sort of help would be
appreciated.
--
There are none so blind as those who will not see.
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andreas Schuldei 2004-06-15, 5:55 pm |
| * Nathanael Nerode (neroden@twcny.rr.com) [040615 21:27]:
> I request that the people planning to take over kernel maintenance comment
> ASAP on their plans regarding this issue, and what sort of help would be
> appreciated.
would it be ok if they hand in their humble suggestions at your
front desk in written form? or would you rather want them to
report to you in person?
hint: your tone is not appropriet and you are not the one to
order them around.
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Matthew Wilcox 2004-06-15, 5:55 pm |
| On Mon, Jun 14, 2004 at 12:57:15PM -0400, Nathanael Nerode wrote:
> I ask because such a file is present in Debian's Linux kernel sources and
> there seems to have been no attempt to remove it, despite the upload of
> new versions since the bug report.
>
> drivers/usb/misc/emi26_fw.h:
>
> * The firmware contained herein is Copyright (c) 1999-2002 Emagic
> * as an unpublished work. This notice does not imply unrestricted
> * or public access to this firmware which is a trade secret of Emagic,
> * and which may not be reproduced, used, sold or transferred to
> * any third party without Emagic's written consent. All Rights Reserved.
>
> This is the subject of bug #242895, which was (incorrectly) merged with
> other bugs.
I would think you would be better off asking on debian-legal rather than
here. My non-lawyerly opinion is that since it's factually incorrect
(it has been published, it is not a trade secret), these are probably
incompetent people who couldn't organise a piss-up in a brewery, let
alone a lawsuit.
--
"Next the statesmen will invent cheap lies, putting the blame upon
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince
himself that the war is just, and will thank God for the better sleep
he enjoys after this process of grotesque self-deception." -- Mark Twain
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Matthew Wilcox 2004-06-16, 5:56 pm |
| On Tue, Jun 15, 2004 at 05:51:03PM -0500, Joe Wreschnig wrote:
> On Tue, 2004-06-15 at 14:38, Andreas Schuldei wrote:
>
> Then who is? This problem has been known about for a long time, and a
> lot of developers would like to see some progress on it. No one should
> *need* to order Debian developers to follow copyright law in their
> Debian work; it should be the default. Plus, I'd rather Nathanael do it
> than the US government. (I also found his tone civil, if curt.)
>
> It's not even a request to remove it, just one asking for some
> indication on what kind of schedule or plans are going on, and *an offer
> for help*.
The correct thing to do was what I did -- contact the USB maintainer, who
was previously unaware of this problem:
<gregkh> willy: cool, I got back a response with an email from emagic giving the blessing to include the drivers in linux.
<gregkh> willy: just looks like I need to add the following:
<gregkh> Permission is hereby granted for the distribution of this firmware
<gregkh> image as part of a Linux or other Open Source operating system kernel
<gregkh> in text or binary form as required.
<gregkh> as the keyspan drivers have.
So, problem resolved. No need to remove anything.
--
"Next the statesmen will invent cheap lies, putting the blame upon
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince
himself that the war is just, and will thank God for the better sleep
he enjoys after this process of grotesque self-deception." -- Mark Twain
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Don Armstrong 2004-06-16, 5:56 pm |
| On Wed, 16 Jun 2004, Matthew Wilcox wrote:
> The correct thing to do was what I did -- contact the USB maintainer, who
> was previously unaware of this problem:
>=20
> <gregkh> willy: cool, I got back a response with an email from emagic giv=
ing the blessing to include the drivers in linux.
> <gregkh> willy: just looks like I need to add the following:
> <gregkh> Permission is hereby granted for the distribution of this firmwa=
re=20
> <gregkh> image as part of a Linux or other Open Source operating =
system kernel=20
> <gregkh> in text or binary form as required.=20
> <gregkh> as the keyspan drivers have.
>=20
> So, problem resolved. No need to remove anything.
This appears to solve the immediate legality issue, but it still
doesn't resolve the DFSG issue.[1]
I assume that the text of the email will be included in
debian/copyright of the relevant package. Might be nice to respond to
one of these messages with it also.
Thanks again to whoever went and talked to emagic about this issue. If
possible, please consider asking them to specifically release the
source to this firmware under a license that is free according to the
DFSG as well.
Don Armstrong
1: Specifically, this grant of permision doesn't allow modification
(DFSG =A73) and dissallows specific classes of derived works (eg, those
that aren't a Linux or Open Source OS kernel.)
--=20
Our days are precious, but we gladly see them going
If in their place we find a thing more precious growing
A rare, exotic plant, our gardener's heart delighting
A child whom we are teaching, a booklet we are writing
-- Frederick R=FCkert _Wisdom of the Brahmans_=20
[Hermann Hesse _Glass Bead Game_]
http://www.donarmstrong.com
http://rzlab.ucr.edu
| |
| Eduard Bloch 2004-06-16, 5:56 pm |
| #include <hallo.h>
* Joe Wreschnig [Tue, Jun 15 2004, 09:01:52PM]:
>
> At best that solves a third of the problem. What about all the other
> copyright holders of the kernel, have they agreed to link with the
> non-GPLd code? (Before someone tells me kernel developers don't care, or
> "this isn't linking", [0]) And once the copyright issue is dealt with,
> there's still the issue of meeting the DFSG.
What exactly are you trying to proove with the mentioned link? The same
guesses about guesses about assumptions about expectations as we have on
debian-devel with some confused GPL fans (trying to construct a
DSFG-violation case for free GPLed data files).
Regards,
Eduard.
--
Ich denke, also verschwind ich.
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andrew Suffield 2004-06-16, 5:56 pm |
| On Wed, Jun 16, 2004 at 07:54:34AM +0200, Eduard Bloch wrote:
> #include <hallo.h>
> * Joe Wreschnig [Tue, Jun 15 2004, 09:01:52PM]:
>
>
> What exactly are you trying to proove with the mentioned link? The same
> guesses about guesses about assumptions about expectations as we have on
> debian-devel with some confused GPL fans (trying to construct a
> DSFG-violation case for free GPLed data files).
No, that we have a firm statement from a significant copyright holder
that they consider this to be a violation of the GPL, much like we've
had from people within Debian experienced in interpreting it.
People who are familiar with copyright law will understand the
significance of this.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Thiemo Seufer 2004-06-16, 5:56 pm |
| Joe Wreschnig wrote:
[snip]
>
> People who hold copyrights on the Linux kernel view distribution of the
> kernel with proprietary firmware to be a violation of their license.
> Period. This is a fact: _Copyright holders of material Debian is
> distributing believe we are doing so in violation of the license they
> have granted us_.
They can believe what they want. But for legal relevance they have
to show how exactly the firmware was derived from the rest of the
code (or vice versa). If they can't, it is merely a collection of
works.
[snip]
> Debian has a policy of trying to honor the copyright holder's
> interpretation of a license, both because we want to avoid lawsuits, and
> because (I like to think) we like to be nice to the developers whose
> software we distribute.
And this should as well respect the interpretation of the copyright
holder who put the firmware in his driver, instead of accusing him
to breach copyright law without proof.
Thiemo
| |
| Andrew Suffield 2004-06-16, 5:56 pm |
| On Wed, Jun 16, 2004 at 03:18:32PM +0200, Thiemo Seufer wrote:
> Joe Wreschnig wrote:
> [snip]
>
> They can believe what they want. But for legal relevance they have
> to show how exactly the firmware was derived from the rest of the
> code (or vice versa). If they can't, it is merely a collection of
> works.
Don't be absurd. Any resulting binary is obviously derived from both.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Matthew Wilcox 2004-06-16, 5:56 pm |
| On Tue, Jun 15, 2004 at 09:01:52PM -0500, Joe Wreschnig wrote:
> At best that solves a third of the problem.
It solves the problem at hand -- that Debian has no permission to
distribute the file. You can now go back to wanking about firmware all
you like. I shan't bother with that.
--
"Next the statesmen will invent cheap lies, putting the blame upon
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince
himself that the war is just, and will thank God for the better sleep
he enjoys after this process of grotesque self-deception." -- Mark Twain
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andrew Suffield 2004-06-16, 5:56 pm |
| On Wed, Jun 16, 2004 at 09:42:43AM -0400, Michael Poole wrote:
> Andrew Suffield writes:
>
>
> Why is that obvious?
>
> I have a binary on my bookshelf that is a combination of works with
> conflicting licenses. Is it a work derived from all of them, or is
> it just an unofficial Debian CD install?
In conventional unix nomenclature, a "binary" is an executable.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Andrew Suffield 2004-06-16, 5:56 pm |
| On Wed, Jun 16, 2004 at 02:56:00PM -0300, Humberto Massa wrote:
> @ 16/06/2004 14:31 : wrote Joe Wreschnig :
>
>
> No, no, and no.
> Firmware with _any_ distributable license + kernel (GPL) = distributable
> even if non-free.
> Firmware and Kernel are agregating only, not derived works. They don't
> link together; firmware is not a derived work of the kernel nor
> /vice-versa/.
"Link together" is completely irrelevant to "derivative work", except
that it sometimes happens at the same time. It is neither a necessary
nor a sufficient condition.
A ''derivative work'' is a work based upon one or more preexisting
works, such as a translation, musical arrangement, dramatization,
fictionalization, motion picture version, sound recording, art
reproduction, abridgment, condensation, or any other form in which a
work may be recast, transformed, or adapted. A work consisting of
editorial revisions, annotations, elaborations, or other
modifications which, as a whole, represent an original work of
authorship, is a ''derivative work''.
-- Section 101, Title 17, United States Code
EU copyright law is a little different, and does it like this:
...the translation, adaptation, arrangement and any other alteration
of a computer program and the reproduction of the results thereof,
without prejudice to the rights of the person who alters the program...
-- Article 4(b), Council Directive 91/250/EEC on the legal
protection of computer programs
The compiled kernel is almost certainly a derivative of the firmware
included in it. A good lawyer might be able to get you out of
this. Debian can *not* afford to assume that it would win such a case,
not least because of a lack of funding for good lawyers.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Don Armstrong 2004-06-16, 5:56 pm |
| [Humberto: Can you please fix your MUA so that it provides
In-Reply-To:, References:, and doesn't break threads?]
On Wed, 16 Jun 2004, Humberto Massa wrote:
> Firmware with _any_ distributable license + kernel (GPL) =
> distributable even if non-free.
This is not clear a priori.
> Firmware and Kernel are agregating only, not derived works. They
> don't link together; firmware is not a derived work of the kernel
> nor /vice-versa/.
The kernel+firmware work is a derived work of the kernel work and the
firmware work. I'll assume that you're not arguing that it's not. The
question is whether the GPL's exception for mere aggregates applies to
such a work.
Since that exception only seems to apply to "mere aggregation" on a
storage medium, it's quite likely that it doesn't apply at all.
In cases like these, unless you can provide clear and convincing
arguments that this is "mere aggregation" (which should go to -legal)
we should assume that the combined, derivative work cannot be
distributed unless both works can satisfy the DFSG.
Don Armstrong
--
DIE!
-- Maritza Campos http://www.crfh.net/d/20020601.html
http://www.donarmstrong.com
http://rzlab.ucr.edu
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| viro@parcelfarce.linux.theplanet.co.uk 2004-06-16, 5:56 pm |
| On Wed, Jun 16, 2004 at 03:21:38PM -0500, Joe Wreschnig wrote:
[firmware as mere aggregation]
> Kernel copyright holders think otherwise, as do many other people.
Out of curiosity, could you please show an email from such copyright
holder (with some references to the code in kernel contributed by
that person) that would say so?
It's not impossible that such a thing exists; however, I've heard the
quoted statement a *LOT* and so far it always turned out either "FOAF
heard about that somewhere" or "well, there's got to be at least one!".
Care to show evidence for your statement? alt.folklore.urban rules,
please.
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Thiemo Seufer 2004-06-17, 5:53 pm |
| Joe Wreschnig wrote:
[snip]
> When you compile a kernel, the firmware is included in it. When you
> distribute that compiled binary, you're distributing a work derived from
> the kernel and the firmware. This is not a claim that the firmware is a
> derivative of the Linux kernel, or vice versa. Rather, the compiled
> binary is a derivative of both.
>
> For someone to claim that data compiled into a program but not executed
> is "mere aggregation" is nonsense. Is a program that prints the source
> code to GNU ls (stored as a string constant in the program, not an
> external file) a derivative of GNU ls? Of course it is. This is
> *exactly* analogous to the situation with firmware.
Could you please explain how exactly the derivation works in this case?
And please bring forward some more convincing arguments than "this is
nonsense", "this is obvious", or some broken analogy.
Thiemo
| |
| Raul Miller 2004-06-17, 5:53 pm |
| > Joe Wreschnig wrote:
On Thu, Jun 17, 2004 at 01:48:08AM +0200, Thiemo Seufer wrote:[vbcol=seagreen]
> Could you please explain how exactly the derivation works in this case?
> And please bring forward some more convincing arguments than "this is
> nonsense", "this is obvious", or some broken analogy.
It's a compilation work.
[Some people might think that "compilation" and "aggregation" are the
same thing -- but the GPL goes to great lengths to specify that it does
apply where the compilation is a program and not where the compilation
is not a program.]
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andrew Suffield 2004-06-17, 5:53 pm |
| On Wed, Jun 16, 2004 at 04:49:27PM -0400, Michael Poole wrote:
> Andrew Suffield writes:
>
>
> Anyone who wanted to go after Debian would have a hard time of it:
> They would have to prove not only that the combined kernel is a
> derived work, but also that there were no binary blobs in the kernel
> when they started contributing and that they still have standing to
> make a claim.
That probably wouldn't hold. They merely have to demonstrate that they
were not aware that this was a license violation - that's easy, they
can do it on their own testimony.
> Estoppel would bar a claim if the plaintiff first
> contributed code to a kernel that already had binary blob components.
> A merely decent lawyer may be able to invoke laches depending on how
> long an author was silent after the first binary blob was added to the
> kernel, but merely decent lawyers are not much cheaper than good
> lawyers.
These are stock defenses that are invoked as a matter of course, in
case something useful turns up during discovery, and which are
notoriously difficult to prove. You need to prove *malicious*
*intent*, not incompetence, for these defences to work. Not a good
idea to bet on them.
> By way of example, the DGRS driver firmware was added to Linux
> sometime before linux-2.0 was released in June 1996. It stayed more
> than seven years. That kind of delay in discovering alleged rights
> reminds me more of a certain Lindon, Utah-based company than of
> serious developers, and smacks of revisionist history.
We have not always been paying such close attention as we currently
do. Anybody on -legal can tell you that much. This is partially
because we're getting better at it, and partially because SCO has
demonstrated that it's important we be good at it.
Also, this sort of delay is far from unusual in this sort of
case. That's why the statute of limitations exists (we're nowhere near
the limit here; in the US it's three years from the more recent
infringement for civil copyright cases, five for criminal ones).
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Andrew Suffield 2004-06-17, 5:53 pm |
| On Wed, Jun 16, 2004 at 06:18:14PM -0400, Michael Poole wrote:
> A little Google shows that Yggdrasil has made such an argument:
> http://lists.debian.org/debian-lega...4/msg00130.html
>
> Unfortunately for Mr. Richter, Linux does not seem to contain any
> copyright notices attributable to him or Yggdrasil before 2000. As I
> cited elsewhere, this is at least FOUR YEARS after firmware was
> included in the kernel, so he cannot fairly claim infringement. He
> should have known that binary firmware existed in the kernel before.
(This is wrong, see elsewhere in the thread for details)
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Raul Miller 2004-06-17, 5:53 pm |
| On Wed, Jun 16, 2004 at 09:11:32PM -0400, Michael Poole wrote:
> I think you are confusing language. When the GPL talks about the
> Program, it refers to "any program or other work" licensed under the
> GPL; see section 0. It deals with collective (in contrast to
> derivative) works in just two paragraphs: one exempts "mere
> aggregation" from coverage; the other seems explanatory rather than
> normative. Calling that "great lengths" is a little deceptive.
The GPL only uses the word "collective" in one sentence, not two:
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is
to exercise the right to control the distribution of derivative or
collective works based on the Program.
However, this sentence makes clear that "works based on the Program"
is meant to include both derivative works based on the Program and
collective works based on the Program.
So what is this "deception" you're talking about?
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-17, 5:53 pm |
| > > However, this sentence makes clear that "works based on the Program"
On Wed, Jun 16, 2004 at 11:12:37PM -0400, Michael Poole wrote:[vbcol=seagreen]
> 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.
>
> This sentence deals just as much with collective works, and makes
> clear that "mere aggregation" is not sufficient to invoke GPL coverage
> of the other work(s). What is your point?
My point is that any sentence talking about "a work based on the Program"
is by default talking about both derivative and collective works.
>
> The deception is calling it "great lengths." When I said the GPL
> "deals with collective works in just two paragraphs" you focused on
> the one where they are mentioned by name and entirely ignored the
> other (because you don't like what it says?).
You seem to be ignoring everything the GPL says about "works based on
the Program".
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andreas Barth 2004-06-17, 5:53 pm |
| * Joe Wreschnig (piman@debian.org) [040616 22:25]:
> Kernel copyright holders think otherwise, as do many other people.
There is a company that claims that itself is the copyright holder of
some Unix sources, and that thinks that use of that concepts is a
breach of copyright. Should we accept that?
Cheers,
Andi
--
http://home.arcor.de/andreas-barth/
PGP 1024/89FB5CE5 DC F1 85 6D A6 45 9C 0F 3B BE F1 D0 C5 D1 D9 0C
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andrew Suffield 2004-06-17, 5:53 pm |
| On Wed, Jun 16, 2004 at 10:36:11PM -0400, Michael Poole wrote:
> Andrew Suffield writes:
>
>
> Incompetence (or laziness) on the part of the plaintiff is a perfectly
> adequate reason to invoke either of those defenses. Until you cite
> specific case law, I will disbelieve your claim that proof of intent
> is necessary, since Google finds dismissals on those grounds that
> never mention malicious intent.
Judge's discretion, probably coupled with bad lawyering on the part of
the prosecution.
>
> As of the last filings and rulings I saw, SCO has only demonstrated
> that someone with more money than sense can cost other people a lot in
> legal fees. This is not really news, and as a stock manipulation
> scheme, it is proving to not have legs. If someone else down the road
> wants to involve free software in another frivolous lawsuit, no amount
> of diligence on Debian's part will prevent it.
This is merely a reflection that SCO does not appear to have a
case. It does not change the fact that SCO has demonstrated why it is
important that they must never have a case.
> Yes, Debian should make sure it will not end up on the wrong end of a
> valid lawsuit, but that can be satisfied by other means than claiming
> that widely used packages violate the GPL.
"Yes, Debian should make sure it will not end up on the wrong end of a
valid lawsuit, but that can be satisfied by other means than avoiding
breaking the law"
Umm... how?
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Thiemo Seufer 2004-06-17, 5:53 pm |
| Raul Miller wrote:
>
> On Thu, Jun 17, 2004 at 01:48:08AM +0200, Thiemo Seufer wrote:
>
> It's a compilation work.
Fine. The copyright for the compilation lies by the one who did the
compilation. This is Linus Torvalds, I guess.
Thiemo
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Thiemo Seufer 2004-06-17, 5:53 pm |
| Humberto Massa wrote:
[snip]
>
> not here in BR. Or at least not in the way you _seem_ to be implying.
I referred only to the inclusion of some firmware in the kernel.
[snip: lengthy explanation]
> Easy, huh? extrapolate for ten+ years of patching and aggregating and
> you'll get where we are today.
The firmware typically wasn't patched, and nothing is derived from it.
Thiemo
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Andrew Suffield 2004-06-17, 5:53 pm |
| Troll.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Andrew Suffield 2004-06-17, 5:53 pm |
| On Thu, Jun 17, 2004 at 01:57:05PM +0100, Oliver Elphick wrote:
> On Thu, 2004-06-17 at 13:31, Brian Thomas Sniffen wrote:
>
> Unless someone else put it into the kernel without asking him (was that
> the case?), he gave it to them under the GPL with a request to include
> it in the kernel as it then was. That creates an implied permission to
> do so, since otherwise his request to have it put in the kernel would be
> nonsensical.
There is no such thing as "implied permission" in copyright. Anything
which is not expressly granted is forbidden. That's fundamental.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -><- |
| |
| Raul Miller 2004-06-17, 5:53 pm |
| > Raul Miller writes:
>
On Wed, Jun 16, 2004 at 11:47:48PM -0400, Michael Poole wrote:[vbcol=seagreen]
> Maybe you don't realize it, but a "mere aggregation" is also a work
> based on the Program: it is a collective work.
I do realize that "'mere aggregation' works" are a subset of
"collective works".
> Exactly how many collective works are covered by that phrase is open
> to debate, but you cannot ignore the GPL's treatment of such works.
You have no basis for claiming I'm ignoring this treatment.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-17, 5:53 pm |
| On Thu, Jun 17, 2004 at 12:24:29PM -0300, Humberto Massa wrote:
> No way. The clause #0 of the GPL is crystal clear: << a "work based on
> the Program" means either the Program or any derivative work under
> copyright law >> DERIVATIVE. Under copyright law.
>
> _Not_ collective/compilation/anthology.
False dichotomy.
There's nothing preventing a collective work from being a
derivative work.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-17, 5:53 pm |
| > >False dichotomy.
On Thu, Jun 17, 2004 at 03:24:23PM -0300, Humberto Massa wrote:[vbcol=seagreen]
> No, Raul. The law. USC17, BR copyright law, and probably every copyright
> law following the Geneva convention *does* such a distinction.
I did not say that there was no distinction.
I did say that there was no dichotomy.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-17, 5:53 pm |
| On Thu, Jun 17, 2004 at 03:46:14PM -0300, Humberto Massa wrote:
> But there is. You see, in Law, when you enumerate things, you are
> separating things. (dichotomy = two separated in Greek)
I'm writing in english, not greek.
If you think there is some legally relevant document which means that a
collective work can't be a derivative work (for example, if you think that
an anthology can't be a derivative work based on the contained stories,
or that a subsequent edition of that anthology can't be a derivative
work of an earlier edition), please cite that specific document.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-17, 5:53 pm |
| > > If you think there is some legally relevant document which means that a
....
On Thu, Jun 17, 2004 at 04:41:42PM -0400, Michael Poole wrote:[vbcol=seagreen]
> http://digital-law-online.info/lpdi1.0/treatise6.html discusses the
> differences between derivative works and compilations, and quotes a
> congressional report that elaborates:
....
> See also http://www.copyright.gov/circs/circ14.html, which remarks
> both that the whole of the derivative work must represent an original
> work of authorship, rather than an arrangement of distinct works, and
> that mechanical (non-creative, ergo non-copyrightable) transformation
> of the original does not make a derivative.
Ok, this is good -- I did not know that.
However -- by this definition, the linux kernel is very definitely a
derivative work, and the firmware is content which has been incorporated
into the kernel.
According to what you just cited, the concept of a collective work
doesn't enter into the picture at all.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-17, 11:50 pm |
| On Thu, Jun 17, 2004 at 06:05:06PM -0400, Michael Poole wrote:
> The kernel (I assume as a whole) is a derivative work of what?
Earlier versions of the kernel.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Troy Benjegerdes 2004-06-17, 11:50 pm |
| On Wed, Jun 16, 2004 at 03:21:18AM -0500, Joe Wreschnig wrote:
> On Wed, 2004-06-16 at 00:54, Eduard Bloch wrote:
>
> People who hold copyrights on the Linux kernel view distribution of the
> kernel with proprietary firmware to be a violation of their license.
> Period. This is a fact: _Copyright holders of material Debian is
> distributing believe we are doing so in violation of the license they
> have granted us_. This is a serious problem, both in terms of BTS
> severity and most legal codes.
Who actually objects to this that has contributed signifigant code to
the kernel? Is there a list, and what they have contributed?
>
> Debian has a policy of trying to honor the copyright holder's
> interpretation of a license, both because we want to avoid lawsuits, and
> because (I like to think) we like to be nice to the developers whose
> software we distribute. In some really blatantly stupid cases like SCO,
> we rebut them. But I don't think Adam's case is stupid, and you're
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Christoph Hellwig 2004-06-17, 11:50 pm |
| On Thu, Jun 17, 2004 at 06:46:26PM -0500, Troy Benjegerdes wrote:
>
> Who actually objects to this that has contributed signifigant code to
> the kernel? Is there a list, and what they have contributed?
Adam Richter and me for example. Note that these are both in context of
files that have a GPL-incomaptible license boilerplate on them and not
the usual firmware wankfests.
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Matthew Palmer 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 09:04:18AM -0300, Humberto Massa wrote:
> Repeating, trying to summarize: the current version of the Linux kernel
> is a derivative work of its earlier versions, and an anthology work of
> its separated autonomous parts. Those parts, in principle, would be each
> and every patch that entered the kernel and stayed there.
I would imagine that a lot of the patches in the kernel are derivative works
of the kernel, besides. This is, I would imagine, the major difference
between the kernel and a "standard" anthology.
- Matt
--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-20, 10:24 pm |
| On Thu, Jun 17, 2004 at 04:41:42PM -0400, Michael Poole wrote:
> This is not the way the law works. The presumption is not "this work
> is a derivative work because Raul Miller claims it is." Humberto has
> cited reasons why the kernel tarball (or binary images) should be
> considered a compilation rather than a derivative work. You have only
> claimed that they should be a derivative work, without referring to a
> "why."
I just took a closer look at your cites
http://digital-law-online.info/lpdi1.0/treatise6.html
http://www.copyright.gov/circs/circ14.html
the digital-law-online page says: "There is necessarily some overlapping
between the two". You quoted this yourself.
The copyright.gov page doesn't really say anything that would make me
think differently.
In other words, it's still not a dichotomy. I was too hasty when I
wrote my message saying that it was.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Matthew Wilcox 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 09:02:25AM -0400, Brian Thomas Sniffen wrote:
> I would be much more convinced if I saw an argument from the
> GPL-incompatible-firmware-is-OK side as to why the GPL prohibits
> distributing linkages of GPL'd and GPL-incompatible code.
The interpretation favoured by kernel hackers is that anything that runs
on the host CPU is part of the program, and anything that runs on the
card is just data for the program to operate on.
--
"Next the statesmen will invent cheap lies, putting the blame upon
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince
himself that the war is just, and will thank God for the better sleep
he enjoys after this process of grotesque self-deception." -- Mark Twain
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 02:46:22PM +0100, Matthew Wilcox wrote:
> The interpretation favoured by kernel hackers is that anything that runs
> on the host CPU is part of the program, and anything that runs on the
> card is just data for the program to operate on.
This distinction isn't relevant when discussing copyright. Most
copyrights are copyrights on data.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 10:55:47AM -0300, Humberto Massa wrote:
> What rights do the GPL'd software recipient have? The GPL grants
> some rights not granted by copyrights law. I made an extensive
> document and posted it to d-l, but no-one seemed to listen or to
> understand. All ok. IRT making derived works, the recipient has the
> right of making *some* derived works (respecting 2a and 2c) and to
> redistribute those derived works under the terms of the GPL itself.
> It seems not to permit anthology (collective) works, until you see
> the "mere aggregation" clause (section 2 third paragraph) /which/
> appears to cover anthology works.
That clause only deals with some anthology works, not all. It's an
exception to <<a "work based on the Program" means either the Program or
any derivative work under copyright law: that is to say, a work containing
the Program or a portion of it, either verbatim or with modifications...>>
It's pretty clear that the linux kernel is not a mere aggregation of
works on some volume of storage.
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| William Lee Irwin III 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 11:35:43AM -0400, Raul Miller wrote:
> That clause only deals with some anthology works, not all. It's an
> exception to <<a "work based on the Program" means either the Program or
> any derivative work under copyright law: that is to say, a work containing
> the Program or a portion of it, either verbatim or with modifications...>>
> It's pretty clear that the linux kernel is not a mere aggregation of
> works on some volume of storage.
Any chance you guys could come to some kind of consensus on this so I
know what has to be done for the debian package?
Thanks.
-- wli
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Thiemo Seufer 2004-06-20, 10:24 pm |
| Brian Thomas Sniffen wrote:
> Thiemo Seufer <ica2_ts@csv.ica.uni-stuttgart.de> writes:
>
> Isn't the kernel containing the firmware derivative of it?
AFAICS it contains not a derivative in the legal sense but the
original in a different representation.
> If not,
> why can't I put some GPL-incompatible x86 code into the kernel, load
> it into a device in my system -- the main memory -- and then issue a
> command to the processor to execute it?
You surely can, nothing prevents you from doing so. Distributing that
combination is the problematic part.
> That is, doesn't your interpretation allow arbitrary linking of GPL'd
> programs?
>From the copyright law POV, you most likely need to derive some call
wrapper for this, which needs some compatible licence for further
distribution. For an requirement to GPL the binary you have to prove
it includes important parts of/ideas from a GPL'ed work.
[shuffle]
> as to why the GPL prohibits
> distributing linkages of GPL'd and GPL-incompatible code.
It doesn't. If some work includes a GPL'ed work and is distributed,
then the whole work must be GPL compatible. This doesn't extend to a
collection of works.
For some binary driver, you have to show why a bunch of kernel modules
is included in the kernel, and why the fact you can drop them easily
from a compilation doesn't reinforce the idea it's a collection of
works.
For some firmware, you have to explain how this provision becomes
relevant, that is why the definition of "source" makes sense for it,
why the presence of this firmware constitutes an inclusion when it's
only use is to be loaded in a separate device, and how exactly
moving it to a userland file makes a difference.
> I would be much more convinced if I saw an argument from the
> GPL-incompatible-firmware-is-OK side
I don't say "GPL incompatible" firmware is OK. I say I can't prove it's
_not_ OK, while its copyright holder claims it is. Without proof to the
contrary, I'd rather follow his idea of what his work actually is.
Thiemo
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| William Lee Irwin III 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 04:50:08PM +0000, Brian M. Carlson wrote:
> If it's undistributable, it obviously doesn't belong in main. So please
> remove the undistributable stuff. Second, if it's non-free, it doesn't
> belong in the kernel, which is in main. So remove anything that is
> non-free from the kernel-source. It's really not rocket science, and I
> don't know why people insist upon talking about collective versus
> derivative works, because none of this stuff belongs in main anyway.
> If you need help in determining whether something is free or not, please send
> a message to debian-legal (preferably in a new thread) asking that question.
I have a blacklist of things to remove already I'm stuck carrying and
continuing to remove from what we call "pristine" upstream sources, but
I didn't see any specific suggestion for additions to it or removals
from it. So at the moment I don't have anything to go on as far as what
this thread tells me to do. I'm bothered somewhat by the fact that this
all sounds alarming, but doesn't seem to have anything specific to be
done about it, which is why I asked the participants in this discussion
to try to come to some kind of actual conclusion I could act on.
-- wli
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| William Lee Irwin III 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 12:34:24PM -0500, Joe Wreschnig wrote:
> The current release policy says that all firmware not licensed under
> GPL-compatible licenses needs to be removed. It also says that any
> sourceless firmware needs to be removed.
> http://lists.debian.org/debian-kern...6/msg00064.html
> Depending on the outcome of GRs, it might change so that only the former
> class has to be removed.
> Either way, firmware not licensed under a GPL-compatible license needs
> to be removed.
I could use some clarification as to which of those GR's are pending
and which are already requirements.
Also, this discussion seems to be more far-reaching than that, so it's
not clear to me what the connection is between this discussion and
those GR's or that the conclusion is limited to debian GR's. Some kind
of jointly-written "summarized findings" or similar thing may help me
decide what to do.
I'm getting a different story from every single person I talk to, so
something resembling an authoritative answer would be very helpful.
-- wli
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| Raul Miller 2004-06-20, 10:24 pm |
| On Fri, Jun 18, 2004 at 10:47:50AM -0700, William Lee Irwin III wrote:
> I'm getting a different story from every single person I talk to, so
> something resembling an authoritative answer would be very helpful.
The current GR on debian-vote attempts to resolve some of these
issues.
FYI,
--
Raul
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
| |
| William Lee Irwin III 2004-06-20, 10:24 pm |
| William Lee Irwin III writes:
On Fri, Jun 18, 2004 at 01:55:34PM -0400, Michael Poole wrote:[vbcol=seagreen]
> For Debian's purposes, I believe that Joe's summary is correct: DFSG
> requires that anything without source be removed. As far as I know,
> that covers all the firmware under discussion. I don't have a list of
> affected files/drivers.
> I believe that chunks with licenses that appear GPL-incompatible are
> being worked on upstream (Greg K-H made a posting to linux-kernel
> earlier this week mentioning the particular case that spawned this
> thread), so I hope his first set becomes empty.
> The broader discussion is not whether Debian should remove binary
> firmware, but whether the upstream kernel has to remove it too.
Now can I get more than 1 person to agree on this? The trouble is not
what the conclusion is, but rather, that everyone has their own personal
conclusion they communicate to me, and none of them resemble each other.
-- wli
--
To UNSUBSCRIBE, email to debian-legal-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
|
|
|
|
|