Re: cdrtools - GPL code with CDDL build system

2006-03-19 Thread Måns Rullgård
Eduard Bloch [EMAIL PROTECTED] writes:

 #include hallo.h
 * Måns Rullgård [Sun, Mar 19 2006, 01:50:24AM]:
 Sam Morris [EMAIL PROTECTED] writes:

 These are the bits I'm referring to, from cdrecorc.c (sorry for the
 long lines, but that's how it's written):
 
 ---BEGIN QUOTE---
  /*
   * Begin restricted code for quality assurance.
   *
   * Warning: you are not allowed to modify or to remove the
   * Copyright and version printing code below!
   * See also GPL § 2 subclause c)
 ...
 For completeness, here's GPL 2c:
 
 ---BEGIN QUOTE---
 c) If the modified program normally reads commands interactively
 when run, you must cause it, when started running for such
 interactive use in the most ordinary way, to print or display an
 announcement including an appropriate copyright notice and a
 notice that there is no warranty (or else, saying that you provide
 a warranty) and that users may redistribute the program under
 these conditions, and telling the user how to view a copy of this
 License.  (Exception: if the Program itself is interactive but
 does not normally print such an announcement, your work based on
 the Program is not required to print an announcement.)
 ---END QUOTE---
 
 Take note that cdrecord is never interactive, so GPL 2c doesn't apply.

 Wrong. It displays messages by default and tells you how to stop the
 command etc. IMO this can clearly be interpreted as interaction.

It does not read commands interactively, which is the provision for
GPL 2c applying.  If every program that exits when it receives certain
signals is interactive, then all programs are interactive (think SIGKILL).

 And since it does print such an announcement by default then it
 should be kept. However, I disagree on the level appropriateness -
 stuff like This is a broken Linux system does not belong to the
 disclaimer/copyright category.

It is clearly not a copyright notice or disclaimer, and not being this
restricting its removal is non-free.

-- 
Måns Rullgård
[EMAIL PROTECTED]



Re: Problematic distribution of P2P clients in France

2006-03-19 Thread Måns Rullgård
Mike Hommey [EMAIL PROTECTED] writes:

 On Sun, Mar 19, 2006 at 12:35:16PM +0100, Marco d'Itri [EMAIL PROTECTED] 
 wrote:
 [EMAIL PROTECTED] wrote:
 
 If courts were to go for the first interpretation, my opinion is that
 french Debian (and other distributions) mirrors could be endangered.
 This is a problem for French mirror operators, not for Debian.

 It is also a problem for any Debian Developer that would come to France.

What?  Do the French lock you up for things you did outside of France,
even if they were legal there?

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: cdrtools - GPL code with CDDL build system

2006-03-19 Thread Måns Rullgård
Bernhard R. Link [EMAIL PROTECTED] writes:

 * Mns Rullgrd [EMAIL PROTECTED] [060319 01:14]:
 Don Armstrong [EMAIL PROTECTED] writes:
  Not just linking; it's the creation of a derivative work of a GPLed
  work. Frankly, I don't see how you can argue that cdrecord is not a
  derivative work of the GPLed part of cdrecord and the build system.
 
 I disagree.  The final executable is no more a derivative of the build
 system than it is of the compiler.  After all, no parts of the
 makefiles end up inside the executable.

 I think derivative or not is not the question here, but the GPL 
 explicitly demands that the build system is available under
 GPL-compatible changes.

 from Section 2 of the GPL:
 # 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.

 from Section 3 of the GPLL:

 # Accompany it with the complete corresponding machine-readable
 # source code, which must be distributed under the terms of Sections
 # 1 and 2 above on a medium [...]

 # For an executable work, complete source
 # code means all the source code for all modules it contains, plus any
 # associated interface definition files, plus the scripts used to
 # control compilation and installation of the executable.

 scripts used to control compilation is clearly what we currently
 refer to as build system.

Point taken.  The obvious solution is to replace the build system
with an acceptably licensed one.  While at it, one could also make it
work properly.  Incidentally, this is what the dvdrtools folks have
already done.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Problematic distribution of P2P clients in France

2006-03-19 Thread Måns Rullgård
Mike Hommey [EMAIL PROTECTED] writes:

 On Sun, Mar 19, 2006 at 01:39:12PM +, Måns Rullgård [EMAIL PROTECTED] 
 wrote:
 Mike Hommey [EMAIL PROTECTED] writes:
 
  On Sun, Mar 19, 2006 at 12:35:16PM +0100, Marco d'Itri [EMAIL PROTECTED] 
  wrote:
  [EMAIL PROTECTED] wrote:
  
  If courts were to go for the first interpretation, my opinion is that
  french Debian (and other distributions) mirrors could be endangered.
  This is a problem for French mirror operators, not for Debian.
 
  It is also a problem for any Debian Developer that would come to France.
 
 What?  Do the French lock you up for things you did outside of France,
 even if they were legal there?

 I think you can get charged for exporting illegal stuff to France
 whenever you touch the french soil, this being legal in your own
 country or not. That applies to a lot of countries, I believe.

There ought to be some requirement for said export to be somewhat
active or deliberate for the law to kick in.  Simply offering stuff
for download being sufficient is insane.  Unfortunately, the system is
insane, as evidenced by some cases involving Nazi uniforms.

-- 
Måns Rullgård
[EMAIL PROTECTED]



Re: cdrtools - GPL code with CDDL build system

2006-03-19 Thread Måns Rullgård
Francesco Poli [EMAIL PROTECTED] writes:

 On Sat, 18 Mar 2006 22:05:53 + Måns Rullgård wrote:

 Eduard Bloch [EMAIL PROTECTED] writes:
 
  Hello debian-legal experts ;-),
 
  I need a bit support to clarify the issue with cdrtools' build
  system.
 
  Summary: a while ago, Joerg Schilling (upstream) replaced the
  copyright headers in the files of his build system inside of the
  cdrtools package with references to a CDDL license context.
 
 D-L v. JS, now that's a flame war I'd like to see ;-)
 
 Flaming aside, this is a non-issue.  The source for cdrecord contains
 invariant sections (those obnoxious warnings about using device
 names), so it's certainly not DFSG-free.

 Aaargh!  :-(
 I was hoping this problem had been fixed...

[...]

 How can such a package still be in main with this non-freeness?
 I'm surprised that nobody filed a serious bug for this...

I guess everyone has just gotten used to Schilling's inane ramblings,
and ignores him without further thought.

 Just use dvdrtools instead.

 ITYM dvd+rw-tools,

That's what I use for burning DVDs.  It doesn't handle CDs though.

 since dvdrtools is in non-free too...

Someone's launched an investigation into the reason for this.  Let's
wait and see what comes back.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Results for Debian's Position on the GFDL

2006-03-19 Thread Måns Rullgård
Adam McKenna [EMAIL PROTECTED] writes:

 On Sun, Mar 19, 2006 at 01:36:14AM -0500, Anthony DeRobertis wrote:
 Adam McKenna wrote:
 But you can only use one copy at a time.  You could make a good
 argument that the copies not in use are backup copies.  (Remember,
 we're talking about documents here.)
   
 Well, US copyright law at least gives the right to make a backup copy so 
 long as such new copy or adaptation is for archival purposes only. 
 Clearly, if you're regularly using it, its no longer for archival 
 purposes only.

 That would need to be decided by a court.  Obviously if you can only
 use one copy at a time, and your backup strategy involves keeping
 multiple copies on multiple machines, someone would have to *prove*
 that you were using more than one copy at a time, and convice the
 cort that your backup strategy does not comply with the license.

Next we know, they'll be counting mirrored disks as multiple copies,
and probably as using all the copies at once too.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: cdrtools - GPL code with CDDL build system

2006-03-20 Thread Måns Rullgård
Francesco Poli [EMAIL PROTECTED] writes:

 On Mon, 20 Mar 2006 01:21:08 + Måns Rullgård wrote:

 Francesco Poli [EMAIL PROTECTED] writes:
 
  On Sat, 18 Mar 2006 22:05:53 + Måns Rullgård wrote:
 
  Just use dvdrtools instead.
 
  ITYM dvd+rw-tools,
 
 That's what I use for burning DVDs.  It doesn't handle CDs though.

 Ouch! It seems that we are left with *no* DFSG-free tools to burn CDs in
 Debian, until someone forks cdrtools (from a version prior to the
 license clarifications) or implements an equivalent program suite...

JS insists that his clarifications also apply to previous versions
of cdrecord...

  since dvdrtools is in non-free too...
 
 Someone's launched an investigation into the reason for this.  Let's
 wait and see what comes back.

 Assuming that its debian/copyright file is accurate (in Debian sid), it
 seems that the same license (and added restrictions, passed off as
 license interpretation) applies.
 If this is the case, my bet is that dvdrtools is in non-free for pretty
 the same reason why cdrtools should not be in main (at least, not as it
 is now)...

The dvdrtools web page has this paragraph:

  dvdrtools is a fork of cdrtools, with the primary goals of remaining
  100% Free Software (dvdrtools is a fork of the last version of
  cdrtools without any you are not allowed to modify this section
  comments), and adding support for DVD-R/DVD-RW drives and media.

Checking the source, none of the troublesome bits from cdrtools are
there, and the build system is replaced with automake.  I can't see
any problem with it.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: MIT licenses

2006-03-20 Thread Måns Rullgård
Francesco Poli [EMAIL PROTECTED] writes:

 On Mon, 20 Mar 2006 02:39:56 + MJ Ray wrote:

 I think that's the one. There are several often called MIT. Someone
 has moved the copy on X.org to which
 http://www.fr.debian.org/legal/licenses/ links - has anyone
 a new URL besides the failed open source initiative, please?

 AFAIK, the two most famous licenses ambiguously called MIT license,
 are:

  a) the Expat license (http://www.jclark.com/xml/copying.txt)

  b) the X11 license (formerly  http://www.x.org/Downloads_terms.html)

 The latter URL (http://www.x.org/Downloads_terms.html) has been useful
 for long time, but now seems to have vanished.
 Another URL (http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3) includes
 something that resembles to the X11 license, but it rather seems to be
 an Expat license with an added final no-advertising/no-promotion clause
 similar to the one found in the X11 license.

 I'm pretty sure the real X11 license (the one once found at 
 http://www.x.org/Downloads_terms.html, now vanished) has other
 differences that distinguish it from the Expat license.
 Namely:
 a) the condition explicitly mentions that the copyright notice and
 license text must be included in supporting documentation (as well as in
 the Software or any substantial portion)
 b) permission to sublicense is not explicitly mentioned

 These two differences seem to be negligible, since the term Software
 includes associated documentation files and the list of granted rights
 is exemplary, not exhaustive (Permission is hereby granted [...] to
 deal in the Software without restriction, including without limitation
 the rights to [...]).
 Nonetheless, the X11 and Expat license texts differ in those details too
 (not only for the presence of the final no-advertising/no-promotion
 clause).

 Has anyone found another URL for the real X11 license (the one that
 used to live at http://www.x.org/Downloads_terms.html)?

Gentoo has a fairly comprehensive collection of license texts at
http://www.gentoo.org/cgi-bin/viewcvs.cgi/licenses/, including an
MIT and an X11 license that are quite distinct.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: cdrtools - GPL code with CDDL build system

2006-03-23 Thread Måns Rullgård
Francesco Poli [EMAIL PROTECTED] writes:

 On Tue, 21 Mar 2006 18:19:52 +0100 Eduard Bloch wrote:

 Don't count much on dvdrtools, it has no active upstream at all (no, I
 don't mean the guys whoes only heroic act was the replacement of the
 Schilly build system with autodev-stuff).

 That's a good reason to find someone willing to take over dvdrtools
 maintenance and development...
 We should really seek someone interested.

Umm... they released a new version just a couple of weeks ago.  What
do you require of a project to count it as active?

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: license of schema files

2006-04-04 Thread Måns Rullgård
Noèl Köthe [EMAIL PROTECTED] writes:

 Hello debian-legal and openldap maintainers,

 the upstream kolabd package includes rfc2739.schema:

 http://kolab.org/cgi-bin/viewcvs-kolab.cgi/server/kolabd/kolabd/rfc2739.schema?rev=1.1.1.1content-type=text/vnd.viewcvs-markup

 kolabd was rejected by ftpmaster because of this schema license text
 (this schema is now removed but it would be better to get it included
 again):

 ...
 However, this
 #  document itself may not be modified in any way, such as by removing
 #  the copyright notice or references to the Internet Society or other
 #  Internet organizations, except as needed for the purpose of
 #  developing Internet standards in which case the procedures for
 #  copyrights defined in the Internet Standards process must be
 #  followed, or as required to translate it into languages other than
 #  English.
 ...

 document itself may not be modified in any way is the main point. The
 following are examples and more information. It looks like its just a
 copy of a RFC license e.g. ftp://ftp.rfc-editor.org/in-notes/rfc2821.txt

IMHO permission to modify standards is uninteresting.  The document is
only useful in it's original form anyway.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: MPEG-4 patent license issues - libfaad* and libx264* and other codecs.

2006-04-30 Thread Måns Rullgård
Don Armstrong [EMAIL PROTECTED] writes:

 On Sun, 30 Apr 2006, Josselin Mouette wrote:
 Le samedi 29 avril 2006 à 23:37 +0100, Matthew William Solloway Bell a
 écrit :
  The packages libxine1, ffmpeg,  include libfaad*, libx264* or another
  codec which implement the MPEG-4 Advanced Audio Coding and Advanced
  Video Coding standards.
 
 I have already stated this: Debian shouldn't have anything to do
 with regard to patents. We should entirely ignore them unless
 directly threatened, and such issues that depend so much on the
 country should be up to the end-user to deal with.

 Like it or not, we can't completely ignore them in Debian. Each mirror
 operator needs to make their own decisions about their liability in
 their own country, but Debian itself needs to be careful about
 distributing works which have patents in the United States[1] where
 the patent holders are well known and have been inolved in patent
 litigation against individuals using their patents.

Correct me if I'm wrong, but isn't it use, not distribution, that
requires a patent license?

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: MPEG-4 patent license issues - libfaad* and libx264* and other codecs.

2006-05-01 Thread Måns Rullgård
Steve Langasek [EMAIL PROTECTED] writes:

 On Sat, Apr 29, 2006 at 11:37:39PM +0100, Matthew William Solloway
 Bell wrote:
 The packages libxine1, ffmpeg,  include libfaad*, libx264* or another
 codec which implement the MPEG-4 Advanced Audio Coding and Advanced
 Video Coding standards. Unfortunately, these are patent encumbered in at
 least the USA, and many other countries. To distribute code implementing
 any of these patents, a license is required[1], assuming that the
 claimed patents are valid. This license requires signing an agreement
 and the payment of royalties, which hasn't been done AFAIK, and is
 contrary to policy. 
 There is evidence of prior attempts of enforcement, specifically against
 FAAD at AudioCoding.com[2].

 This appears to refer to enforcement of patents covering encoding using the
 codecs in question.  Do libxine1 and ffmpeg implement encoding of these, or
 just decoding?  Is there a history of enforcement of patents on decoding of
 the codecs in question?

FFmpeg has an AVC decoder.  It does not include an AVC encoder, and it
neither encodes nor decodes AAC.  FFmpeg can optionally use external
libraries for these tasks.

I'm not familiar with libxine enough to comment on it.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: MPEG-4 patent license issues - libfaad* and libx264* and other codecs.

2006-05-01 Thread Måns Rullgård
Matthew William Solloway Bell [EMAIL PROTECTED] writes:

 On Sat, Apr 29, 2006 at 11:37:39PM +0100, Matthew William Solloway Bell 
 wrote:
  The packages libxine1, ffmpeg,  include libfaad*, libx264* or another
  codec which implement the MPEG-4 Advanced Audio Coding and Advanced
  Video Coding standards. Unfortunately, these are patent encumbered in at
  least the USA, and many other countries. To distribute code implementing
  any of these patents, a license is required[1], assuming that the
  claimed patents are valid. This license requires signing an agreement
  and the payment of royalties, which hasn't been done AFAIK, and is
  contrary to policy. 
  There is evidence of prior attempts of enforcement, specifically against
  FAAD at AudioCoding.com[2].
 
 This appears to refer to enforcement of patents covering encoding using the
 codecs in question.  Do libxine1 and ffmpeg implement encoding of these, or
 just decoding?  Is there a history of enforcement of patents on decoding of
 the codecs in question?

 Hmmm, I think I have missed something; what makes you draw this
 conclusion? AudioCoding.com has removed all binaries including those
 related to decoding. I see no reference to encoding only in [2]. The
 licensing authorities in [1] have licenses that cover decoders. I did
 look at their patent portfolio, but is was brief and shallow. I'm having
 a closer look now.

 libxine: libfaad (AAC decoder)
 vlc: libfaad (AAC decoder); libx264 (AVC decoder)
 libavcodec0: libfaad (AAC decoder); libx264 (AVC decoder)

 AFAIK, libx264 is a decoder only but the decoding functions are called
 x264_encoder_?

x264 is an AVC *encoder* only.  libavcodec can optionally call it for
AVC encoding.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: DFSG as Licence?

2006-06-11 Thread Måns Rullgård
Michelle Konzack [EMAIL PROTECTED] writes:

 Hello *,

 Since I have read tonns of different licences I do not realy know
 what to do.  Since I am using Debian/main only (with the exception
 of libdvdcss2) since more then 7 years now I want to say, that my
 Software any Licence which comply with the DFSG.

 Question:

 Is there allready a licence which use the term DFSG as licence?

 I do not fully agree with the FSF and the GPL. v2.0 maybe ok,
 but I have complains against the new one.

If you do not like gpl3, use gpl2 without the or later option, if
that does what you want.  The FSF won't like you if you do, but nobody
is under any obligation to please them.  Personally, I'm allergic to
more than two paragraphs of legalese, and I don't want to release my
work under terms I do not fully understand, so I release my stuff
under the MIT license.  It gives a little more permission than the
GPL, but I don't really care if someone uses my code in a commercial
application.  It doesn't interfere with my reasons for releasing it in
the first place, and it lets any free software project use it, without
any concerns about being GPL compatible.  All the fuss about open
source licenses being incompatible is, IMHO, contradictory to the
spirit of free software, and spending time on such issues is
counter-productive.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: DFSG as Licence?

2006-06-11 Thread Måns Rullgård
George Danchev [EMAIL PROTECTED] writes:

 On Sunday 11 June 2006 19:25, Måns Rullgård wrote:
 Michelle Konzack [EMAIL PROTECTED] writes:
  Hello *,
 
  Since I have read tonns of different licences I do not realy know
  what to do.  Since I am using Debian/main only (with the exception
  of libdvdcss2) since more then 7 years now I want to say, that my
  Software any Licence which comply with the DFSG.
 
  Question:
 
  Is there allready a licence which use the term DFSG as licence?
 
  I do not fully agree with the FSF and the GPL. v2.0 maybe ok,
  but I have complains against the new one.

 If you do not like gpl3, use gpl2 without the or later option, if
 that does what you want.  The FSF won't like you if you do, but nobody
 is under any obligation to please them.  Personally, I'm allergic to
 more than two paragraphs of legalese, and I don't want to release my
 work under terms I do not fully understand, so I release my stuff
 under the MIT license.  It gives a little more permission than the
 GPL, but I don't really care if someone uses my code in a commercial
 application.  

 GPL allows commercial applications, but what GPL does not allow is
 becoming a 'proprietary application' (non-free). E.g. you are not

OK, bad choice of words.  I don't much care if someone uses my code in
a proprietary application either.

 allowed to grab a GPL'ed source code, modify it and distribute the
 modified binaries only. In that case GPL force you to publish the
 your source modifications, which is perfectly in the spirit of free
 software ... e.g. what is give is what you get.

What I'm talking about is different, each on their own free, licenses
being deemed incompatible with each other.  Examples are the GPL, the
OpenSSL license, and the Open Software License.  I find it hard to
believe that most authors who choose to release under the GPL do so in
order to prevent their code being used in a program released under the
OSL.  Neither of these two licenses (GPL and OSL) allows for
proprieterization of code.  However, I see it as a loss to the free
software world as a whole, that the open source code is divided into
several islands, between which no code sharing is allowed.  This leads
to time and efforts being wasted in reimplementing perfectly good
code, only because the existing version has slightly different terms
of use and distribution.

How many cases of Foo is under GPL, Foo uses libcurl, libcurl can be
linked with OpenSSL, hence Foo is non-distributable have been
discussed on this list?  I have no figures, but it is a recurring
topic.  Does anyone seriously believe that the authors of Foo
intentionally created those situation?

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Debian-approved creative/content license?

2007-03-11 Thread Måns Rullgård
Ken Arromdee [EMAIL PROTECTED] writes:

 On Sun, 11 Mar 2007, Ben Finney wrote:
 Those licenses can apply to any software, not just programs. So, if
 the software is an audio work or picture, a software license like GPL
 or Expat can apply to it.

 Actually, there's one big problem.  The GPL's preferred form for
 modification clause.

 Unless the creators of the podcast directly edit the MP3--which is rather
 unlikely--the MP3 is not the preferred form for modification and putting the
 MP3 under GPL without releasing the raw audio files grants no rights at all.
 GPLing video has a similar problem.

The preferred form for modification for a film director is often a
reshoot of the scene.  I guess this means that a GPL video would have
to ship with (a copy of) Tom Cruise if he happens to be one of the
actors in the film.  Sounds difficult to fulfill.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Final text of GPL v3

2007-07-04 Thread Måns Rullgård
Ben Finney [EMAIL PROTECTED] writes:

 Stephen Gran [EMAIL PROTECTED] writes:

 You continually miss the point that the GPL is explicitly noted as a
 free license, which means that anything in the GPL is DFSG free.

 No. It means that works licensed under the GPL are considered free
 software under the DFSG. That does *not* mean that anything in the
 GPL is DFSG free outside the context of a work licensed under the
 GPL.

It may also be worth noting that GPLv2 *has* to be considered DFSG
free, or there would be little left to call Debian...

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: LGPL v3 compatibilty

2007-07-16 Thread Måns Rullgård
Walter Landry [EMAIL PROTECTED] writes:

 Francesco Poli [EMAIL PROTECTED] wrote:
 On Sat, 14 Jul 2007 21:56:27 -0700 (PDT) Walter Landry wrote:
 
  Francesco Poli [EMAIL PROTECTED] wrote:
   On Mon, 2 Jul 2007 12:31:13 -0400 Anthony Towns wrote:
   
   [...]
Note that _if_ we do stick to the view we've taken up until now,
when we have a LGPLv3 only glibc in the archive, we'll no longer
be able to distribute GPLv2-only compiled executables.
   
   Unless the GPLv2-only work copyright holder(s) add(s) a special
   exception, similar to the one needed to link with the OpenSSL
   library, right?
   
   This scenario is worrying me...  :-(
  
  Is this going to be a problem for the kernel?  It is definitely not
  going to go to GPLv3.
 
 Is the Linux kernel linked with any LGPL'd work?
 AFAIUI, it is not, so no problem for the kernel.

 Doesn't the kernel get its implementations for pow(), sqrt(),
 printf(), and the rest of the C standard library from glibc, which is
 LGPL'd?

No.  The kernel is completely self-contained.  Some code may of course
have been borrowed from glibc at some point, but that's irrelevant.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: LGPL v3 compatibilty

2007-07-16 Thread Måns Rullgård
Walter Landry [EMAIL PROTECTED] writes:

 Måns_Rullgård [EMAIL PROTECTED] wrote:
 Walter Landry [EMAIL PROTECTED] writes:
 
  Francesco Poli [EMAIL PROTECTED] wrote:
  On Sat, 14 Jul 2007 21:56:27 -0700 (PDT) Walter Landry wrote:
  
   Francesco Poli [EMAIL PROTECTED] wrote:
On Mon, 2 Jul 2007 12:31:13 -0400 Anthony Towns wrote:

[...]
 Note that _if_ we do stick to the view we've taken up until now,
 when we have a LGPLv3 only glibc in the archive, we'll no longer
 be able to distribute GPLv2-only compiled executables.

Unless the GPLv2-only work copyright holder(s) add(s) a special
exception, similar to the one needed to link with the OpenSSL
library, right?

This scenario is worrying me...  :-(
   
   Is this going to be a problem for the kernel?  It is definitely not
   going to go to GPLv3.
  
  Is the Linux kernel linked with any LGPL'd work?
  AFAIUI, it is not, so no problem for the kernel.
 
  Doesn't the kernel get its implementations for pow(), sqrt(),
  printf(), and the rest of the C standard library from glibc, which is
  LGPL'd?
 
 No.  The kernel is completely self-contained.  Some code may of course
 have been borrowed from glibc at some point, but that's irrelevant.

 Are you sure that it is self-contained?  Grepping through the sources
 of 2.6.22.1, I do not see an implementation of stdarg.h or
 stdio.h.  I do see string.h, and math.h is never included.

Inside the kernel stdio is meaningless, so I'd hardly expect to find
that header there.  The only places in the kernel source where stdio.h
is included are in tools, such as kconfig, only to be used for
building the kernel or for testing purposes.  This use of standard C
header files is of course completely outside the scope of any license.

As for stdarg.h, it is provided by the compiler, and generally
(including GCC) licensed without restrictions on use.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: algorithm copyright? what's that?

2007-10-03 Thread Måns Rullgård
Antti-Juhani Kaijanaho [EMAIL PROTECTED] writes:

 Is it a kind of algorithm copyright?

 No.

In some countries there is.  They call it a patent.

 IANAL etc

Neither am I.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: rescuing code from the GPL

2007-11-11 Thread Måns Rullgård
Shriramana Sharma [EMAIL PROTECTED] writes:

 Wesley J. Landaker wrote:
 2. Y modifies this program to use Qt (under the GPL), creating
 02-qt-nothirdvar.cpp, and distributes it under both the BSDL and GPL.
 Well, they could distribute the source code under the BSD, as the
 source code isn't a derivative work of Qt just by using it. But they
 could not legally distribute a compiled binary that included
 copyrightable parts of GPL-only Qt. You could distribute binaries
 under the GPL.

 I don't agree with some points in this.

 The question is not whether a work *includes* parts of Qt or not. The
 very fact that it is dependent on Qt for its functioning makes it a
 derivative work, and it *must* be licensed under the GPL when
 distributed, whether in source form or compiled form.

 Please point out the flaw in this reasoning. Thank you.

Suppose I write from scratch a new library, Tq, that is source and
binary compatible with Qt (huge task, but that's beside the point).
Every app written to use Qt can now instead use Tq without even a
recompile.  Are these apps now suddenly derivatives of Tq as well as
Qt?  I think not.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: TrueCrypt License 2.3

2008-01-12 Thread Måns Rullgård
Patrick Matthäi [EMAIL PROTECTED] writes:

 Hello,

 I wanted to package maybe truecrypt for Debian.
 There was an older discussion on l.d.legal for an older version of the
 TrueCrypt license, where the most developers said, that it is not
 distributeable.

 But as I know TrueCrypt has modified the license, so that more
 distributions could ship it.

 Here it is: http://www.truecrypt.org/license.php

 I'm not a lawler, so what do you mean, is this license free or when
 not could I distribute it in non-free?

At a glance, the bits that might be controversial appear to be a few
naming restriction clauses and some advertising clauses.  I don't see
anything restricting use or distribution.

IANAL

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: TrueCrypt License 2.3

2008-01-13 Thread Måns Rullgård
Patrick Matthäi [EMAIL PROTECTED] writes:

 Måns Rullgård schrieb:
 Patrick Matthäi [EMAIL PROTECTED] writes:

 Hello,

 I wanted to package maybe truecrypt for Debian.
 There was an older discussion on l.d.legal for an older version of the
 TrueCrypt license, where the most developers said, that it is not
 distributeable.

 But as I know TrueCrypt has modified the license, so that more
 distributions could ship it.

 Here it is: http://www.truecrypt.org/license.php

 I'm not a lawler, so what do you mean, is this license free or when
 not could I distribute it in non-free?
 At a glance, the bits that might be controversial appear to be a few
 naming restriction clauses and some advertising clauses.  I don't see
 anything restricting use or distribution.
 IANAL

 Thanks,

 sounds good :)

Hold on.  I didn't say it was free or anything, only that I didn't see
anything that struck me as obviously non-free.  Some consider renaming
clauses non-free (although I don't), and the plethora of licenses used
for Truecrypt could be conflicting, even if each on its own is free.
Wait for another opinion before drawing any conclusions.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: TrueCrypt License 2.3

2008-01-13 Thread Måns Rullgård
Iain Nicol [EMAIL PROTECTED] writes:

 VI. General Terms
 
 1. You may not use, modify, reproduce, derive from, (re)distribute, or
 sublicense This Product, or portion(s) thereof, except as expressly
 provided under this License. Any attempt (even if permitted by
 applicable law) otherwise to use, modify, reproduce, derive from,
 (re)distribute, or sublicense This Product, or portion(s) thereof,
 automatically and immediately terminates Your rights under this License.

 This paragraph explicitly denies rights available under fair use or fair 
 dealing. Hopefully a non-op (?), but not good.

If it were a contract, such a clause could be valid.  Whether licenses
like this are to be considered contracts is matter for debate.

Either way, the license has a clause about unenforcable terms:

  4. If any term of this License is found to be invalid or
  unenforceable under applicable law, You agree that it shall not
  affect the validity or enforceability of any other terms of this
  License that are found to be valid and enforceable under applicable
  law.

 All the above was about the TrueCrypt License version 2.3. The other 
 license I have trouble with is a short one.
 
 
 This is an independent implementation of the encryption algorithm:
 
 Twofish by Bruce Schneier and colleagues
 
 which is a candidate algorithm in the Advanced Encryption Standard
 programme of the US National Institute of Standards and Technology.
 
 Copyright in this implementation is held by Dr B R Gladman but I hereby
 give permission for its free direct or derivative use subject to

If the copyright is held be Dr Gladman, how can I (Schneier?) grant
any permission pertaining to it?

 acknowledgment of its origin and compliance with any conditions that the
 originators of the algorithm place on its exploitation.

 I know the reference implementation for Twofish is in the public domain, 
 and it's not been patented. But what happens, hypothetically, if Bruce 
 Schneier were to publicly assert that people should not use the 
 algorithm, say for moral reasons. Or what if he said people should not 
 use this algorithm [as it is no longer considered secure enough. Could 
 those situations not revoke my license to use this software?

Note that the text says algorithm, not implementation.  If it is
not patented, there is nothing the originators of the algorithm can
do to stop it being used.

IANAL

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Do web applications disable GPL obligations?

2008-03-11 Thread Måns Rullgård
Mark Tyler [EMAIL PROTECTED] writes:

 Dear all,
 I hope this is the right list to discuss GPL related issues. (Or where
 would be a better place to get help with the following question?)

Your question doesn't relate to Debian, so strictly speaking it is
off-topic here.

 A friend and I have developed a web application for an online shop
 with some pretty advanced features that were requested by our
 customers. We included GPL sources into the application and for that
 reason distributed the application according to the same license
 (GPLv3).

 Now one of the customers has modified large parts of our shop without
 mentioning the origin, the GPL license or the authors, and without
 publishing the modified sources. Program icons and dialog boxes where
 copied identically.

 When we asked him to comply with the GPL he only replied that web
 applications are not affected, because the application is only used,
 but not distributed. It is true, that the application is installed on
 the server side, but parts of it, namely java classes, are distributed
 to the clients (=the shop visitors) in binary form.

 We are really annoyed because we spent months fulfilling our
 customers' wishes at an almost negligible price (assuming that we were
 contributing to open source software), and now our work as well as the
 original authors' work is exploited to generate large turnovers.

 Can you judge from this information, whether we are right or the
 customer who claims to have the right to publish our java classes and
 icons within the web application as if he owned the copyright?

It is my understanding that source is only required to be supplied to
those who receive binaries.  In your case, it seems pretty clear that
the Java classes should be accompanied with source (or instructions
how to obtain it).  However, source for software that only runs on the
server need not be supplied to users of the service.

There were early plans to have the GPLv3 require source distribution
with web services and similar, but these requirements were dropped
from the final version.

Disclaimers: IANAL, TINLA.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: k3b monkey audio plugin

2008-03-11 Thread Måns Rullgård
Robin Heron [EMAIL PROTECTED] writes:

 Hi,
I am unclear of the license and exactly which section a deb package
 should belong. I am currently building k3b plugin for monkeys audio
 codec. The author has licenced under GPL 2 so no problem there. 
 The actual codec however uses the following:

   Monkey's Audio Source Code License Agreement

It may be of interest to you that FFmpeg has an independent
implementation of this decoder licensed under the LGPL 2.1.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Do web applications disable GPL obligations?

2008-03-12 Thread Måns Rullgård
M. Tyler [EMAIL PROTECTED] writes:

 Måns Rullgård-3 writes:
Mark Tyler [EMAIL PROTECTED] writes:

 Dear all,
 I hope this is the right list to discuss GPL related issues. (Or where
 would be a better place to get help with the following question?)

Your question doesn't relate to Debian, so strictly speaking it is
off-topic here.
 We are talking about Debian servers, of course ;-)
 Fact is that I haven't found a list or a newsgroup where GPL questions are
 answered nearly as competent as here. So I thank all writers for sharing
 their experience.

It is my understanding that source is only required to be supplied to
those who receive binaries.  In your case, it seems pretty clear that
the Java classes should be accompanied with source (or instructions
how to obtain it).
 This is exactly how I see the situation. But two questions remain to be
 clarified:
 1) I have not mentioned explicitly that the Java classes are published as
 applets on a website. They are not offered as standalone download. I don't
 see any difference in that the source code of GPLd applets must still be
 made available, do you?

Anyone who can get their hands on the Java class files has a right to
get the source code as well.  It doesn't matter how they obtain those
files, or whether whomever they were obtained from was aware of the
files being distributed.  It is the responsibility of the server
operator to keep track of what is distributed from his server, and
ensure that any conditions attached to such distribution are adhered
to.

 2) What about icons which are delivered with any GPLd software (not
 necessarily our situation, but if it is different, please mention the
 crucial points as well)? I haven't found a conclusive answer in previous
 discussions on this list. I don't see how icons are affected by the GPL,
 because the GPL deals with source code and binary program code, not with
 media files.

This is a (common) misconception.  The conditions of the GPL can be
applied to anything, not just program code.

 For this reason I assume that icons remain copyright of their
 creator, such as any data which is published without any license at
 all. Is this assumption true, or what are the obligations if
 somebody copies icons onto his website, icons that we created for
 and distribute with our GPLd software?

It is usually assumed, unless otherwise noted, that artwork included
in a package is under the same license as the source code.  It is,
however, not all too uncommon with special conditions for logos, icons
and the like.  C.f. the Mozilla projects.

However, source for software that only runs on the
server need not be supplied to users of the service.
 This is evident.

Disclaimers: IANAL, TINLA.
 BTW, is there a reason why many of the list contributions end with these
 acronyms? Do they have any legal significance or do they only show that you
 don't want to be misunderstood to tell the final truth?

You can never be too careful in legal matters.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Bug#476644: bring back H.261 encoding

2008-04-18 Thread Måns Rullgård
Fabian Greffrath [EMAIL PROTECTED] writes:

 Package: ffmpeg
 Severity: wishlist
 Version: 0.svn20080206-1

 Hi,

 according to
 http://blogs.sun.com/openmediacommons/entry/oms_video_a_project_of
 and other resources, SUN develops a new codec called OMS, which is a
 royalty-free codec loosely based on the h.26x codec family. In the
 FAQ the question asking Why have you started from h.261? is answered
 with the following reply:

   H.261 was finalized in 1989, outside the (17-year) patent
   window. Key tool strategies and prior art were already
   established in that era.

 Wouldn't this mean that we are also free to ship the built-in H.261
 encoder in ffmpeg packages?

Even though the spec dates from 1989, it is possible to use newer
algorithms in an encoder, e.g. for motion estimation, quantisation, or
any other aspect where the spec doesn't detail how values are to be
computed from input data.  I have no idea whether any patents are
applicable to the FFmpeg H.261 encoder, but I wouldn't discount the
possibility without a thorough examination.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Copy vs. (re)distribute

2008-05-13 Thread Måns Rullgård
Cyril Brulebois [EMAIL PROTECTED] writes:

 Hi,

 since I've got upstreams (having copied some code from others, that's
 why they aren't spelling it out directly) that aren't convinced that
 having the rights to copy, use, modify is insufficient to meet the DFSG.
 From what I recall having read during NM, I've never seen any discussion
 comparing the rights to copy and to (re)distribute. Could someone please
 shed some light, and/or share pointers to some nice reading on that
 topic?

 I also believe that You may use this program as desired without
 restriction can't be understood as one is free to modify and
 redistribute it; but I'd appreciate your views (anyway, I believe it was
 meant to be free for real, just not worded adequately -- this one
 dates back to 1986 -- but I'd like to make sure it's not DFSG-free
 as-is).

There is another important aspect you didn't mention.  In addition to
granting rights to use, copy, and redistribute, the license must allow
you to pass on these rights to those receiving copies or modified
versions from you.

I agree that the rights to use, copy, and redistribute are distinct,
and that none of them implicitly include any of the others.

IANAL

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Licence query

2008-09-14 Thread Måns Rullgård
Francesco Poli [EMAIL PROTECTED] writes:

 On Sun, 14 Sep 2008 17:10:17 +0100 Julian Gilbey wrote:

 Hi all!

 Hi!  :)

 
 I'm looking at a package (GLFW, bindings to OpenGL from Haskell) which
 has the following licence.  I think that it is DFSG-free, but I'd like
 someone else to check over it, as I've not seen a licence exactly like
 this one before.

 The license you quoted (thanks for quoting its text in full) is
 word-by-word identical to the zlib license:
 http://www.gzip.org/zlib/zlib_license.html

I thought it looked familiar, but couldn't quite place it.

There is one thing about that license that strikes me as slightly odd.

  Permission is granted to anyone to use this software for any purpose,
  including commercial applications, and to alter it and redistribute it
  freely, subject to the following restrictions:

In the above grant of permissions, I see no explicit grant to
distribute modified versions.  It is fairly obvious from the remainder
of the license that such permission was intended, but it should still
be explicitly mentioned.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Licence query

2008-09-14 Thread Måns Rullgård
Stephen Gran [EMAIL PROTECTED] writes:

 This one time, at band camp, Måns Rullgård said:
 There is one thing about that license that strikes me as slightly odd.
 
   Permission is granted to anyone to use this software for any purpose,
   including commercial applications, and to alter it and redistribute it
   freely, subject to the following restrictions:
 
 In the above grant of permissions, I see no explicit grant to
 distribute modified versions.  It is fairly obvious from the remainder
 of the license that such permission was intended, but it should still
 be explicitly mentioned.

 Permission is granted to ... alter it and redistribute it freely seems
 like it does just that?

The first it is clearly referring to the unmodified source.  The
second it has no other noun to refer to, so must also be referring
to the unmodified source.  Had the text said something like and
redistribute it freely, with or without modification, all would be
much clearer.  The BSD license uses this precise phrase.

One can never be too careful with legal language.

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Licence query

2008-09-14 Thread Måns Rullgård
Stephen Gran [EMAIL PROTECTED] writes:

 This one time, at band camp, Måns Rullgård said:
 Stephen Gran [EMAIL PROTECTED] writes:
 
  This one time, at band camp, Måns Rullgård said:
  There is one thing about that license that strikes me as slightly odd.
  
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
  
  In the above grant of permissions, I see no explicit grant to
  distribute modified versions.  It is fairly obvious from the remainder
  of the license that such permission was intended, but it should still
  be explicitly mentioned.
 
  Permission is granted to ... alter it and redistribute it freely seems
  like it does just that?
 
 The first it is clearly referring to the unmodified source.  The
 second it has no other noun to refer to, so must also be referring
 to the unmodified source.  Had the text said something like and
 redistribute it freely, with or without modification, all would be
 much clearer.  The BSD license uses this precise phrase.
 
 One can never be too careful with legal language.

 One can also try to be slightly sensible.

Try telling that to the lawyers.

 English is an inexact language at the best of times.  In this
 context, the meaning is clear enough - it's a logical and operation.
 Of course it's possible that some legalistic moron could find a way
 to argue that the intent of the license is the opposite of what it
 apparently says, but I doubt it will stand up in any court that
 counts.

Even the Eastern District Court of Texas?

-- 
Måns Rullgård
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: GPL photographies, eg for backround

2008-12-29 Thread Måns Rullgård
Ken Arromdee arrom...@rahul.net writes:

 On Mon, 29 Dec 2008, Josselin Mouette wrote:
 More precisely: if you are the copyright owner, you can publish it in
 whatever format you like, and if under a free license (e.g. the GPL), it
 will be acceptable for Debian.

 Say what?

 If you GPL a program and don't provide source code, Debian doesn't
 even have the right to distribute it, since they have to distribute
 it with the source code and they don't have that.  The fact that you
 are the copyright owner means that your sending it *to* Debian is
 legal, but that doesn't grant Debian permission to distribute it any
 further.

More precisely, Debian has the right to distribute such a work, but
chooses not to do so.

-- 
Måns Rullgård
m...@mansr.com


-- 
To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Re: GPL photographies, eg for backround

2008-12-29 Thread Måns Rullgård
Don Armstrong d...@debian.org writes:

 On Mon, 29 Dec 2008, Måns Rullgård wrote:
 More precisely, Debian has the right to distribute such a work, but
 chooses not to do so.

 If a work is GPLed and we do not have the complete source for the
 work, we cannot distribute it under the GPL.

If the work as distributed *by the author* lacks something one might
call source, a recipient may still redistribute whatever he received.
When *redistributing* a work, possibly modified, received under the
terms of the GPL, you may of course not omit any parts that would be
considered source for any part of what you do distribute.

 [For non-copyleft works, however, your statement is correct.]

Yes, obviously.

-- 
Måns Rullgård
m...@mansr.com


-- 
To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Re: GPL photographies, eg for backround

2008-12-29 Thread Måns Rullgård
Don Armstrong d...@debian.org writes:

 On Tue, 30 Dec 2008, Måns Rullgård wrote:
 Don Armstrong d...@debian.org writes: 
  On Mon, 29 Dec 2008, Måns Rullgård wrote:
  More precisely, Debian has the right to distribute such a work, but
  chooses not to do so.
 
  If a work is GPLed and we do not have the complete source for the
  work, we cannot distribute it under the GPL.
 
 If the work as distributed *by the author* lacks something one might
 call source, a recipient may still redistribute whatever he
 received.

 That's not correct, unless you're in a locality that has some form of
 the First Sale doctrine. Debian doesn't ever distribute under the
 first sale doctrine, and furthermore, Debian modifies everything that
 is distributed (even if just to package it), so it doesn't apply
 either. [And we certainly don't distribute in 1:1 ratio from the
 copies we obtain from original author.]

 Under GPL v3, when we convey a work in a non-source form, we must
 satisfy all of 6d. That requires making the Corresponding Source
 available, which we cannot.

 Under GPL v2, we distribute under 3(a), and that also requires
 distributing the corresponding machine-readable source code.

 If we don't have the corresponding source, we can't satisfy the GPL,
 so we cannot distribute (GPLv2 §4, GPLv3 §8).

Your argument, if it can be called that, assumes that the requirements
of the GPL, or any license, extend backwards, prior to the point it
was applied.  The extent to which this is true has to be determined by
real lawyers.

For photographs, the argument about what constitutes source can
easily become absurd.  I can easily imagine a photograph where the
preferred form for modification is the depicted scene itself, rather
than its depiction.  To created a modified photo, the photographer
would rearrange the scene and make a new photo, not alter an existing
one.  Does this mean a photo of this scene cannot be distributed under
the GPL (unless the physical scene is also included)?

Similarly, when I write a computer programme, a lot of ideas,
structures, etc. that could be seen as source remain as thoughts in
my brain, never to be written down.  Does this make my programmes
non-distributable?

-- 
Måns Rullgård
m...@mansr.com


-- 
To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Re: The Software shall be used for Good, not Evil.

2010-03-26 Thread Måns Rullgård
Josselin Mouette j...@debian.org writes:

 Le vendredi 26 mars 2010 à 18:43 +0100, Thomas Koch a écrit : 
 Yes, it's this topic again. I've just had a short mail exchange
 with crockford himself. His final answer: If you cannot tolerate
 the license, then do not use the software.
 
 Could you please give me a definitive Yes or No for the below license?

 The Software shall be used for Good, not Evil.

 Definitely non-free, and the author's clarification removes any doubt.

It is certainly a bizarre licence, and it is clearly best to regard it
as non-free.  However, I suspect he'd have a very hard time enforcing
it in court.

IANAL

-- 
Måns Rullgård
m...@mansr.com


-- 
To UNSUBSCRIBE, email to debian-legal-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/yw1xiq8isx7h@unicorn.mansr.com



<    1   2   3