FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Debian > Debian Development

 
 
LinkBack Thread Tools
 
Old 03-04-2009, 10:33 AM
 
Default xcdroast does no longer work with wodim: Who to blame?

Goswin von Brederlow <goswin-v-b@web.de> wrote:

> And the LGPL is by design clearly GPL compatible. Even if you could
> find some fault in that the intention of the license is clearly to
> allow GPL and non GPL programs to link to an LGPL library and any such
> bug would get clarified by the FSF. Any such faults please bring to
> the attention of the FSF for further clarification. Don't bother us
> with it.

The LGPL is incompatible with GPL by design. I encourage you to inform yourself.....

Jrg

--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) Jrg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni)
joerg.schilling@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/
URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 03-04-2009, 10:38 AM
Jeremiah Foster
 
Default xcdroast does no longer work with wodim: Who to blame?

On Mar 4, 2009, at 11:48 AM, Goswin von Brederlow wrote:


Joerg.Schilling@fokus.fraunhofer.de (Joerg Schilling) writes:



makes libc a derived work of the program "hello world"?

Jrg


Please do read all of the mail and try to follow each step. And if you
have a counter argument please first check that it isn't negated
further on.

The point you are not getting is that the program "hello world" comes
under the GPL because

<quote>


[snip]


So again we end up with the conclusion that the "hello world" program
requires that both hello world source and libc must be GPL. And the
same argument works for mkisofs linked against libschilly and libscg.


Wow, great explanation Goswin, thanks very much. You made parts of the
GPL much clearer to me.


Jeremiah


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 03-04-2009, 01:25 PM
Darren Salt
 
Default xcdroast does no longer work with wodim: Who to blame?

I demand that Goswin von Brederlow may or may not have written...

> Joerg.Schilling@fokus.fraunhofer.de (Joerg Schilling) writes:
[snip]
>> Well despite the claims from some people that try to prevent a solution,
>> there in fact is only a very minor disagreement.

That would appear to include "despite Schilling's own claims".

>> This disagreement is based on the attempt from some people to interpret
>> some meaning into the "system exception" that is not in the GPL text.

> The "system exception" can not be applied for Debian so whatever
> differences there are about interpreting it are completely irrelevant. The
> whole clause does not apply.

Oh, it can. It applies when it suits Schilling. [1]

[snip]
> And it still doesn't apply to Debian. So whatever permission to omit
> things it gives you is void. It doesn't apply to Debian.

Situation normal wrt cdrschillings, then...


[1] <49ae61d1.hee70xnRYZq1zOCn%Joerg.Schilling@fokus.f raunhofer.de>

--
| Darren Salt | linux or ds at | representing
| RISC OS, Linux | youmustbejoking,demon,co,uk | self only
| + RIPA NOTICE: NO CONSENT GIVEN FOR INTERCEPTION OF MESSAGE TRANSMISSION

I know what I wrote yesterday. Note to whom this is NOT sent... :-


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 03-04-2009, 04:18 PM
Bill Unruh
 
Default xcdroast does no longer work with wodim: Who to blame?

On Wed, 4 Mar 2009, Josselin Mouette wrote:


Le mardi 03 mars 2009 �0:41 -0800, Bill Unruh a �it :

Well, no, there is a problem. Whether that problem is due to a misreading of
the law, differing laws (Under US, the concept of derivative work is a very
important and strong concept. Publishers have been successfully sued for using less than
.3% of another work in their work.



From your comments, it appears that German copyright law does not contain that
concept. Unfortunately all Linux distributions are world wide and have to
worry about the law in many jurisdictions, especially the USA.


Indeed, the concept of derived work does not exist in European law. By
default, a derived work is a counterfeit. Only when the author allows it
(using e.g. a Free Software license) is it allowed at all.


Sorry, but that means that the concept does exist. The binary of a program is
a totally different animal from the original. Not a shred of similarity in
expression exists between the two. If I grep phrases from the original in the
binary, I will not find them. Thus the binary, like a translation, is NOT
a copy of the original. Thus the European law must also have some idea of what
the relation is between say a translation and the original under copyright
law. That is what "derivative works" is all about. Exactly how far does the
copyright in the original extend over something which is clearly not a copy of
expression (which is all copyright includes, it does not include ideas) but is
also clearly based on the original and on the form of expression of the
original.






--
William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273
Physics&Astronomy | Advanced Research | Fax: +1(604)822-5324
UBC, Vancouver,BC | Program in Cosmology | unruh@physics.ubc.ca
Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/
 
Old 03-04-2009, 04:38 PM
Bill Unruh
 
Default xcdroast does no longer work with wodim: Who to blame?

On Wed, 4 Mar 2009, Goswin von Brederlow wrote:


Joerg.Schilling@fokus.fraunhofer.de (Joerg Schilling) writes:


Matthew Johnson <mjj29@debian.org> wrote:


On Tue Mar 03 11:07, Joerg Schilling wrote:


The rules of the GPL end at "work" limit and neither libc nor
libschily or libscg are part of the "work" mkisofs. For this reason,
there is no problem with the fact that mkisofs links against libschily
and libscg.


The FSF certainly believes (and I think it is supported by at least US
copyright law) that the complete work of mkisofs linked against
libschily and libscg (i.e. the binary form, rather than the source) is a
single work which is a derivative work of all three individual (source)
works. Therefore, it must be distributed under terms which are
compatible with the licences of all three.


Repeating false claims does not make them correct.

In order to create a derived work, you need to add own code of a sufficient
creation level. The simple act of compiling does of course not create a derived
work.


You do not even have to create derived work. Only a "Program" as in
<quote>
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: that is to say, a work
containing the Program or a portion of it, either verbatim or with
modifications and/or translated into another language.
</quote>


Sorry, but the GPL derives whatever force it has from copyright law. It cannot
exceed copyright law in its scope. Copyright law controls the copying of a
original work, and under derivative works, it controls also derivative works.
It does not control anything else and the GPL cannot extend it control over
anything else. Ie, all statements in the GPL HAVE to be read under copyright
law. You simply cannot base your reading only on your interpretation of the
GPL. That interpretation must be consistant with copyright law.





In addition: if ever, mkisofs could be a derived work if libschily but not vice
vcersa.


Do you really like to tell us that compiling:

main()
{
printf("hello world
");
}

makes libc a derived work of the program "hello world"?

J�>

Please do read all of the mail and try to follow each step. And if you
have a counter argument please first check that it isn't negated
further on.

The point you are not getting is that the program "hello world" comes
under the GPL because

<quote>
b) You must cause any work that you distribute or publish, that in
whole or in part contains or is derived from the Program or any
part thereof, to be licensed as a whole at no charge to all third
parties under the terms of this License.
</quote>

Clearly the "hello world" program contains "in whole or in part" your
main() function. To make arguments simple lets first assume you have
build a static "hello world". So the statically linked "hello world"
as a whole must be under GPL. And that includes any part of libc you
linked into it.

It does not mean libc must be GPL (compatible) but only if it is you
may distribute the statically linked "hello world". Otherwise you
violate the licenses.


But your next argument surely is that you don't link statically and
the "hello world" source and libc are independent and separate works
in themself. This is also mentioned in the GPL:

<quote>
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works.
</quote>

This seems to agree with your interpretation. But behold, read on:

<quote>
But when you distribute the same sections as part of a whole which
is a work based on the Program, the distribution of the whole must
be on the terms of this License, whose permissions for other
licensees extend to the entire whole, and thus to each and every
part regardless of who wrote it.
<quote>

And we are back again at everything must be GPLed. The fact that you
have 2 seperate works (hello world and libc), even if packaged as 2
seperate *.deb files, is negated by it being distributed together.

[I do see that it says 'modified work' above while we could distribute
an 'unmodified work'. But the DFSG reqires that we are able to patch
sources. So we must be able to distribute 'modified work' and one could
even say that packaging itself modifies the work already. So for the
purpose of Debian the above applies.]


So now why does that not mean everything in Debian must be GPLed or
nothing at all? Reading on in the GPL you have one exception:

<quote>
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.
</quote>

So the final question is:

Is hello-world.deb and libc.deb distributed together a mere aggregation
of works?

There is a simple test you can do for yourself. If they are
independent works in themselves then I should be able to take either
one of them and have a functioning work:

Take libc.deb:
mrvn@frosties:~% /lib/libc-2.9.so
GNU C Library stable release version 2.9, by Roland McGrath et al.
Copyright (C) 2008 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
Compiled by GNU CC version 4.3.3.
Compiled on a Linux >>2.6.28-1-amd64<< system on 2009-02-22.
Available extensions:
crypt add-on version 2.1 by Michael Glad and others
GNU Libidn by Simon Josefsson
Native POSIX Threads Library by Ulrich Drepper et al
BIND-8.2.3-T5B
For bug reporting instructions, please see:
<http://www.gnu.org/software/libc/bugs.html>.

That seems to be a functioning work on its own.


But "hello world" is not. A dynamically linked "hello world" binary is
not functioning without libc.


So hello-world.deb and libc.deb are no mere aggregation of
works. Therefore both hello-world.deb and libc.deb must be under GPL.


But hey, what about the exception for system libraries:

<quote>
However, as a special exception, the source code distributed need
not include anything that is normally distributed (in either source
or binary form) with the major components (compiler, kernel, and so
on) of the operating system on which the executable runs, unless
that component itself accompanies the executable.
</quote>

Note the _unless_that_component_itself_accompanies_the_exec utable_.

Since Debian distributes compiler, kernel, libraries, programs all
together the special exception can not be used.


So again we end up with the conclusion that the "hello world" program
requires that both hello world source and libc must be GPL. And the
same argument works for mkisofs linked against libschilly and libscg.

MfG
Goswin





--
William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273
Physics&Astronomy | Advanced Research | Fax: +1(604)822-5324
UBC, Vancouver,BC | Program in Cosmology | unruh@physics.ubc.ca
Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/
 
Old 03-05-2009, 04:21 AM
Russell Coker
 
Default xcdroast does no longer work with wodim: Who to blame?

On Thu, 5 Mar 2009, Bill Unruh <unruh@physics.ubc.ca> wrote:
> Sorry, but that means that the concept does exist. The binary of a program
> is a totally different animal from the original. Not a shred of similarity
> in expression exists between the two. If I grep phrases from the original
> in the binary, I will not find them.

In future please do some tests before making wild claims (or even just
consider how a compiler works).

In all the common compilers every text string (IE something that is enclosed
in "" in the source) will appear exactly the same in the binary. Variations
on the compressed executable theme (which was common in MS-DOS days) will of
course be exceptions to this rule.

$ gcc test.c
$ strings a.out
/lib/ld-linux.so.2
__gmon_start__
libc.so.6
_IO_stdin_used
puts
__libc_start_main
GLIBC_2.0
PTRh
[^_]
hello world
$ ./a.out
hello world
$ cat test.c
#include <stdio.h>
int main() { printf("hello world
"); return 0; }
$

--
russell@coker.com.au
http://etbe.coker.com.au/ My Main Blog
http://doc.coker.com.au/ My Documents Blog


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 03-05-2009, 05:05 AM
Bill Unruh
 
Default xcdroast does no longer work with wodim: Who to blame?

On Thu, 5 Mar 2009, Russell Coker wrote:


On Thu, 5 Mar 2009, Bill Unruh <unruh@physics.ubc.ca> wrote:

Sorry, but that means that the concept does exist. The binary of a program
is a totally different animal from the original. Not a shred of similarity
in expression exists between the two. If I grep phrases from the original
in the binary, I will not find them.


In future please do some tests before making wild claims (or even just
consider how a compiler works).


Oh dear, Yes, I did think of that and assumed that the readers were
intelligent enough not to need obvious caveates. My programs have no string
literals. Yes, string
litereals are sometimes preserved. And the point is? YOu think that those
string literals will convert a non-copyrighted binary to a copyrighted one
(Read "Fair use")? "If the program has theree letters in common with the
source then the program is derivative and if two it is not?"





In all the common compilers every text string (IE something that is enclosed
in "" in the source) will appear exactly the same in the binary. Variations
on the compressed executable theme (which was common in MS-DOS days) will of
course be exceptions to this rule.


Who cares in this context?



$ gcc test.c
$ strings a.out
/lib/ld-linux.so.2
__gmon_start__
libc.so.6
_IO_stdin_used
puts
__libc_start_main
GLIBC_2.0
PTRh
[^_]
hello world
$ ./a.out
hello world
$ cat test.c
#include <stdio.h>
int main() { printf("hello world
"); return 0; }
$


That program is probably not copyrightable. It has long gone into the public domain.

A binary is a translation of the source. Does it fall under the copyright on
the source or does it not? That is the issue of "derivative work". How close
does the "translation" or the use of the original have to be to still fall
under the original copyright?

And how different does it have to be not to do so.










--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 03-05-2009, 05:23 AM
Russell Coker
 
Default xcdroast does no longer work with wodim: Who to blame?

On Thu, 5 Mar 2009, Bill Unruh <unruh@physics.ubc.ca> wrote:
> > In future please do some tests before making wild claims (or even just
> > consider how a compiler works).
>
> Oh dear, Yes, I did think of that and assumed that the readers were
> intelligent enough not to need obvious caveates.

After this discussion I don't think that you are in a position to criticise
the intelligence of anyone.

> My programs have no string literals. Yes, string
> litereals are sometimes preserved.

String literals are almost always preserved.

> And the point is? YOu think that those
> string literals will convert a non-copyrighted binary to a copyrighted one
> (Read "Fair use")? "If the program has theree letters in common with the
> source then the program is derivative and if two it is not?"

Copyright applies to strings. In fact when copyright was first invented one
of the main aims was to protect collections of strings (books).


PS What contributions are you making to any free software projects? Please
note that trolling this mailing list doesn't count as a contribution.

--
russell@coker.com.au
http://etbe.coker.com.au/ My Main Blog
http://doc.coker.com.au/ My Documents Blog


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 03-05-2009, 07:58 AM
Matthew Johnson
 
Default xcdroast does no longer work with wodim: Who to blame?

On Thu Mar 05 17:23, Russell Coker wrote:

> PS What contributions are you making to any free software projects? Please
> note that trolling this mailing list doesn't count as a contribution.

I think you're being a little harsh Russell, but you're right

Bell, I don't think anyone anywhere disagrees that the copyright on
program source also applies to the binary form of the program, or to
other forms it may be converted into. Otherwise I would be able to
pirate Microsoft office without it being illegal. After all, I'm not
copying the source. Ditto with other forms of non-text works, I don't
think any judge in the world would consider the mp3 of the latest hits
any less copyrighted than the original and that's even a lossy
transformation.

It therefore follows that any binary inherits the copyright of all the
source components which went to create it, otherwise I could clearly
steal anyone's work merely by adding enough useless code to it.

Hence we reach the same conclusion that Debian did some time ago.

matt

--
Matthew Johnson
 
Old 03-05-2009, 02:08 PM
Bill Unruh
 
Default xcdroast does no longer work with wodim: Who to blame?

On Thu, 5 Mar 2009, Matthew Johnson wrote:


On Thu Mar 05 17:23, Russell Coker wrote:


PS What contributions are you making to any free software projects? Please
note that trolling this mailing list doesn't count as a contribution.


I think you're being a little harsh Russell, but you're right

Bell, I don't think anyone anywhere disagrees that the copyright on
program source also applies to the binary form of the program, or to
other forms it may be converted into. Otherwise I would be able to
pirate Microsoft office without it being illegal. After all, I'm not
copying the source. Ditto with other forms of non-text works, I don't
think any judge in the world would consider the mp3 of the latest hits
any less copyrighted than the original and that's even a lossy
transformation.


That is the question that "derivative works" raises, and my only claim was
that it was a question that all copyright law, whether in the US or in Europe
addresses in some for or another. However, exactly what constitutes derivative
work is an incredibly contentious issue and there are no clear rules. It is
quite possible that a judge could fine that linking program A to library B
does NOT make the result a derivative work of library B. Or he could find it
does. It is at present simply not known.

As I said, the SCO case revolves around what consititutes derivative work.




It therefore follows that any binary inherits the copyright of all the
source components which went to create it, otherwise I could clearly
steal anyone's work merely by adding enough useless code to it.


Of course. That is precisely what copyright to some extent is supposed to do,
encourage others to build on the work. The consitition gives the Federal Gov't
in the US the right to set up copyright to encourage the creation of new
works, not just by protecting the author but by making the work public so
others can build on it. Copyright is a delicate balancing act, trading a
detested (in capitaist circles at least) for of ownership-- monopoly rights--
for public exposure exactly so others can build on it. When does that building
infringe on the monopoly rights? That is tough question which the courts are
constantly battling with.



Hence we reach the same conclusion that Debian did some time ago.



Debian's conclusion comes uncomfortably near to SCO's conclusion, that
derivative works are created if there is any whiff of prior contact.
A library is written to be used and to be linked to other programs. That is

its purpose. To argue that in so doing it infects the linking program and
makes it a derivative work could be dangerous, but at the least could be
argued, and has been argued by reasonable people.




matt




--
William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273
Physics&Astronomy | Advanced Research | Fax: +1(604)822-5324
UBC, Vancouver,BC | Program in Cosmology | unruh@physics.ubc.ca
Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 

Thread Tools




All times are GMT. The time now is 10:25 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org