GPL violation by CorAccess?

GPL violation by CorAccess?

[Karel Kulhavy]

Hello

I have seen a device by CorAccess which apparently uses Linux and didn't find
anything that would suggest it complies to GPL, though I had access to the
complete shipping package. Does anyone know about known cause of violation by
this company or should I investigate further?

CL<

Re: GPL violation by CorAccess?

[Lennart Sorensen]

On Tue, Apr 19, 2005 at 05:57:43PM +0000, Karel Kulhavy wrote:
> I have seen a device by CorAccess which apparently uses Linux and didn't find
> anything that would suggest it complies to GPL, though I had access to the
> complete shipping package. Does anyone know about known cause of violation by
> this company or should I investigate further?

Well what is the case if you use unmodified GPL code, do you still have
to provide sources to the end user if you give them binaries?  I would
guess yes, but IANAL.

As far as I can tell their system is a geode GX1 so runs standard x86
software.  Maybe they didn't have to modify any of the linux kernel to
run what they needed.  Their applications are their business of course.
It looks like they use QT as the gui toolkit, which I don't off hand
know the current license conditions of.  Then there is the web browser
and such, which has it's own license conditions.  Of course for all I
know their user manual has an offer of sending a CD with the sources if
you ask.  Does anyone actually have their product that could check for
that?

Len Sorensen

Re: GPL violation by CorAccess?

[Charles Cazabon]

Lennart Sorensen <lsorense@csclub.uwaterloo.ca> wrote:
> 
> Well what is the case if you use unmodified GPL code, do you still have
> to provide sources to the end user if you give them binaries? 

Yes, or a written offer to provide sources, plus a copy of the GPL.  It's all
spelled out pretty clearly in the GPL itself.

Charles
-- 
-----------------------------------------------------------------------
Charles Cazabon                            <linux@discworld.dyndns.org>
GPL'ed software available at:               http://pyropus.ca/software/
-----------------------------------------------------------------------

Re: GPL violation by CorAccess?

[Chris Friesen]

Charles Cazabon wrote:
> Lennart Sorensen <lsorense@csclub.uwaterloo.ca> wrote:
> 
>>Well what is the case if you use unmodified GPL code, do you still have
>>to provide sources to the end user if you give them binaries? 
> 
> 
> Yes, or a written offer to provide sources, plus a copy of the GPL.  It's all
> spelled out pretty clearly in the GPL itself.

Or a nonprofit organization can simply point to the original 
source...but that doesn't apply in this case.

Chris

Re: GPL violation by CorAccess?

[Richard B. Johnson]

Violation? They proudly reply in their article in
 	http://www.linuxdevices.com
that they use Linux, that they embedded a version
of Red Hat, etc.

It's likely that they didn't modify anything in the kernel and
just used some stripped-down C-libraries to make everything fit.

On Tue, 19 Apr 2005, Lennart Sorensen wrote:

> On Tue, Apr 19, 2005 at 05:57:43PM +0000, Karel Kulhavy wrote:
>> I have seen a device by CorAccess which apparently uses Linux and didn't find
>> anything that would suggest it complies to GPL, though I had access to the
>> complete shipping package. Does anyone know about known cause of violation by
>> this company or should I investigate further?
>
> Well what is the case if you use unmodified GPL code, do you still have
> to provide sources to the end user if you give them binaries?  I would
> guess yes, but IANAL.
>
> As far as I can tell their system is a geode GX1 so runs standard x86
> software.  Maybe they didn't have to modify any of the linux kernel to
> run what they needed.  Their applications are their business of course.
> It looks like they use QT as the gui toolkit, which I don't off hand
> know the current license conditions of.  Then there is the web browser
> and such, which has it's own license conditions.  Of course for all I
> know their user manual has an offer of sending a CD with the sources if
> you ask.  Does anyone actually have their product that could check for
> that?
>
> Len Sorensen
> -
> 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,
Dick Johnson
Penguin : Linux version 2.6.11 on an i686 machine (5537.79 BogoMips).
  Notice : All mail here is now cached for review by Dictator Bush.
                  98.36% of all statistics are fiction.

Re: GPL violation by CorAccess?

[Chris Friesen]

Richard B. Johnson wrote:
> 
> Violation? They proudly reply in their article in
>     http://www.linuxdevices.com
> that they use Linux, that they embedded a version
> of Red Hat, etc.
> 
> It's likely that they didn't modify anything in the kernel and
> just used some stripped-down C-libraries to make everything fit.

Right.  They're distributing products licensed under the GPL, so 
sections 3-a and 3-b of the GPL apply.

Thus they can either a) accompany it with the source code, or b) 
accompany it with a written offer to give any third party a copy of the 
source code.

Chris

Re: GPL violation by CorAccess?

[Richard B. Johnson]

On Tue, 19 Apr 2005, Chris Friesen wrote:

> Richard B. Johnson wrote:
>>
>> Violation? They proudly reply in their article in
>>     http://www.linuxdevices.com
>> that they use Linux, that they embedded a version
>> of Red Hat, etc.
>>
>> It's likely that they didn't modify anything in the kernel and
>> just used some stripped-down C-libraries to make everything fit.
>
> Right.  They're distributing products licensed under the GPL, so
> sections 3-a and 3-b of the GPL apply.
>
> Thus they can either a) accompany it with the source code, or b)
> accompany it with a written offer to give any third party a copy of the
> source code.
>
> Chris

No. Accompany it with a written offer to __provide__ the source
code for any GPL stuff they used (like the kernel or drivers).
Anything at the application-level is NOT covered by the GPL.
They do not have to give away their trade-secrets.

Cheers,
Dick Johnson
Penguin : Linux version 2.6.11 on an i686 machine (5537.79 BogoMips).
  Notice : All mail here is now cached for review by Dictator Bush.
                  98.36% of all statistics are fiction.

Re: GPL violation by CorAccess?

[Chris Friesen]

Richard B. Johnson wrote:

> No. Accompany it with a written offer to __provide__ the source
> code for any GPL stuff they used (like the kernel or drivers).
> Anything at the application-level is NOT covered by the GPL.
> They do not have to give away their trade-secrets.

GPL'd applications would still be covered by the GPL, no?

If I buy their product, I should be able to ask them for the source to 
all GPL'd entities that are present in the system, including the kernel, 
drivers, and all GPL'd userspace apps.

Any *new* apps that they wrote they would of course be free to keep private.

Re: GPL violation by CorAccess?

[Richard B. Johnson]

On Tue, 19 Apr 2005, Chris Friesen wrote:

> Richard B. Johnson wrote:
>
>> No. Accompany it with a written offer to __provide__ the source
>> code for any GPL stuff they used (like the kernel or drivers).
>> Anything at the application-level is NOT covered by the GPL.
>> They do not have to give away their trade-secrets.
>
> GPL'd applications would still be covered by the GPL, no?
>

You mean like `ls` and `init` ??? Sure. I don't think any serious
embedded stuff would use that, though. Typically an embedded
system would start with a new application called 'init'. It
wouldn't use a SYS-V startup and certainly wouldn't have a shell.
The new init would do everything including mounting any file-
systems and initializing networking all by itself without
any help from the usual applications. It might fork-off a few
different tasks to handle different things. For instance,
the system shown probably handles the furnace and air-conditioner
as a separate task. The shades and blinds are probably another
and, certainly, communicating with the robot that mows the lawn
would require a separate task just to handle GPS.

> If I buy their product, I should be able to ask them for the source to
> all GPL'd entities that are present in the system, including the kernel,
> drivers, and all GPL'd userspace apps.
>
> Any *new* apps that they wrote they would of course be free to keep private.
>

Yep.


Cheers,
Dick Johnson
Penguin : Linux version 2.6.11 on an i686 machine (5537.79 BogoMips).
  Notice : All mail here is now cached for review by Dictator Bush.
                  98.36% of all statistics are fiction.

Re: GPL violation by CorAccess?

[Bernd Petrovitsch]

On Tue, 2005-04-19 at 17:37 -0600, Chris Friesen wrote:
> Richard B. Johnson wrote:
> 
> > No. Accompany it with a written offer to __provide__ the source
> > code for any GPL stuff they used (like the kernel or drivers).
> > Anything at the application-level is NOT covered by the GPL.

That depends on the software used there.

> > They do not have to give away their trade-secrets.

Unless they coded them into GPL software ...

> GPL'd applications would still be covered by the GPL, no?

Good question: Strictly speaking if you omit the GPL in the delivered
ssoftware/product/whatever, you violated the GPL yourself and - thus -
loose all rights which are "given" to you through the GPL.

> If I buy their product, I should be able to ask them for the source to 
> all GPL'd entities that are present in the system, including the kernel, 
> drivers, and all GPL'd userspace apps.

ACK.

> Any *new* apps that they wrote they would of course be free to keep private.

As long as they do not statically link against LGPL (or GPL) code and as
long as they do not link dynamically agaist GPL code. And there are
probably more rules .....

	Bernd
-- 
Firmix Software GmbH                   http://www.firmix.at/
mobil: +43 664 4416156                 fax: +43 1 7890849-55
          Embedded Linux Development and Services

Re: GPL violation by CorAccess?

[Steven Rostedt]

On Wed, 2005-04-20 at 09:30 +0200, Bernd Petrovitsch wrote:

> 
> As long as they do not statically link against LGPL (or GPL) code and as
> long as they do not link dynamically agaist GPL code. And there are
> probably more rules .....
> 

Actually, I believe that the LGPL allows for static linking as well. As
long as you only interact with the library through the defined API, it
is OK.

>From the LGPL preamble:

 The precise terms and conditions for copying, distribution and
modification follow.  Pay close attention to the difference between a
"work based on the library" and a "work that uses the library".  The
former contains code derived from the library, whereas the latter must
be combined with the library in order to run.


Point number 5 of TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION:

 5. A program that contains no derivative of any portion of the
Library, but is designed to work with the Library by being compiled or
linked with it, is called a "work that uses the Library".  Such a
work, in isolation, is not a derivative work of the Library, and
therefore falls outside the scope of this License.


So, I would say that the LGPL _does_ allow statically linked to non GPL
work.


-- Steve

Re: GPL violation by CorAccess?

[Arjan van de Ven]

On Wed, 2005-04-20 at 08:49 -0400, Steven Rostedt wrote:
> On Wed, 2005-04-20 at 09:30 +0200, Bernd Petrovitsch wrote:
> 
> > 
> > As long as they do not statically link against LGPL (or GPL) code and as
> > long as they do not link dynamically agaist GPL code. And there are
> > probably more rules .....
> > 
> 
> Actually, I believe that the LGPL allows for static linking as well.

it does, as long as you provide the .o files of your own stuff so that
the end user can relink with  say a bugfixed version of library.

Re: GPL violation by CorAccess?

[Steven Rostedt]

On Wed, 2005-04-20 at 14:57 +0200, Arjan van de Ven wrote:
> On Wed, 2005-04-20 at 08:49 -0400, Steven Rostedt wrote:
> > On Wed, 2005-04-20 at 09:30 +0200, Bernd Petrovitsch wrote:
> > 
> > > 
> > > As long as they do not statically link against LGPL (or GPL) code and as
> > > long as they do not link dynamically agaist GPL code. And there are
> > > probably more rules .....
> > > 
> > 
> > Actually, I believe that the LGPL allows for static linking as well.
> 
> it does, as long as you provide the .o files of your own stuff so that
> the end user can relink with  say a bugfixed version of library.

I don't see that in the license.  As point 5 showed: "Such a
work, in isolation, is not a derivative work of the Library, and
therefore falls outside the scope of this License." So you don't need to
do anything more than supply the source of the LPGL work. In fact, it
may not be a good idea to add a bugfixed version of the libary without
going through the vendor. You don't know if the application that uses
this depends on the side effects of the bug.

-- Steve

Re: GPL violation by CorAccess?

[Michael Poole]

Steven Rostedt writes:

> On Wed, 2005-04-20 at 14:57 +0200, Arjan van de Ven wrote:
>> On Wed, 2005-04-20 at 08:49 -0400, Steven Rostedt wrote:
>> > On Wed, 2005-04-20 at 09:30 +0200, Bernd Petrovitsch wrote:
>> > 
>> > > 
>> > > As long as they do not statically link against LGPL (or GPL) code and as
>> > > long as they do not link dynamically agaist GPL code. And there are
>> > > probably more rules .....
>> > > 
>> > 
>> > Actually, I believe that the LGPL allows for static linking as well.
>> 
>> it does, as long as you provide the .o files of your own stuff so that
>> the end user can relink with  say a bugfixed version of library.
>
> I don't see that in the license.  As point 5 showed: "Such a
> work, in isolation, is not a derivative work of the Library, and
> therefore falls outside the scope of this License."

"Such a work" refers to "A program that contains no derivative of any
portion of the library."  A program that is statically linked against
the library clearly contains part or all of the library, and cannot
qualify for the lower threshold of section 5.  Section 5 is talking
about late binding to the library; dynamic linking is one example.

For programs distributed as object code that does contain part of the
library, the distributor must -- sooner or later -- comply with 6(a)
(allow the user to relink) or 6(b) (use dynamic linking).

Michael Poole

Re: GPL violation by CorAccess?

[Arjan van de Ven]

On Wed, 2005-04-20 at 09:07 -0400, Steven Rostedt wrote:
> On Wed, 2005-04-20 at 14:57 +0200, Arjan van de Ven wrote:
> > On Wed, 2005-04-20 at 08:49 -0400, Steven Rostedt wrote:
> > > On Wed, 2005-04-20 at 09:30 +0200, Bernd Petrovitsch wrote:
> > > 
> > > > 
> > > > As long as they do not statically link against LGPL (or GPL) code and as
> > > > long as they do not link dynamically agaist GPL code. And there are
> > > > probably more rules .....
> > > > 
> > > 
> > > Actually, I believe that the LGPL allows for static linking as well.
> > 
> > it does, as long as you provide the .o files of your own stuff so that
> > the end user can relink with  say a bugfixed version of library.
> 
> I don't see that in the license.  As point 5 showed: "Such a
> work, in isolation, is not a derivative work of the Library, and

you missed the point "in isolation". If you do NOT link against the lib,
eg your app in isolation, you don't have to care abuot the LGPL. That is
what it says. The moment you do link you are no longer "in isolation".

Re: GPL violation by CorAccess?

[Pavel Machek]

Hi!

> > I have seen a device by CorAccess which apparently uses Linux and didn't find
> > anything that would suggest it complies to GPL, though I had access to the
> > complete shipping package. Does anyone know about known cause of violation by
> > this company or should I investigate further?
> 
> Well what is the case if you use unmodified GPL code, do you still have
> to provide sources to the end user if you give them binaries?  I would
> guess yes, but IANAL.
> 
> As far as I can tell their system is a geode GX1 so runs standard x86
> software.  Maybe they didn't have to modify any of the linux kernel to
> run what they needed.  Their applications are their business of course.
> It looks like they use QT as the gui toolkit, which I don't off hand
> know the current license conditions of.  Then there is the web browser
> and such, which has it's own license conditions.  Of course for all I
> know their user manual has an offer of sending a CD with the sources if
> you ask.  Does anyone actually have their product that could check for
> that?

QT is GPLed, IIRC. Not LGPL-ed, meaning you can't link it with
proprietary application without license from trolltech.
								Pavel
-- 
Boycott Kodak -- for their patent abuse against Java.