Re: Opinion about GPL-2 exception

[Josselin Mouette]

Le mercredi 30 janvier 2013 à 02:03 +0100, Giulio Paci a écrit : 
> During a package review it came out that the software license includes this 
> statement:
> "Should a provision of no. 9 and 10 of the GNU General Public License be 
> invalid or become invalid, a valid provision is deemed to have been agreed 
> upon which comes closest
> to what the parties intended commercially. In any case guarantee/warranty 
> shall be limited to gross negligent actions or intended actions or fraudulent 
> concealment."
> 
> I contacted the original author and he explained that this statement was a 
> request of their legal department to avoid the possibility that third party 
> can change the
> license of the software (I guess a misinterpretation of the GPL-2 clauses 9 
> and 10). Unfortunately the author is not working anymore for the copyright 
> holder and I am
> having some trouble contacting someone that is allowed to remove the 
> exception.
> 
> What is your opinion about this exception? Is this exception acceptable for a 
> Debian package in main?

It looks like complete nonsense to me.

Since it starts with “should a provision be invalid or become invalid”,
and these provisions have no reason to become invalid (§9 and §10 are
purely informational), I don’t see it a problem, though.

However, regarding compatibility with other GPL components (if there are
any), I wouldn’t be so sure.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'
  `-


--
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/1359550365.27214.614.camel@pi0307572

Re: Unsure If Array 30 Chinese Input Method is DFSG Free or Not.

[Josselin Mouette]

Le mardi 20 décembre 2011 à 11:35 +0800, Yao Wei (魏銘廷) a écrit : 
> The postcard part is removed just today.
> Could you help me reconfirm this license?
> 
> http://www.array.com.tw/company/array_license.pdf
> 
> Though I am not sure if the author wants to adopt a well-known
> license, if it is DFSG free, at least packages including array input
> method could be kept in main.

This new version looks fine to me.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'
  `-


--
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/1324375341.7963.221.camel@pi0307572

Re: a Free Platform License?

[Josselin Mouette]

Le jeudi 24 novembre 2011 à 14:55 -0400, David Prévot a écrit : 
> > I'm looking for a software license, which Debian would
> > support, that actively encourages use of free platforms;
> > and consequently restricts proprietary platforms. 
> 
> If you're looking for a license that “discriminate against [a] group of
> persons” [DGSG5], such as proprietary platforms users, that's not going
> happen in Debian.

WTF?

This is a very, very personal interpretation of DFSG5. One that would
lead to reject any package with a copyleft license.

Actually, it is already widely accepted that the system libraries
exception does not apply to packages in Debian, so a license that would
be the same as the GPL but without this exception wouldn’t make a single
difference for us.

But it would make the hell of a difference for Android/OSX/Windows, of
course…
And because of that, overall I don’t think it’s a bad idea. But you’d
have to convince a lot of people to use it, because it only makes sense
when applied to a killer application, and you can’t know by advance
which ones will be.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'
  `-


signature.asc
Description: This is a digitally signed message part

Re: On using the name Kinect, and fetching a binary firmware

[Josselin Mouette]

Le vendredi 30 septembre 2011 à 12:13 +0200, Antonio Ospite a écrit : 
> 1) Can I use the word "Kinect" in my project name? I read that the
>  name "Kinect" is a trademark of Microsoft Corporation[3], in the
>  project description I make clear that the software is related to
>  the _Microsoft_ Kinect device, but maybe this is not enough, I
>  don't know.

I don’t think that’s a problem, it’s fair use of the trademark.

> 2) For the audio to work a binary firmware is needed; it can be taken
>  from the Micosoft Kinect for Windows SDK[4] but since that is not
>  re-distributable I decided to use a script[5] which downloads and
>  extracts the firmware, the script is going to be called at
>  installation time, is that OK?

It is OK but on 2 conditions: 
  * it needs to check the integrity of the firmware using a secure
hash or equivalent technology; 
  * the package will need to go in contrib.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'
  `-


--
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/1317395902..870.camel@pi0307572

Re: unsourced pdf in tarball; src available from ftp site;

[Josselin Mouette]

Le mardi 29 mars 2011 à 07:25 +0800, Paul Wise a écrit : 
> .doc files are usually binary so you won't be able to include it as a
> patch. Instead I think you can use dpkg-source v3 and include a second
> orig.tar.gz named orig-docsrc.tar.gz (check the dpkg-source manual
> page for info on that). You can then use the upstream pristine
> tarball.

With dpkg-source v3 you can also include binaries in the debian/
directory. Just mention them in debian/source/include-binaries.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling



signature.asc
Description: This is a digitally signed message part

Re: Re: Question about GPL and DFSG Compatibility of a Proposed Amendment to the W3C Document Licence

[Josselin Mouette]

Le jeudi 28 avril 2011 à 17:41 +0200, Lachlan Hunt a écrit : 
> However, since GPL compatibility is a requirement for the licence, I'm 
> hoping the W3C can be convinced to give up on these alternatives if it 
> can be proven that they are not compatible.

If GPL compatibility is a requirement, they should add an option 4: the
GPL.

End of trouble.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'
  `-


--
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/1304064968.32026.72.camel@pi0307572

Re: distributing a restricted branding icon OK?

[Josselin Mouette]

Le mardi 15 mars 2011 à 11:54 -0500, Gabriel Burt a écrit : 
> Is changing http://www.emusic.com/favicon.ico to a PNG "modifying" it?
> 
> Assume it's not, would we be OK including that image in our Debian
> package of Banshee?

No, it would not. This icon is not free, in terms of copyright - and
that’s regardless of any trademark issues.

-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1300210196.30617.179.camel@meh

Re: Bug#570621: Parsing output = derivative work?

[Josselin Mouette]

Le mercredi 09 mars 2011 à 11:27 +, Noel David Torres Taño a
écrit : 
> > Miriam Ruiz  wrote:
> > In general, I do agree with Miriam that parsing the output of another
> > program does not make a derivative work.  But just to give an example
> > of where it does happen, git is largely comprised of many small
> > utilities that communicate over pipes and command-line arguments.
> 
> It happens (to me) to be as simple as this:
> 
> If the parsed program is susbtituted by a clone will the parsing program 
> continue working?

If the shared library is substituted by a clone reimplementing the same
API/ABI, will the program linking to it continue working?

> If the answer is yes, since it is absurd that the parsing program is a 
> derivative work of all possible clones at the same time, then clearly it is 
> not a derivative work. If the answer is no, then clearly the parsing program 
> depends on the parsed one 'in an intimate way'.
> 
> Can a clone or sucessor of nmap be used with gnetworktester?

I agree that if such clone exists (and that holds for libraries too), it
is not a derivative work.

Nothing specific to parsing the output of another program here.

-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1299680294.21760.94.camel@meh

Re: The "Evil Cookie Producer" case

[Josselin Mouette]

Hi, and thanks for the effort of providing the complete information.

Le lundi 07 mars 2011 à 09:31 +, Andrew Ross a écrit : 
> The software itself is the current version of iText, which is licensed
> under the AGPL with the following additional term:
> 
> "In accordance with Section 7(b) of the GNU Affero General Public
> License, you must retain the producer line in every PDF that is created
> or manipulated using iText."

I’m not sure how to interpret this restriction. If it is in accordance
to Section 7b, it means you cannot remove from iText the code that
produces this line in PDF files. You are still free to remove the line
in the generated PDF file if you want.

If the intent is to restrict modifications that can be made on generated
PDF files themselves, the license is clearly non-free (and considered a
“further restriction” in the AGPL).

Cheers,
-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1299599115.18970.15.camel@meh

Re: Bug#570621: Parsing output = derivative work?

[Josselin Mouette]

Le mardi 08 mars 2011 à 07:30 -0800, Ken Arromdee a écrit : 
> >> Parsing the output of a program doesn’t make a derivative work. However,
> >> if this parsing is vital for the operation of the application and makes
> >> it useless without that program, what is the difference with dynamic
> >> linking to a library? To a programmer, there might be one, but to a
> >> court, there wouldn’t be any.
> 
> By this reasoning, if I write a program which converts another word 
> processor's
> output to Microsoft Word format, then that program is a derivative of
> Microsoft Word, at least until Open Office gets a filter good enough to read
> it.

This is a completely unrelated case. Functionally, such a program can
work without Microsoft Word.

> Moreover, by this reasoning, if I write a program that runs only on Windows,
> or which interfaces with some proprietary Windows protocol, Microsoft can
> legitimately claim that I am violating their copyright by creating an
> unauthorized derivative of their work.

Microsoft gives you explicit permission to link to the system libraries
provided with Windows.

> This definition of "derivative work" is something which the FSF claims, but
> which many people outside the FSF are skeptical of precisely because of
> absurd consequences like these.

If you want to prove something is absurd, please point to absurdities
first.

-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1299598463.18970.8.camel@meh

Re: The "Evil Cookie Producer" case

[Josselin Mouette]

Le dimanche 06 mars 2011 à 12:43 +0100, Bruno Lowagie a écrit :
> I'm confronted with a very special case for which I've written a
> metaphor. I'd like your opinion on this case.

This is not a metaphor. It is about something that is not software, for
which you try to apply software reasoning. I fail to see where you want
to go with this.

> Now suppose that nuts are part of the ingredients list of the recipe
> created by Company A. To protect cookie consumers, Company A adds an
> extra term to the AGPL: as long as there are traces of nuts down the
> production line, all products that may contain traces of nuts must
> retain the notice "may contain traces of nuts".

Copyright (and copyleft) isn’t here to protect consumers, it is here to
protect your intellectual property.

This kind of protection should be provided by law. In many countries
there a rule that forces resellers to warn consumers when there are
traces of nuts in their products. You can’t take the place of such a law
by using your copyright license as a vessel.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling



signature.asc
Description: This is a digitally signed message part

Re: Parsing output = derivative work? (was: RFS: gnetworktester)

[Josselin Mouette]

Le dimanche 06 mars 2011 à 12:04 +0100, W. Martin Borgert a écrit : 
> (out of curiosity moved to debian-legal)
> 
> On 2011-03-05 23:46, Timo Juhani Lindfors wrote:
> > gnetworktester seems to parse the output of nmap and nmap upstream at
> > http://insecure.org/nmap/data/COPYING gives me the impression that
> > gnetworktester would thus be "derivative work".
> 
> IANAL, but since when parsing the output of another program
> constitutes a derivative work? 

The distinction between a derivative work and a separate work is not
based on technology but on functionality. 

Parsing the output of a program doesn’t make a derivative work. However,
if this parsing is vital for the operation of the application and makes
it useless without that program, what is the difference with dynamic
linking to a library? To a programmer, there might be one, but to a
court, there wouldn’t be any.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling



signature.asc
Description: This is a digitally signed message part

Re: S3TC patent violation in multiple packages?

[Josselin Mouette]

Le lundi 27 décembre 2010 à 23:22 +, Wols Lists a écrit :
> But it shouldn't be too hard, in UK courts at least, to file to have it
> dismissed in summary judgement. I don't know about other European
> jurisdictions, but we do have case law to back up the European Patent
> Treaty ban on software patents.

Patents on picture compression are among the few kinds of software
patents that can be valid in Europe. They would, of course, be much
harder to enforce than in the US, but they are, unfortunately, legal.

-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1293539690.14833.26.ca...@meh

Re: S3TC patent violation in multiple packages?

[Josselin Mouette]

Le samedi 25 décembre 2010 à 13:58 +0100, Rudolf Polzer a écrit : 
> the S3TC texture compression scheme is covered by an US patent, as described 
> in
> http://lists.freedesktop.org/archives/nouveau/2010-March/005439.html
> 
> The following Debian packages in "main" seem to contain S3TC decompression 
> code:

Don’t forget that compression and decompression are different
algorithms. Regardless of the complexity of a compression algorithms,
decompression ones are usually trivial in comparison, and not covered by
patents. This is for example why we ship tons of MPEG decoders in main.

As for the compression algorithms, I can only agree with Julien: we can
only care if the patent is being actively enforced and considered valid.
Otherwise we might as well stop distributing anything.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: Inappropriate use of Debian logo.

[Josselin Mouette]

Le mercredi 01 décembre 2010 à 17:30 +0100, Mike Hommey a écrit :
> > There is no question in my mind that the blue logo under debate is a copy of
> > the Debian logo, followed by a color change.  There are too many degrees of
> > freedom (beginning of the swirl path, brush thickness, etc).
> 
> You swirled it too much. Other than that, it's the same.

The point is that to get the exact result, pixel per pixel, you have to
use exactly the same settings, and there are too many of them for it to
be a coincidence.

-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1291221697.1756.6.ca...@meh

Re: Inappropriate use of Debian logo.

[Josselin Mouette]

Le lundi 29 novembre 2010 à 11:23 +0100, Josselin Mouette a écrit :
> Le dimanche 28 novembre 2010 à 10:36 +0100, Alessandro Rubini a écrit :
> > > http://imgur.com/gFKfs.jpg
> > 
> > Thank you for making this jpeg, it's very clear.
> > 
> > > [...]
> > > "The comapny Logo was created by photoshop and Logo software, we desgined 
> > > it
> > > from the stretch. if you have somethins to say, give us a call."
> > 
> > Unfortunately, they may be right and in good faith.
> > This message confirms the swirl is just one of the defaults:
> > http://lists.debian.org/debian-legal/2005/06/msg00340.html
> 
> This is, again, *completely irrelevant*.
> 
> I can reproduce the GAP logo in 3 clicks in Inkscape. Does it mean I can
> use it for my own clothing shop? NO, of course.
> 
> A trademark is a trademark. If we don’t enforce it, we lose it.
> 
> I hope this time, for such blatant violation, SPI takes action.

Stefano just mentioned that it would be better to CC him if we want
something to be done to do something. So here we go.

Cheers,
-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1291124425.25881.7.ca...@meh

Re: Inappropriate use of Debian logo.

[Josselin Mouette]

Le dimanche 28 novembre 2010 à 10:36 +0100, Alessandro Rubini a écrit :
> > http://imgur.com/gFKfs.jpg
> 
> Thank you for making this jpeg, it's very clear.
> 
> > [...]
> > "The comapny Logo was created by photoshop and Logo software, we desgined it
> > from the stretch. if you have somethins to say, give us a call."
> 
> Unfortunately, they may be right and in good faith.
> This message confirms the swirl is just one of the defaults:
>   http://lists.debian.org/debian-legal/2005/06/msg00340.html

This is, again, *completely irrelevant*.

I can reproduce the GAP logo in 3 clicks in Inkscape. Does it mean I can
use it for my own clothing shop? NO, of course.

A trademark is a trademark. If we don’t enforce it, we lose it.

I hope this time, for such blatant violation, SPI takes action.

-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1291026208.16643.25.ca...@meh

Re: Problem with Squeeze Artwork

[Josselin Mouette]

Le samedi 13 novembre 2010 à 00:04 +0100, Francesco Poli a écrit : 
> However, I think there's an issue that would prevent its inclusion in
> Debian main, unless some images are modified.
> A number of images in this theme include a derivative work of the
> Debian Open Use Logo with "Debian":
> http://www.debian.org/logos/openlogo.svg
> which, unfortunately, does *not* comply with the DFSG:
> http://www.debian.org/logos/#open-use

Note that this is not a regression from lenny, since moreblue-orbit has
the same problem.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: CodeIgniter license

[Josselin Mouette]

Le lundi 01 novembre 2010 à 10:24 +0100, Francesco Poli a écrit : 
> Back to the initial topic: I personally think that a license which
> forces me to disclose my own identity in order to distribute a file
> modified by me fails DFSG#1, since being forced to disclose one's own
> identity can be a fee.

This is a very far-fetched interpretation of DFSG#1. I would say that it
fails DFSG#5, maybe #7.

However, It would be interesting to know in which countries this is a
real concern. In French copyright law, pseudonyms are legally an
identity when it comes to copyright. So a license that requires you to
disclose your identity doesn’t actually require you to give the name on
your passport.

If a local law doesn’t acknowledge pseudonyms for copyright assignment,
what happens to works distributed under a pseudonym? Are they
automatically illegal? made public domain?

-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: is RtMidi license DFSG-free?

[Josselin Mouette]

Le dimanche 17 octobre 2010 à 21:00 +0100, MJ Ray a écrit : 
> Miriam Ruiz wrote:
> > Especially this part: "Any person wishing to distribute modifications
> > to the Software is requested to send the modifications to the original
> > developer so that they can be incorporated into the canonical
> > version.", can it be considered DFSG-free?
> 
> It looks like a request not a requirement, so it's fine as far as I
> understand the DFSG.  I have not tried a wdiff against other licences
> to see if there's another subtle change.

It’s a request, but it’s listed in the conditions. I think you should
really ask upstream for clarification here.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: CDDL/GPL and Nexenta (with CDDL libc)

[Josselin Mouette]

Le jeudi 23 septembre 2010 à 20:42 +0100, Stephen Gran a écrit : 
> This one time, at band camp, Florian Weimer said:
> > * Don Armstrong:
> > 
> > > CDDL'ed libc (and other System Library) and GPLv3+ work: OK
> > 
> > I think the FSF wants us not to be able to use the System Library
> > exception.  It is only intended for proprietary operating systems.
> 
> Does no one else see a license exception that is easier for a
> proprietary OS than a free one as problematic?

Well, it would be if that was the case. But if a proprietary software
vendor wanted to distribute a GPL program built on proprietary
libraries, he could not. 

The constraints for a CDDL’ed OS are the same as for a proprietary one.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: [Lame-dev] LAME license

[Josselin Mouette]

Le jeudi 26 août 2010 à 23:27 +0200, Gabriel Bouvigne a écrit :
> > http://permalink.gmane.org/gmane.comp.video.ffmpeg.devel/115654
> > 
> > Do you agree with this claim?

> Without ressorting to any court, I share the opinion that every file 
> featuring the usual LPGLv2 header is only convered by the LPGLv2, 
> without any consideration of the two problematic notes located within 
> the readme file.
> To me, those should only apply to "headerless" source files. It is even 
> likely that several of us haven't even noticed/considered those two 
> notes when commiting new files, faithfully willing to commit new files 
> only under LGPLv2. (at least that is the case for me).

I think this is a valid claim, but it also makes lame as a whole
undistributable, since you cannot link together code that is licensed
under regular LGPLv2+ and code that is under LGPLv2++restrictions, since
the restrictions are incompatible with the LGPL.

> > As a last resort, I think we can still redistribute lame under the terms
> > of the LGPLv3, which seems the be permitted by the LAME license. The
> > GPLv3 contains terms that are very similar to the modifications quoted
> > above in §11. This should address the "additional restrictions" concern.
> > 
> > Do you agree with this theory?
> 
> I think that you can not redistribute LAME only under LGPLv3. According 
> to LGPLv2 you can choose to apply LGPLv3 to LAME, but you should not be 
> able to remove its LGPLv2-ability.

I don’t think this is true. When a work is licensed under the LGPL vX
“or any later version”, you can simply start redistributing it under a
later version.

But anyway, as I understand it, the additional restrictions to the LGPL
are regardless of the LGPL version.

> BTW, while the note #2 is an "activist" one, note #1 is a "pragmatic 
> one". LGPLv2 theoritically prevents any patent holder or patent to 
> distribute LAME. That is a huge problem for some companies, and goes far 
> beyond only LAME.

It only prevents redistribution by holders of a patent that applies to
LAME, or by those who have bought a patent license. I think this is a
desirable side effect.

And anyway, if it only applies to headerless files, the LGPL clauses
apply anyway.

Cheers,
-- 
 .''`.
: :' : “You would need to ask a lawyer if you don't know
`. `'   that a handshake of course makes a valid contract.”
  `---  J???rg Schilling


--
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/1282894922.25321.26.ca...@meh

Re: Plugins for non-free software in orig.tar.gz

[Josselin Mouette]

Le dimanche 04 juillet 2010 à 12:34 +0200, Francesco Poli a écrit :
> I wonder whether those "shaders" should be considered as purely
> functional in nature, and thus not copyrightable: I seem to recall that
> there's U.S. case law about a functionality exception to copyright
> protection, and the following was stated:
> 
> |  not much we can do about that, if we dont' ship them we're not 
> | compliant.
> 
> which seems to suggest that those "shaders" are necessary to comply to
> a software standard (namely, the RenderMan interface standard defined
> by Pixar Animation Studios).

These are not only interface descriptions like C headers, which are
indeed not copyrightable. However the code pieces are absolutely
trivial.

IMHO the “clean” solution is to rewrite the shaders from scratch by only
looking at the standard. Given the triviality you will end up writing
the same files with only non-algorithmic changes.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: Plugins for non-free software in orig.tar.gz

[Josselin Mouette]

Le jeudi 01 juillet 2010 à 00:40 +0200, Manuel A. Fernandez Montecelo a
écrit :
> > Of course, the quickest and easiest solution, until the licensing is
> > clarified, is dropping the scripts from the package...
> 
> First of all, I need to talk to the author personally when he's available, 
> but I think that this script is not useful to be shipped in a binary anyway, 
> so it's not clear to me why it should be there.  Think of it as a plugin 
> that you have to copy/import/whatever-is-called into Blender modeler so 
> Bender is able to export to .xad format, or invoke Aqsis as rendered of the 
> scene previously modeled in Blender.  Or a GIMP plugin to save in new image 
> format .ghda.  As I understand it, it's of no use unless you have Houdini 
> installed, and then you have to install it in Houdiny as a plugin (not sure 
> if compiled in some way, or not processed at all).
> 
> Maybe they should provide a separate package with it and that's all, and not 
> bothering shipping it in normal source packages of the rest of Aqsis, and 
> thus avoid installing it with normal CMake building system.

As long as the script itself is free, there is no point in removing it
from the source package. Just don’t ship it in the binary. The
requirement to remove non-free files is for non-free files. Not for
files that relate to non-free software.

For a similar example, we ship Visual Studio build files in a lot of
packages. Believe me, upstream is not going to remove them, and we are
not going to remove them either. We just don’t use them to build the
package, and we don’t ship them in the binaries.

> 1) I should still remove it from orig.tar.gz, otherwise Debian would 
> continue distributing it and potentially breaking the license, right?

You should remove it only if it is non-free. Otherwise the rule is to
keep the upstream tarball.

> 2) Should I do anything special other than that?  Explain the case in a 
> README.Debian?  Name the orig.tar.gz in a specific way?  Use 'dfsg' in the 
> name of orig.tar.gz, source or binary packages in some ways.

If you remove some files, it is appreciated to append "+dfsg" to the
upstream version, and to explain what you did in README.source.

> $ head -7 shaders/surface/metal.sl 
> /* metal.sl - Standard metal surface for RenderMan Interface.
>  * (c) Copyright 1988, Pixar.
>  *
>  * The RenderMan (R) Interface Procedures and RIB Protocol are:
>  * Copyright 1988, 1989, Pixar.  All rights reserved.
>  * RenderMan (R) is a registered trademark of Pixar.
>  */
> 
> I tried to find the answers in Pixar's website, but I found none.  I think 
> that it's a kind of OpenGL, but I don't know if things like "All rights 
> reserved" means that they can forbid people using them, like the (already 
> several) FOSS implementations (aqsis, pixie, jrman).

This looks much more worrisome to me. There is no license for use or
redistribution here. I’m surprised the FTP masters let this slip
through.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: One-line licence statement

[Josselin Mouette]

Le mercredi 21 avril 2010 à 19:28 +0200, Franck Joncourt a écrit : 
> --
> Copyright (c) 1999 by Megginson Technologies.
> Copyright (c) 2003 Ed Avis 
> Copyright (c) 2004-2010 Joseph Walton 
> 
> No warranty.  Commercial and non-commercial use freely permitted.
> --

This is clearly non-free, since it doesn’t allow modification and
redistribution.

> As it is intended to be as close to public domain as legally possible, he
> pointed me out to the following URL:
> 
> >From <http://linuxmafia.com/faq/Licensing_and_Law/public-domain.html>:
> 
> > All alleged advantages of a "public domain dedication" can be
> > gained without uncertainty using a regular one-line licence statement,
> > e.g., "Copyright (C) 2008 Owner Name. Do whatever you want with this work."
> 
> Quoting upstream:
> "This is exactly the form, and intent, of the original XML::Writer licence:
> you may "Do whatever you want with this work."

No. You may not, since it only permits use.

> As I am not sure, which form of language would be the best to achieve this
> goal?

The simplest way to achieve that is probably the WTFPL.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

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

[Josselin Mouette]

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.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'  “If you behave this way because you are blackmailed by someone,
  `-[…] I will see what I can do for you.”  -- Jörg Schilling


signature.asc
Description: This is a digitally signed message part

Re: CMU LTI Licence

[Josselin Mouette]

Le vendredi 22 janvier 2010 à 21:37 +1100, Ben Finney a écrit :

As already said, the rest is very similar to a 4-clause BSD.

> ##  5. Any commercial, public or published work that uses this data  ##
> ## must contain a clearly visible acknowledgment as to the   ##
> ## provenance of the data.   ##

In the spirit it’s very similar to the advertising clause, but it’s at
the same time more broad in the application, and more vague in what is
required.

For example, what is required for someone who sells Debian CDs
containing this data? Should the acknowledgment go into the package
descriptions? In the documentation? On the CD itself?

I’d be more confortable with shipping this in Debian if the clause was
more clear; or at least, if it was asking to refer to the data only in
relevant places.

Cheers,
-- 
 .''`.      Josselin Mouette
: :' :
`. `'   “A handshake with whitnesses is the same
  `- as a signed contact.”  -- Jörg Schilling


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

Re: mass bug filling about moving toi main

[Josselin Mouette]

Le jeudi 24 septembre 2009 à 10:22 +0200, Bastien ROUCARIES a écrit : 
> It is a nice news :)
> 
> I plan to post a mass bug filling moving all contrib package depending on 
> cmap 
> to main.

Whatever you are talking about, it looks like good news, but do you have
any details? Do you know whether this also affects stuff like
poppler-data which are in non-free because of CMap files?

Please also post, per developer-reference, a dd-list of affected
packages to debian-devel.

Thanks,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: Internet2 licence

[Josselin Mouette]

Le samedi 05 septembre 2009 à 18:52 +0200, Francesco Poli a écrit : 
> Well, not completely a copyleft license, IMHO.
> It says:
> 
> |  If you choose to provide your enhancements, or if you choose to
> |  otherwise publish or distribute your enhancements,
> [in source form and without doing something special],
> |  then you thereby grant Internet2 and its contributors a
>   ^^
> |  non-exclusive, royalty-free, perpetual license to
> [do many things, including sub-license].
> 
> Hence, it does _not_ require you to grant *any recipient* the *same*
> rights you received.
> It says that, when you distribute enhanced source (without doing a
> special action), you will automatically grant the *original authors* and
> contributors *more* rights than those you received from them.
> 
> This is not a classical (strong or weak) copyleft mechanism, AFAICT.
> This is something different.

Indeed, I have overlooked that. Even if everyone would be granted those
rights, they would be subject to the copyleft while the original authors
are not.

So in short, it allows the original authors to create a proprietary
derivative while it does not allow others to do it. I agree with your
reasoning and the conclusion this clearly violates DFSG #3.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: Internet2 licence

[Josselin Mouette]

Le mercredi 02 septembre 2009 à 16:03 +0200, Carsten Wolff a écrit : 
> Hi,
> 
> I have a project here based on code under the "Internet2" license and I was 
> wondering, if it is any different from the 3-clause BSD license.
> 
> If it's no different, can I leave out the additional explanatory section 
> ("You 
> are under no obligation[..]in binary and source code form.") in derivative 
> work and just license the whole derivative under BSD?

It is very different since this additional section makes it a copyleft
license, which makes it incompatible with most other copyleft licenses.

> You are under no obligation whatsoever to provide any enhancements to
> Internet2 or its contributors.  If you choose to provide your
> enhancements, or if you choose to otherwise publish or distribute your
> enhancements, in source code form without contemporaneously requiring
> end users to enter into a separate written license agreement for such
> enhancements, then you thereby grant Internet2 and its contributors a
> non-exclusive, royalty-free, perpetual license to install, use,
> modify, prepare derivative works, incorporate into the software or
> other computer software, distribute, and sublicense your enhancements
> or derivative works thereof, in binary and source code form.

The “Internet2 and its contributors” choice of words is poor, but
otherwise it sounds like a reasonable and free copyleft license.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: Serious problem with geoip - databases could not be build from source

[Josselin Mouette]

Le mercredi 26 août 2009 à 11:09 +0200, Bjørn Mork a écrit : 
> You are aware of the GPLv3 licensed database at
> http://software77.net/geo-ip/ ?

Woohoo, nice. Combine this with Julien’s idea to implement support for a
new database format, and I think you have the correct solution.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: License of CORBA Interface Definition Files published by the Object Management Group

[Josselin Mouette]

Le mardi 25 août 2009 à 02:52 +0200, Ludovic Brenta a écrit : 
> The source package orbit2_2.14.17.orig.tar.gz shipped by Debian contains
> the following files that concern me:
> 
> src/idl/CORBA_PIDL/CORBA_Request.idl
> src/idl/CORBA_PIDL/pseudo_orb.idl
[snip]

> The debian/copyright file in the package does not explicitly state a
> license for those files but implies that the license is the GPL.  The
> package is in main.

My opinion is that the headers themselves are not subject to copyright.
They are just the formal description of a specification, there is
nothing creative in them. However the comments are, so maybe we have to
strip the comments from those files.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: Is "IPA Font license" DFSG-Free?

[Josselin Mouette]

Le dimanche 31 mai 2009 à 20:52 +0900, Hideki Yamane a écrit :
>  I've ITPed IPAfont as otf-ipafont package.

>  You can see its license at http://www.opensource.org/licenses/ipafont.html
>  Please give me your feedback (Please add CC to me). Thanks.

The only things that looks suspicious are the name change clauses.

For derived works:
No one may use or include the name of the Licensed Program as a
program name, font name or file name of the Derived Program.

And for redistribution without modification:
The Recipient may not change the name of the Licensed Program.

So if there are any changes, the name must be changed, and it must not
be changed if there are no changes. For a regular computer program, that
would imply iceweaselization, but for a font this seems reasonable: we
have no practical reason to patch it in our packages, and most font
systems make it easy to alias a font with another one so it’s fine for
those who modify it.

Otherwise, it’s a simple license with a strong copyleft, which should be
fine for Debian.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: legal questions regarding machine learning models

[Josselin Mouette]

Le mercredi 27 mai 2009 à 00:36 +0200, Francesco Poli a écrit :
> > Of course, the decision is up to the FTP masters, but I think this
> > should be accepted for the sake of consistency with things we already
> > cannot decently exclude from the archive.
> 
> I instead think that FTP masters should change their minds about 2D
> images rendered from 3D models.

I suggest you start your own distribution, in which you won’t ship:
  * xfonts-* (bitmap renderings of non-free vector fonts)
  * all icons shipped without SVG source 
  * all pictures shipped without XCF/PSD source (oh yeah, that makes
a lot)
  * actually, all pictures that are initially photographs of an
object (the preferred form of modification is the original
object; if you want to see it at another angle, you need to take
another photograph)
  * all sound files shipped without the full genetic code of the
speaker

You could call it something like gNewSense, and you could discuss during
hours with RMS how much better it is this way.

> Disclaimers, of course: IANADD, TINASOTODP (and IANAL, TINLA).

If you really feel the urge to add meaningless acronyms to all your
emails, please do so in your signature.

-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: legal questions regarding machine learning models

[Josselin Mouette]

Le mercredi 27 mai 2009 à 01:17 +0900, Mathieu Blondel a écrit :
> For efficient storage, the model may be stored in binary format but
> human-readable formats (such as XML) may be used, thus allowing easy
> access to the parameters of the models.
> 
> My first question is : is it possible to distribute the model under a
> free software license without distributing the original data that were
> used to train the model? Likewise, is it possible to package directly
> a model in Debian? Although it's very unlikely, I could pretend that I
> found the parameters of the models by hand. In that case, the
> parameters can be seen as "magical numbers" with no explanation
> whatsoever as to how I found them.

This looks very similar to distributing a picture which is a 2D
rendering of a 3D model without distributing the original model. This is
already accepted in the archive, and the reason is that a 2D picture is
its own source, and can serve as a base for modified versions this way.

The same reasoning applies to the model: as long as it is useful to tune
the parameters by hand to produce derived versions, there’s no reason
not to consider it as the source.

Of course, the decision is up to the FTP masters, but I think this
should be accepted for the sake of consistency with things we already
cannot decently exclude from the archive.

> My second question is: Given the difficulty to prove what data were
> actually used to train a model, how can we prevent non-free software
> to use free data such as those of Voxforge?

A widely-used technique is to cleverly hide some minor bugs in the data.
If a non-free model shows the same bugs, you can prove the data was used
illegally. Of course this only works if you manage to keep the bugs
secret.

Cheers,
-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: Dirent.cpp - Is it DFSG-free GPLv2 only compatible

[Josselin Mouette]

Le vendredi 24 avril 2009 à 02:02 +0100, Dmitrijs Ledkovs a écrit :
> It looks similar to BSD license but it doesn't say anything about
> binary / source distribution. Please tell me if you identify this
> license by a known name.

> Permission to use, copy, modify, and distribute this software and its
> documentation for any purpose is hereby granted without fee, provided
> that this copyright and permissions notice appear in all copies and
> derivatives, and that no charge may be made for the software and its
> documentation except to cover cost of distribution.

“No charge may be made” → clearly non-free

-- 
 .''`.  Josselin Mouette
: :' :
`. `'   “I recommend you to learn English in hope that you in
  `- future understand things”  -- Jörg Schilling


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: GPL2 vs. GPL3

[Josselin Mouette]

Le mardi 14 avril 2009 à 18:53 +0200, dominik.smat...@gmail.com a
écrit :
> Dear legal gurus,
> 
> is it possible to publish software under GPL3,  if this software is
> depending on some GPL2 libraries?
> 
> Sorry, I was trying to read GPL3, but my English is not good enough to
> figure this out by myself.

Very short summary: if the libraries are “GPL version 2 or later”, you
can. If they are only GPL version 2, you can’t.

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: GCC 4.4 run-time license and non-GPLv3 compilers

[Josselin Mouette]

Le vendredi 10 avril 2009 à 14:35 +0200, Florian Weimer a écrit :
> At least with a strict interpretation, the run-time exception suffers
> from a significant issue with compilers which are not licensed under a
> GPLv3-compatible license (such as the GPLv2, or the QPL), and which
> are implemented in the language itself.  (One such compiler is
> Objective Caml.)  After compilation with GCC 4.4, such a compiler is a
> "work based on GCC" because it links to the GCC run-time library.

No, the exception is here precisely to make it not so. The ocamlopt
binary was generated through an eligible compilation process, and as
such you can choose your terms for redistribution.

> Therefore, it's output cannot use the run-time library exception (it's
> not the result of an Eligible Compilation Process because it's neither
> the result of running "GCC, alone or with other GPL-compatible
> software," nor "it is done without using any work based on GCC"), and
> the resulting binary is covered by the GPLv3 (potentially among other
> licenses).

I don’t think you can say this is a work based on GCC just because of
that.

Anyway, clarification from the FSF would be better, but I don’t think
this is violating the spirit of the license, so this can be fixed later
on, and this is not a reason not to make GCC 4.4 the default.

Cheers,
-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: Zimbra and Yahoo Public License

[Josselin Mouette]

Le lundi 23 mars 2009 à 17:16 +0100, Cedric Fachinetti a écrit :
>   * 3.2 - In any copy of the Software or in any
> Modification you create, You must retain and
> reproduce, any and all copyright, patent, trademark,
> and attribution notices that are included in the
> Software in the same form as they appear in the
> Software. This includes the preservation of
> attribution notices in the form of trademarks or logos
> that exist within a user interface of the Software.

This looks like a restriction in modifications, a distorted way to
introduce invariant data.

If the software includes such logos or trademarks in the user interface,
I think this is clearly non-free. Otherwise, it is not a problem.

> Term and Termination 
>   * 6.1 - This Agreement will continue in effect unless
> and until terminated earlier pursuant to this Section
> 6.
>   * 6.2 - In the event You violate the terms of this
> Agreement, Yahoo! may terminate this Agreement.

This looks like a lawyerbomb, as the wording suggests Yahoo! can reserve
the rights to revoke the license if they think you have violated it.

> All disputes arising out of this Agreement involving
> Yahoo! or any of its subsidiaries shall be subject to
> the jurisdiction of the federal or state courts of
> northern California, with venue lying in Santa Clara
> County, California. 

This is a choice of venue clause. There’s controversy over it, and many
people consider it non-free. I disagree in the general case, but
together with the 6.2 termination clause, it puts an unacceptable burden
on the recipient and clearly fails the “tentacles of evil” test.

Therefore I think this license is non-free.

Cheers,
-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The copyright of a keyboard mapping and its implementation

[Josselin Mouette]

Le mardi 17 mars 2009 à 23:13 -0700, Steve Langasek a écrit :
> Unless there are copyrightable comments, which can be stripped out, I don't
> think there's anything here that we should recognize as covered by copyright
> at all.  Keep the keyboard mapping we currently have in X if it's useful
> (correcting the claim that there are restrictions on modification), or
> replace it with one that works better for the users.

The best for users is probably to have both, and that’s what several
people have been requesting to upstream:
  * One is compatible with the Windows mapping.
  * One has been here for longer and some users consider it better
as it doesn’t require dead keys.

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: The copyright of a keyboard mapping and its implementation

[Josselin Mouette]

Le mardi 17 mars 2009 à 14:54 +, MJ Ray a écrit :
> In general, I'd support this.  Further, j...@debian's implementation
> included at X.org is not subject to the other holder's copyright, so
> I'd report the bogus CC-NC-ND layout claim as a bug to X.org.
> 
> Has that happened?  This thread hinted at previous discussions at
> X.org, but I haven't found it yet - where was it?

http://bugs.freedesktop.org/show_bug.cgi?id=15407

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message numériquement signée

Re: The copyright of a keyboard mapping and its implementation

[Josselin Mouette]

Le lundi 16 mars 2009 à 11:18 -0700, Don Armstrong a écrit :
> Is there any hope of getting Leboutte to license this under CC without
> the NC and ND clauses or retract his claims?

I don’t think so, but maybe an open source evangelist would have better
luck.

> Alternatively, can someone generate a clean-room implementation of the
> appropriate layout?

What do you mean by “clean-room”? Both X11 implementations were made
from scratch, it’s just that Francis Leboutte claims they are a derived
work of his layout.

> I'm of the opinion that if we are to distribute the layout, and we are
> unable to do the two things above, we need to get legal advice, and
> then specificially ignore the ND clause by distributing a derived
> version. I'm slightly concerned about this layout hanging around and
> then a small company who uses it because it was distributed in Debian
> being sued.

I think we’re not at risk of anyone being sued as long as we don’t
distribute a derived version. However this particular requirement makes
the layout non-free.

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message	numériquement signée

The copyright of a keyboard mapping and its implementation

[Josselin Mouette]

Hi,

there are two available layouts for French Dvorak keyboards. One of
them, which is the cause of my concern, was written by Francis Leboutte,
and was originally distributed as a (non-free) Windows driver.

I started then to make another implementation of the same mapping, for
X11. It soon turned out that the mapping forces to use dead keys, which
sucks, so I used a variant that removes this need. Several people asked
me of a French Dvorak layout, so I started to distribute it under the
X11 license, and it finally ended up in the official X.org tarballs.

Later, Francis Leboutte started to license the *layout* (not only the
Windows driver) under a clearly non-free license (CC-NC-ND), and asked
X.org to remove the driver, arguing that the X11-licensed version is
illegal, being a derivative work of his layout. The X.org guys finally
agreed to distribute the original variant instead, with the following
licensing header:
// Licence  : X11 (the layout itself is released under CC-NC-ND licence)

It is my opinion that, in European law, the copyright on a keyboard
mapping does not affect that of its implementation, because, among other
things, of the interoperability exception – in the same way a function
prototype is not subject to copyright while the function itself is.
Francis Leboutte believes the opposite and claims that drivers
implementing a derivative layout are illegal. I don’t think this
disagreement will resolve, since it could only happen in a court, and we
both have better things to do than suing each other for a keyboard
mapping.

However, the Francis Leboutte mapping is now included in Debian. This
means we should settle on this issue: if we consider it non-free, we
must remove this layout (and all derivatives) from the distribution; if
we don’t, there’s no barrier against including some variants. I’d tend
to say we should opt for the conservative approach and remove them;
despite the fact that I like the mapping, we shouldn’t include software
with such an unclear copyright status.

Thoughts anyone?

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Why is OpenSSL not in non-free?

[Josselin Mouette]

Le mercredi 25 février 2009 à 14:24 +0200, Adrian Bunk a écrit :
> > Even the FSF considers it free.
> 
> The FSF also considers the GFDL with invariant sections as free...

They clearly don’t consider it as a free software license. The FSF
argues that documentation doesn’t need the same freedoms as software.

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Why is OpenSSL not in non-free?

[Josselin Mouette]

Le mercredi 25 février 2009 à 12:46 +0200, Adrian Bunk a écrit :
> - the 4-clause BSD license with the advertising clause is considered
>   non-free

No.

Even the FSF considers it free.

-- 
 .''`.  Debian 5.0 "Lenny" has been released!
: :' :
`. `'   Last night, Darth Vader came down from planet Vulcan and told
  `-me that if you don't install Lenny, he'd melt your brain.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: enabling transport and on storage encryption in bacula on debian build

[Josselin Mouette]

Le dimanche 11 janvier 2009 à 21:25 +0100, Hendrik Weimer a écrit :
> The only
> case I am aware of where another distro refuses to distribute a
> package found in Debian is Fedora's stance on afio. If you know of
> other cases, I would be interested to learn about them.

There’s also the case of MP3 decoders in Red hat.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: BSD license with Mozilla-style name clause

[Josselin Mouette]

Le mercredi 07 janvier 2009 à 23:19 +0100, Francesco Poli a écrit :
> > I think it would also be enough to obtain a permission from the authors
> > to call Debian modified versions "Alice", as long as renaming it is easy
> > otherwise. We have allowed such things in the past.
> 
> I don't think the situation is crystal clear.

I don’t think either, but you are really nitpicking too much. So you
can’t call a derived version FreeAlice nor Alice? Who cares? The same
goes with trademarks of several packages we distribute.

Luke, if you are in touch with upstream, I think you could recommend
them to remove this clause and replace it by a trademark policy on the
"Alice" name. It is more efficient and more flexible at the same time.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: BSD license with Mozilla-style name clause

[Josselin Mouette]

Le mercredi 07 janvier 2009 à 09:25 -0500, Luke Faraone a écrit :
> Hi, I'm interested in packaging Alice (RFP:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=500648), but it's
> license  has naming
> restrictions similar to Mozilla.
> 
> Would retitling the package "carol" or "wonderland" be sufficient to
> make the package DFSG-free?

Yes, it’s otherwise an old-style BSD license.

I think it would also be enough to obtain a permission from the authors
to call Debian modified versions "Alice", as long as renaming it is easy
otherwise. We have allowed such things in the past. 

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GPL photographies, eg for backround

[Josselin Mouette]

Le mercredi 31 décembre 2008 à 09:15 -0800, Ken Arromdee a écrit :
> > Indeed, but we are not talking of a program but of pictures here.
> 
> The same applies if you don't provide the source code for the picture.

No. If you’re the copyright owner, you get to decide what is the
preferred form of modification.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GPL photographies, eg for backround

[Josselin Mouette]

Le mercredi 31 décembre 2008 à 15:50 +0100, Francesco Poli a écrit :
> > Indeed, but we are not talking of a program but of pictures here.
> 
> I am convinced that this distinction is (almost) irrelevant from the
> GPL point of view.

The relevance comes from the fact that pictures can be their own source
code, regardless of the format.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GPL photographies, eg for backround

[Josselin Mouette]

Le lundi 29 décembre 2008 à 10:44 -0800, Ken Arromdee a écrit :
> 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

Indeed, but we are not talking of a program but of pictures here.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GPL photographies, eg for backround

[Josselin Mouette]

Le lundi 29 décembre 2008 à 13:52 +0100, Marco d'Itri a écrit :
> > * To upload a "background source package", is it mandatory to use 
> >   an uncompressed format, such as tiff, for photographies, or a 
> E.g. this is bullshit.

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.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: deluge and GeoIP database license

[Josselin Mouette]

Le vendredi 21 novembre 2008 à 00:33 +0100, Cristian Greco a écrit :
> 1) Is this license really suitable for distribution in main now?

Yes, it’s a BSD license with advertising clause. Theoretically you
cannot link GPL software without exception to it; however the library is
merely using it as data and is not strictly depending on the database,
so my guess would be that this is OK.

> 2) If so, what about ktorrent? Should they update their copy of the
> database in order to avoid source repackaging?

If possible, ktorrent and libgeoip should use the same database to avoid
duplication.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Public Domain for Germans

[Josselin Mouette]

Le samedi 15 novembre 2008 à 14:25 +, MJ Ray a écrit :
> Also, I'm disappointed that WTFPLv2 is so long.  Why do people need to
> care about Sam Hocevar's name, address and permission to change it?
> It seems obviously below the creative threshold for copyright...
> 
> I'm no longer sure whether this is a joke.

What made you think this was a joke to begin with?

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GNU Free Documentation License v1.3

[Josselin Mouette]

Le lundi 03 novembre 2008 à 18:28 +0100, Simon Josefsson a écrit :
> 2. VERBATIM COPYING

> You may not use
> technical measures to obstruct or control the reading or further
> copying of the copies you make or distribute.

I wonder how we should consider the fact they did not remove nor
rephrase this obnoxious clause. Back in the FDL discussions, it was
commonly accepted that this was a honest mistake and that it was going
to be fixed in the next version.

Now the next version is here, and it hasn’t been fixed. How many years
will it take before we can chmod -r our directories and close our
drawers?

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: EllisLab, Inc. CodeIgniter license

[Josselin Mouette]

Le samedi 01 novembre 2008 à 17:59 +0100, Francesco Poli a écrit :
> First of all, a no-name-including-this-string clause may well be more
> restrictive than default trademark law.  A case where this is
> particularly apparent is the PHP one: I am not allowed to use any name
> including the string "PHP" for a derivative work of PHP, not even
> something like RALPHPANTHER or TELEGRAPHPOLE!
> Those example names are not confusingly similar to PHP, as far as I can
> tell, and trademark laws usually insist on avoiding confusion, rather
> than on substrings...
> On the other hand, it should also be noted that a clause in a copyright
> license cannot prevent me from writing a (non-derivative) work from
> scratch and calling it "PHP++": so maybe such clauses do not even fully
> reach their goal...

This is annoying for PHP, which is a short name, but frankly I fail to
see how this could be a problem for CodeIgniter.

> Secondly, one thing is having something forbidden by trademark law, one
> different thing is adding copyright violation to the picture.
> If trademark law already forbids something, why should I prevent it
> through copyright law as well?

I’m not questioning the uselessness of this clause. Nevertheless, its
practical consequences are limited enough so that we can reasonably say
it doesn’t violate the DFSG.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: EllisLab, Inc. CodeIgniter license

[Josselin Mouette]

Le mercredi 29 octobre 2008 à 23:34 +0100, Francesco Poli a écrit :
> > 4. Any files that have been modified must carry notices stating the nature
> >of the change and the names of those who changed them.
> 
> An obligation to maintain a sort of change-log is normally acceptable.
> The second part of the clause is more troublesome: depending on how it
> should be interpreted, it could meet the DFSG or fail to do so.
> If I am allowed to modify a file, document the nature of the change and
> put a nickname (or a pseudonym, or even the term "anonymous") as the
> name of the modifier, then I think this clause complies with the DFSG.
> 
> On the other hand, if I am compelled to disclose my own identity in
> order to distribute a file modified by me, then I think that this
> clause fails DFSG#1, since being forced to disclose one's own identity
> can be a fee.
> See also  http://lists.debian.org/debian-legal/2007/05/msg00015.html
> for a more detail explanation of the issue (found in another license).

This may depend on local law, but I don’t think this prevents you from
using a pseudonym as "name". A name is different from an identity, so I
don’t think this clause causes any trouble.

> > 6. Products derived from the Software may not be called "CodeIgniter",
> 
> This is considered acceptable (as a compromise!) per DFSG#4.
> 
> >nor may "CodeIgniter" appear in their name, without prior written
> >permission from EllisLab, Inc.
> 
> IMHO, this goes beyond what is permitted (as a compromise!) by DFSG#4,
> since it forbids an infinite set of names, rather than a single one.
> I cannot use any of the following names for a derived product:
> CodeIgniterNG, CodeIgniter++, SuperCodeIgniter, TinyCodeIgniter, ...

In all cases you will not be allowed to use such names unless you obtain
a trademark exception, so I don’t think this is problematic either.

> > INDEMNITY
> > You agree to indemnify and hold harmless the authors of the Software and
> > any contributors for any direct, indirect, incidental, or consequential
> > third-party claims, actions or suits, as well as any related expenses,
> > liabilities, damages, settlements or fees arising from your use or misuse
> > of the Software, or a violation of any terms of this license.
> 
> Warning: indemnification clause: is it acceptable?
> It smells as non-free, but I would like to know the opinion of other
> debian-legal regulars...

At least this is already accepted in main, see e.g. postfix.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: 25+2 packages with (Glade) generated C source files without the source

[Josselin Mouette]

Le dimanche 31 août 2008 à 04:17 +0300, Sami Liedes a écrit :
> I went through some of these and checked them by hand, and generally
> couldn't find the glade project anywhere in the source tarball (it
> might be in the diff, I didn't check for that - would that BTW be OK,
> to have source code in diff only?). The only questionable case I found
> by this sampling is dia, where the file is "generated by Glade and
> then hand-coded to make GNOME optional and add the underline for
> accelerated buttons".

I’m pretty sure many of the list are in similar cases. Now loading the
UI directly into the application is the standard, but not so long ago
people generated template code with glade and then edited it by hand.
The .glade file was removed simply because it has become irrelevant.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Is AGPLv3 DFSG-free?

[Josselin Mouette]

Le jeudi 21 août 2008 à 11:16 +0200, Miriam Ruiz a écrit :
> Yes, I'm saying that it might be failing it. If you use a program
> (possibly modified) covered by AGPLv3 that uses some kind of network,
> and you cannot convey its source code to the remote people you're
> interacting with through that network itself, possibly due to
> technical reasons (like, for example, bandwidth or connection time
> limits), you must make it available to them by some means, all of
> which I can think of mean in some way or another identifying yourself
> (yes, even storing the source code in a public server probably means
> that).

Whoa? I really don’t get you here.

The “problem” with the AGPLv3 is that you can argue the distribution
requirement is onerous. It may be a bit more onerous for a dissident,
but frankly we are talking about ridiculous costs here (and there are
several technical ways to distribute sources at large without being
identified, so in reality your dissident is OK).

As often stated on this list, you cannot always state the DFSG-freeness
of a license, you need to apply the rules to a *work*. If the sources
are several gigabytes large, then the price of distributing them becomes
unacceptable and the work may not be DFSG-free.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Is AGPLv3 DFSG-free?

[Josselin Mouette]

Le jeudi 28 août 2008 à 01:22 +0200, Marco d'Itri a écrit :
> Because it's just not implied, unless you manage to stretch the concepts
> enough.

Maybe we need to rephrase DFSG5 as “The license must not discriminate
against any person or group of persons, except those Marco d’Itri
doesn’t care about.”

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Bootstrapping from binary blob shipped in the source package

[Josselin Mouette]

Le jeudi 31 juillet 2008 à 10:07 +0200, Jens Peter Secher a écrit :
> What do you mean by reliability here?  From a package-building
> perspective, I find it to be more reliable to use the precompiled blop
> because then I can make sure that the package compiles in a minimal
> environment.  Otherwise I need to depend on a *specific* prior version
> of the compiler, thus making the build more fragile (think Ubuntu), or
> I need to depend on a wide range on prior versions of the compiler,
> making testing much harder.

If your code needs a specific version of the compiler, you have a
serious issue here anyway. For example if a bug in the compiler is found
that requires the package to be rebuilt, it will remain broken if it
cannot simply use the latest version.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Zend Optimizer License

[Josselin Mouette]

Le jeudi 10 juillet 2008 à 15:06 -0500, Richard Laager a écrit :
> 7. Indemnity. Licensee will, at its own expense, defend any action
> brought by a third party against Zend to the extent that such action is
> based on a claim arising from or relating to: (a) Licensee's use of the
> Software, (ii) any distribution of the Software by Licensee or by any of
> or by any sub-licensee, regardless of privity of contract and regardless
> of the length of the trail of sublicenses, (iii) any claims based upon
> warranties, guarantees or representations made by Licensee or any of its
> employees, agents or sub-licensees; or (iv) any use of the Software by
> any of the foregoing sub-licensees. Zend shall have the exclusive right
> to control such defense. In no event shall Licensee settle any such
> claim, lawsuit or proceeding without Zend's prior written approval.

This is a lawyerbomb. By distributing the software, Debian and its
mirror network could be held liable for anything that people receiving
the software would be doing. I really don’t think we should distribute
such a thing.

> 10. Miscellaneous. This Agreement is made in and shall be governed by
> the laws of the State of Israel, excluding choice of law principles.
> Venue for all proceedings shall be Tel Aviv, Israel. Notwithstanding the
> foregoing, Zend shall have the right to apply to any court of competent
> jurisdiction for injunctive or other relief. The United Nations
> Convention for the International Sale of Goods shall not apply.

I’m not sure of the implications of this claim, but it is very
suspicious; Israel having ratified the Vienna convention, I don’t think
they can just say it will not apply. Of course that would make moot all
that comes before, since choice of law does not apply in this case.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: codecs and totem

[Josselin Mouette]

Le dimanche 29 juin 2008 à 01:53 -0400, Daniel Dickinson a écrit :
> totem-xine
> vlc
> (I think mplayer)
> 
> can play DVD's as they are in main

totem-gstreamer can also, but without full menu support.

Of course, you need libdvdcss to be able to read 99% of the DVDs.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: BSD/GPL/LGPL and OpenSSL

[Josselin Mouette]

Le jeudi 05 juin 2008 à 18:02 +0200, Vincent Danjean a écrit :
> * LGPL+ssl (LGPL with OpenSSL clause)

There is no need for an OpenSSL exception for a LGPL-licensed work.

> What I'm thinking with a program that links with 2 libraries:
> NOT valid: progA[GPL]{libssl}
> valid: progA[GPL+ssl]{libssl}
> valid: progA[GPL+ssl]{libssl,libB[GPL]}

This is not valid, because you indirectly link libB with libssl. You
need libB[GPL+ssl] for this case.

> valid: progA[GPL+ssl]{libssl,libB[LGPL]}
> valid: progA[GPL+ssl]{libssl,libB[BSD]}
> NOT valid: progA[BSD]{libssl,libB[GPL]}
> valid: progA[BSD]{libssl,libB[LGPL]}
> valid: progA[BSD]{libssl,libB[BSD]}
> 
> And now, more complex cases where this is a library that links to openssl
> 
> valid: progA[BSD]{libB[BSD]{libssl}}
> NOT valid: progA[GPL]{libB[LGPL+ssl]{libssl}}
> valid: progA[GPL+ssl]{libB[LGPL+ssl]{libssl},libC[GPL]}

This one is not valid, for the same reason.

> valid: progA[GPL+ssl]{libB[BSD]{libssl},libC[GPL]}

Ditto.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: DFSG is not suitable for things other than software

[Josselin Mouette]

Le mercredi 21 mai 2008 à 16:12 -0700, Yuhong Bao a écrit :
> DFSG is not suitable for, and should not be applied to, things other than 
> software.

I see dead horses. People beat them. They don’t know they are dead. They
only see what they want to see.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


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

Re: RFR: Inc. logo in game data

[Josselin Mouette]

Le samedi 03 mai 2008 à 17:43 +0800, Wen-Yen Chuang a écrit :
> He thinks that those games should to be put in non-free, unless we can
> remove those logos. [2]
> 
> It is OK for me to pack a non-free package, but it is hard to find a
> sponsor for non-free packages.

The license doesn’t seem to prevent removal of the logos.
4) You may modify the game as you wish.  You may also distribute 
modified
 versions under the terms set forth in this license, but with the 
additional
 requirement that the work is marked with a prominent notice which 
states that
 it is a modified version.

You shouldn’t have any trouble removing them if you add such a notice as
replacement.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Falcon P.L. license (ITP:Bug#460591)

[Josselin Mouette]

On jeu, 2008-03-27 at 18:58 -0700, Sean Kellogg wrote:
> > No one can patent the grammar that you wrote, so this is completely
> > useless. The only point of these clauses seem to claim the copyright on
> > scripts using the language.
> 
> Huh? Why can't someone patent langauge grammar/syntax?

I should have written “no one *else* can pattent the grammar that you
wrote”.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Falcon P.L. license (ITP:Bug#460591)

[Josselin Mouette]

On sam, 2008-03-22 at 22:33 +0100, Giancarlo Niccolai wrote:
> > Have you considered the GNU LGPL (v2.1)?
> >
> Yes, but I encountered strong resistance from FSF when proposing a
> "lighter" (with exceptions) LGPL version.

This is, again, because you are not proposing additional permissions
(for which the LGPL with extra additional permissions would be fine) but
additional *restrictions*.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Falcon P.L. license (ITP:Bug#460591)

[Josselin Mouette]

On ven, 2008-03-21 at 10:09 +0100, Giancarlo Niccolai wrote:
> > This clause makes the license a copyleft one. It is free, but this is a
> > huge restriction compared to the original license. And this turns the
> > license into yet another copyleft license that will be incompatible with
> > other ones.
> >
> I see; you are talking of comma 4.5 in the specific?
> 
> I actually added it after some talk with FSF about "requiring the
> derivative work to be distributed copyleft too". In other words, the
> aim of this comma is to prevent someone from adding a "frobotz"
> statement which nitfols the blorbs, and then distribute an xFalcon as
> a closed source product.

Sure, this is the whole point of copyleft and it is a very good thing.
But copyleft has its drawbacks, especially when it comes to license
compatibility. This is why you should more seriously consider using an
existing copyleft license. License proliferation is not a good thing;
copyleft license proliferation is destructive.

> >>5. *Distribution of Embedding Works and Scripts*.
> >
> > I don’t think you can claim anything on the copyrights of scripts using
> > your language, but this is definitely something you should ask your
> > lawyer.
> This license doesn't state there are copyright claims on the scripts;
> actually it should do the opposite. If there is any point implicitly
> or explicitly stating it, please point it out and it will be fixed.

It does claim the copyright by asking to apply the license to it.

> > If you really want to explicitly tell that you don’t need to follow the
> > license for writing scripts, you should add a notice that this license
> > and your copyright claims don’t apply to the said scripts, and this will
> > be much better.
> The point is that, as previously noted, the patentability of grammar
> sets (i.e. artificial languages) has been recently debated. Including
> the definition of the scripts in this license has the aim to prevent a
> Big Guy to come in, add a frobotz statement and patent the resulting
> language (or, as someone has pointed out, just patent the grammar
> someone else wrote as-is). Or in other words, I did it to maintain
> freedom of the grammar set this language define (it means, freedom for
> everyone to use and extend it).

No one can patent the grammar that you wrote, so this is completely
useless. The only point of these clauses seem to claim the copyright on
scripts using the language.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Falcon P.L. license (ITP:Bug#460591)

[Josselin Mouette]

On mer, 2008-03-19 at 20:34 +0100, Giancarlo Niccolai wrote:
> The license is tightly based on Apache 2, with extra clarifications
> and permissions. 

This is, well, an interesting claim.

>4. *Redistribution of Work and Derivative Works*. You may reproduce
>   and distribute copies of the Work or Derivative Works thereof in
>   any medium, with or without modifications, and in Source or
>   Object form, provided that You meet the following conditions:
>  5. The Derivative Works are distributed under the terms of
> this License, or under terms that do not cause
> infringement of this License.

This clause makes the license a copyleft one. It is free, but this is a
huge restriction compared to the original license. And this turns the
license into yet another copyleft license that will be incompatible with
other ones.

>5. *Distribution of Embedding Works and Scripts*.

I don’t think you can claim anything on the copyrights of scripts using
your language, but this is definitely something you should ask your
lawyer.

If you really want to explicitly tell that you don’t need to follow the
license for writing scripts, you should add a notice that this license
and your copyright claims don’t apply to the said scripts, and this will
be much better.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Desert island test (was: Questions about liblouis)

[Josselin Mouette]

Le mercredi 27 février 2008 à 18:13 -0800, Sean Kellogg a écrit :
> And not grounded in the specific language of the DFSG but rather a shared 
> aspiration of what the document "ought" to say. I have never seen an attempt 
> to tie the three tests to specific points and thus it is impossible to debate 
> and discuss the test themselves... it has become assumed knowledge.

The tests are not meant to be extra guidelines, nor reformulations
specific guidelines, nor assumed knowledge. As the name says, they are
tests: extreme situations in which you can test how the DFSG (all of
them) apply to a piece of software in a specific case.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: License clause, not allowing use of author's name for promoting

[Josselin Mouette]

Le mercredi 23 janvier 2008 à 23:25 +0100, Jose Carlos Garcia Sogo a
écrit :
> Hi, 
> 
> I am concerned about the following license snippet, basically point 3
> which talk about use of authors' name.
> This comes from a source included in conduit (version 0.3.6, not yet
> uploaded)

> 3. The name of the author may not be used to endorse or promote products 
>derived from this software without specific prior written permission.

This is the standard 3rd clause of the BSD license, and there are
probably hundreds of packages with such a clause in the archive.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: TrueCrypt License 2.3

[Josselin Mouette]

Le mercredi 16 janvier 2008 à 14:55 +, MJ Ray a écrit :
> > We have allowed exactly the same conditions by using software with
> > trademarked names.
> 
> Where?  The naming rights asserted above seem much broader than what a
> trademark allows.  Trademarks have many limitations.

If we have named Firefux the modified version of Firefox, I doubt the
Mozilla foundation would have let that pass.

> > In fact, upstream is wrong for putting such
> > restrictions in the license itself instead of the trademark policy, but
> > the net effect is exactly the same as that of the Firefox trademark.
> 
> Didn't we have to change the name to avoid the Firefox trademark+copyright
> combo-knockout?

Indeed, but not all upstreams have such stupid trademark licensing
schemes. See Apache for a good example.

> > > > 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. [...]
> > > This is non-free, as explained by Ken Arromdee in
> > > http://lists.debian.org/debian-legal/2008/01/msg00132.html
> > 
> > Please get out of your US-centric world.
> 
> Does it matter whether it is non-free only for France or only for the
> US?  Doesn't that mere difference make it fail DFSG 5?

I don’t think so. The fact that it doesn’t grant some of the rights that
are usually applicable in the US but not somewhere else doesn’t make the
license in itself non-free. It has been argued several times that DFSG#5
is here to make the software free for everyone. If it is more free for
some groups of persons than for others, good for them. For a similar
example, the GPL with an additional permission for snowboarders to
integrate the software in their proprietary developments would not fail
the test.

> I've not studied the liability debate here, but none of the above responses
> seem to have merit.  They're also unnecessarily you-you-you.  Did Francesco
> Poli run over Josselin Mouette's cat?

Huh? Who’s trying to make things personal?

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: TrueCrypt License 2.3

[Josselin Mouette]

Le lundi 14 janvier 2008 à 22:50 +0100, Francesco Poli a écrit :
> > a. The name of Your Product (or of Your modified version of This 
> > Product)
> > must not contain the name TrueCrypt (for example, the following names 
> > are
> > not allowed: TrueCrypt, TrueCrypt+, TrueCrypt Professional, iTrueCrypt,
> > etc.) nor any of its variations that can be easily confused with the 
> > name
> > TrueCrypt (e.g., True-Crypt, True Crypt, TrueKrypt, TruCrypt, etc.)
> 
> I've argued several times in the past against this kind of broad
> restrictions.  I think they go beyond what is permitted (as a
> compromise!) by DFSG#4.

We have allowed exactly the same conditions by using software with
trademarked names. In fact, upstream is wrong for putting such
restrictions in the license itself instead of the trademark policy, but
the net effect is exactly the same as that of the Firefox trademark.

> See, for instance:
> http://lists.debian.org/debian-legal/2007/11/msg4.html
> http://lists.debian.org/debian-legal/2006/04/msg00181.html

Using yourself as a reference?

> Warning!  Indemnification clause: is it acceptable?  It smells as
> non-free...

Just have a look at the postfix license.

> [...]
> > 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 is non-free, as explained by Ken Arromdee in
> http://lists.debian.org/debian-legal/2008/01/msg00132.html

Please get out of your US-centric world.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Micropolis GPL License Notice

[Josselin Mouette]

On dim, 2008-01-13 at 14:37 +0100, Martin Zobel-Helas wrote:
> | ADDITIONAL TERMS per GNU GPL Section 7
> | 
> | No trademark or publicity rights are granted. This license does NOT give you
> | any right, title or interest in the trademark SimCity or any other 
> Electronic
> | Arts trademark. You may not distribute any modification of this program 
> using
> | the trademark SimCity or claim any affliation or association with Electronic
> | Arts Inc. or its employees.

Sounds fine, but we need to check whether the artwork itself is
DFSG-free. 

> | Any propagation or conveyance of this program must include this copyright
> | notice and these terms.

Already required by the GPL.

> | If you convey this program (or any modifications of it) and assume 
> contractual
> | liability for the program to recipients of it, you agree to indemnify
> | Electronic Arts for any liability that those contractual assumptions impose 
> on
> | Electronic Arts.

These indemnification clauses seem to widespread; currently they are
accepted in the archive.

> | You may not misrepresent the origins of this program; modified versions of 
> the
> | program must be marked as such and not identified as the original program.

Already required by the GPL.

> | This disclaimer supplements the one included in the General Public License.

Blah.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: TrueCrypt License 2.3

[Josselin Mouette]

On sam, 2008-01-12 at 20:27 +0100, Francesco Poli wrote:
> The plain text version of the licence may be found at
> http://www.truecrypt.org/docs/License.txt
> and is pasted below in its entirety.

Thanks. 

Summary:
  * I think this software is fine for main if we package it the
iceweasel way (new name, new artwork).
  * The advertising clause is very obnoxious but still acceptable. 
  * I also have a wonder about the TrueCrypt license, but it can be
easily clarified (at worst by asking Bruce Schneier directly).

Comments about the different licenses follow.

> TrueCrypt License Version 2.3

This looks like a generic copyleft license. Specific clauses follow.

> a. The name of Your Product (or of Your modified version of This Product)
> must not contain the name TrueCrypt (for example, the following names are
> not allowed: TrueCrypt, TrueCrypt+, TrueCrypt Professional, iTrueCrypt,
> etc.) nor any of its variations that can be easily confused with the name
> TrueCrypt (e.g., True-Crypt, True Crypt, TrueKrypt, TruCrypt, etc.)

Name change clause, it is fine.

> All graphics files showing any TrueCrypt logo (including the non-textual
> logo consisting primarily of a key in stylized form) must be removed from
> Your Product (or from Your modified version of This Product) and from any
> associated materials. Logo(s) included in (or attached to) Your Product
> (or in/to associated materials) must not incorporate and must not be
> confusingly similar to any of the TrueCrypt logos or portion(s) thereof.

Which means we need to remove them from the package as well.

> c. Phrase "Based on TrueCrypt, freely available at
> http://www.truecrypt.org/"; must be displayed by Your Product (if
> technically feasible) and contained in its documentation. Alternatively, 
> if
> This Product or its portion You included in Your Product comprises only a
> minor portion of Your Product, phrase "Portions of this product are based
> in part on TrueCrypt, freely available at http://www.truecrypt.org/"; may 
> be
> displayed instead. In each of the cases mentioned above in this paragraph,
> "http://www.truecrypt.org/"; must be a hyperlink (if technically feasible)
> pointing to http://www.truecrypt.org/ and you may freely choose the
> location within the user interface (if there is any) of Your Product 
> (e.g.,
> an "About" window, etc.) and the way in which Your Product will display 
> the
> respective phrase.

This is the most questionable clause. It looks much like a mix between
the OpenSSL advertising clause and the GPL warranty disclaimer. I don’t
like the clause, but I don’t feel it breaks any of the DFSG, especially
because it is reasonable about the requirements (“if technically
feasible”, you’re free to choose how to put it…)

> d. The complete source code of Your Product must be freely and publicly
> available (for exceptions, see Sections III.2 and III.3) at least until 
> you
> cease to distribute Your Product. To meet this condition, it is sufficient
> that You merely include the source code with every copy of Your Product
> that you make and distribute; it is also sufficient that You merely 
> include
> information (valid and correct at least until you cease to distribute Your
> Product) about where the source code can be freely obtained (e.g., an
> Internet address, etc.) with every copy of Your Product that you make and
> distribute. 

This is fine and passes the desert island and dissident tests.

> 2. You are not obligated to comply with Sections III.1.a, III.1.b, III.1.c, 
> and
> III.1.d, if all conditions specified in one of the two following paragraphs 
> are
> met:
> 
> a. Your Product is an operating system distribution, or other aggregate
> software distribution (such as a cover CD-ROM of a magazine) containing
> products from different sources, in which You include either This Product
> without any modifications or file(s) which You obtain by compiling the
> unmodified source code of This Product.

This is a nice clause to allow distributors to keep the name;
unfortunately it requires keeping the non-free logos, so this looks like
an iceweasel case.

> 4. You shall indemnify, defend and hold all (co)authors of This Product, their
> agents and associates, and applicable copyright/trademark owners, harmless
> from/against any liability, loss, expense, damages, claims or causes of 
> action,
> arising out of Your use, inability to use, reproduction, (re)distribution,
> import and/or (re)export of This Product (or portions thereof) and/or Your
> breach of any term of this License.

Indemnification clause, similar to the IBM public license.

> 
> 
> License agreement for Encryption for the Masses.

Simple non-copyleft license with name change clause and advertising
clause.

> This

Re: Choosing a License: GNU APL? AFL 3.0?

[Josselin Mouette]

Le dimanche 30 décembre 2007 à 08:55 +, Sean B. Palmer a écrit :
> I'm looking for a permissive license, of the Modified BSD or MIT
> variety, but I'd like for the copyright notices in each file to be
> protected without having to include the whole license itself, if it's
> more than a few lines.
> 
> The Modified BSD, for example, only protects "the above copyright
> notice, this list of conditions and the following disclaimer", so a
> simple notice that says "Licensed under the Modified BSD License"
> wouldn't be protected unless I included the whole license in each
> file. Therefore it's not sufficient.

I think the WTFPL is exactly what you are looking for. It meets all your
requirements, is DFSG-free, and several Debian packages are already
using it.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Bug#451799: new evince cannot display Japanese characters correctly

[Josselin Mouette]

clone 451799 -1
retitle 451799 evince should depend on poppler-data
reassign -1 wnpp
retitle -1 RFP: poppler-data -- Encoding data for the poppler PDF rendering 
library
block 451799 by -1
thanks

* Package name: poppler-data
  Version : 0.1.1
  Upstream Author : Adobe, Red Hat
* URL : http://poppler.freedesktop.org/
* License : non-free, see below
  Programming Lang: None
  Description : Encoding data for the poppler PDF rendering library

This package contains the encoding data needed to view some PDF
documents with libpoppler.

The copyright is the following:
Copyright 1990-1998 Adobe Systems Incorporated.
All Rights Reserved.

Patents Pending

NOTICE: All information contained herein is the property of Adobe
Systems Incorporated.

Permission is granted for redistribution of this file provided
this copyright notice is maintained intact and that the contents
of this file are not altered in any way from its original form.

PostScript and Display PostScript are trademarks of Adobe Systems
Incorporated which may be registered in certain jurisdictions.

However I don't think there is anything copyrightable in these files;
they only contain series of numbers that describe the mappings. Do you
people think it could be suitable for main?
(Please follow-up on -legal only for licensing discussions.)

Ondrej, are you willing - if the legal problems are settled out - to
package it? Otherwise I guess me or any of the co-maintainers could do
it, the packaging is absolutely trivial.


Le lundi 19 novembre 2007 à 02:25 +0900, Hideki Yamane a écrit :
>  New evince package displays Japanese characters wrong. Broken.
>  Please see attached image file, one is opened CP-06.pdf with evince 
>  (Screenshot-CP-06.pdf.png), and another is with Adobe Reader (shows 
>  fine). I heard same problem from other people on IRC.

The new poppler version needs some specific files that contain some
mappings between Unicode and other encodings, which are in a separate
package.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GPLv3 compatible with OpenSSL?

[Josselin Mouette]

Le mardi 20 novembre 2007 à 12:10 +0100, Matej Vela a écrit :
> Is GPLv3 compatible with the OpenSSL license?

I don't think so.

> 3. All advertising materials mentioning features or use of this
>software must display the following acknowledgment:
>"This product includes software developed by the OpenSSL Project
>for use in the OpenSSL Toolkit. (http://www.openssl.org/)"

This is the actual clause that is GPL-incompatible.

> 6. Redistributions of any form whatsoever must retain the following
>acknowledgment:
>"This product includes software developed by the OpenSSL Project
>for use in the OpenSSL Toolkit (http://www.openssl.org/)"

I don't think this clause is even incompatible with the GPLv2, although
this is debatable.

> might be permitted by GPLv3 section 7:
> 
>   Notwithstanding any other provision of this License, for material you
> add to a covered work, you may (if authorized by the copyright holders of
> that material) supplement the terms of this License with terms:
> 
> b) Requiring preservation of specified reasonable legal notices or
> author attributions in that material or in the Appropriate Legal
> Notices displayed by works containing it; or
> 
> d) Limiting the use for publicity purposes of names of licensors or
> authors of the material; or
> 
> e) Declining to grant rights under trademark law for use of some
> trade names, trademarks, or service marks; or

I'm afraid none of these exceptions apply for advertising material. The
FSF has always considered such clauses to be BAD(tm) (and I can't say I
find them very nice), it's no surprise the GPLv3 doesn't allow them.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of wodim

[Josselin Mouette]

Le mercredi 14 novembre 2007 à 08:30 -0800, Steve Langasek a écrit :
> For all I know he does have a legitimate claim under German law that cdrkit
> infringes his Urheberrecht, but cdrkit is not a German product per se.

The German law doesn't give Jörg Schilling more rights than any other
one, and I don't think there are some that could hurt cdrkit.

The author's rights differ from the copyright mostly by the existence of
moral rights, which are inalienable, imprescriptible and cannot be
transferred under any circumstances. (Note that I'm basing my reasoning
on French law, but TTBOMK there are very few differences with German IP
law.)

These moral rights are:
  * The respect of the name of the author and his quality. Note that
the GPL also explicitly requires this for countries without such
legislation.
  * The right to disclose - or not - the work.
  * The right to withdraw the work, if you can indemnify all
distributors for the financial loss. In France, this right is
not applicable for software - except when the diffusion can
prejudice the author's honor or reputation.

The claims of Mr Schilling are exactly about this prejudice made to his
honor or reputation. Not only would he have a *very* hard time proving
this, as cdrkit clearly indicates it is a modified version, but it isn't
even clear he can do so without indemnifying all distributors (and maybe
users) for the loss.

Given the current number of cdrkit users, I think we could finally
obtain the required financial support to bring libburn to a good
state :)

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Java, GPL and CDDL

[Josselin Mouette]

Le vendredi 16 novembre 2007 à 16:23 +0100, Joerg Schilling a écrit :
> If you talk to lawyers and ask them about the GPL, they will tell you that
> the GPL is a contract offer that needs to be explicitely acepted by the 
> licensee.

This is of course completely wrong. Unless you accept the terms of the
GPL, the author's rights apply by default, so you don't have the right
to use, distribute or modify the software.

-- 
 .''`.   Josselin Mouette/\./\
: :' :   [EMAIL PROTECTED]
`. `'[EMAIL PROTECTED]
  `-  Debian GNU/Linux -- The power of freedom


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Java, GPL and CDDL

[Josselin Mouette]

Le jeudi 15 novembre 2007 à 17:04 +0100, Joerg Schilling a écrit :
> >CarMetal uses colorchooser https://colorchooser.dev.java.net/ wich is
> >CDDL licensed.
> 
> If colorchooser has been developed independently from CaRMetal, and only
> CaRMetal calls colorchooser, it is indeed similar to what happens with mkisofs
> and the license mix you describe is legal too.

Please don't take Jörg Schilling's dreams for reality.

This case is different from cdrtools, and easier to solve. You only need
to get a specific exception from the authors of CarMetal to allow its
linking to colorchooser; much like the OpenSSL exception that is present
in many GPL applications.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: "rescuing" code from the GPL

[Josselin Mouette]

Le dimanche 11 novembre 2007 à 21:25 +0530, Shriramana Sharma a écrit :
> 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.

No, it must be licensed under a GPL-*compatible* license. And the BSD
license is GPL-compatible.

Cheers,
-- 
 .''`.   Josselin Mouette/\./\
: :' :   [EMAIL PROTECTED]
`. `'[EMAIL PROTECTED]
  `-  Debian GNU/Linux -- The power of freedom


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of wodim

[Josselin Mouette]

Le samedi 10 novembre 2007 à 16:39 +0100, Joerg Schilling a écrit :
> So make sure that "wodim" prints something like:
> 
> "This program is known to have bugs that are not present in the original 
> software"
> 
> and it mets the rules.

Sorry, but we are not allowed to display false statements like this one.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of wodim

[Josselin Mouette]

Le samedi 10 novembre 2007 à 12:57 +0100, Joerg Schilling a écrit :
> Josselin Mouette <[EMAIL PROTECTED]> wrote:
> > There is nothing like that in the GPL. It only forbids misrepresentation
> > of the Author's work.
> 
> You seem to missinterpret the GPL.

> If the software is modified by someone else and passed on, we 
> want its recipients to know that what they have is not the original, so 
> that any problems introduced by others will not reflect on the original 
> authors' reputations. 

Which makes indeed clear that correctly representing the different
authors' contributions will avoid affecting their reputations.

> A common misconception is to disregard that the GPL does _not_ void the
> Urheberrecht (http://bundesrecht.juris.de/urhg/index.html). Anything that 
> is permitten in the GPL is permitten only in a Urheberrecht context.
> Urheberrecht allows the Author to forbid publishing of a work if this is made
> in a way that harms the reputation of the author.

Indeed. And as long as the modifications are correctly attributed to
their respective authors, there is no way they could harm the reputation
of another author. So we're all good and people distributing cdrkit
aren't violating any law.

You seem to believe that your reputation is affected by the very
existence of a fork, which would imply that your work isn't perfect. But
this is a problem with your ego, not with the law.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le samedi 10 novembre 2007 à 12:51 +0100, Joerg Schilling a écrit :
> A GPL "work" that uses a CDDL library _may_ be a derived work from the CDDL 
> library. The CDDL library is definitely not a derived work of it's uers.

Of course. But the *combined work* that is constituted by the CDDL
library and the GPL program is derived from *both*. And you cannot
distribute it because it has incompatible requirements.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le samedi 10 novembre 2007 à 12:35 +0100, Joerg Schilling a écrit :
> > > The GPL explicitely allows to use code under other licenses from GPL code.
> >
> > No, it does not. If you think it does, please point the line where it
> > "explicitly" allows it.
> 
> Well, _I_ did already explain why this is the case. You did not give any 
> evidence for your claim.
> 
> GPL §2b) verifies that the GPL is asymmetric.

The fact that you don't understand what "derived work" means is
irrelevant to this discussion.

> http://www.usfca.edu/law/determann/softwarecombinations060403.pdf
> verifies that GPL §2b) is asymmetric.

After reading such basic, giant misunderstandings like:
The GPL does not prohibit licensees from charging money for
selling software copies. However, the GPL itself already
provides everyone with a license free of charge. As a result,
licensees have no realistic chances to commercialize their
copyrights in works that they “derive” from GPLed code.
I will not read any further this document, sorry. You will have to back
up your claims with people who know what they are talking about.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: GPL 3 and derivatives

[Josselin Mouette]

Le samedi 10 novembre 2007 à 06:15 +0530, Shriramana Sharma a écrit :
> Francesco Poli wrote:
> > ... and despite its length, it does not even implement an actually
> > working copyleft mechanism.  :-(
> 
> Francesco, that's very surprising. Can you please elaborate, or have you 
> posted your opinion on this already in previous threads?

It was already posted, and it was also explained at large why this is
not the case.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le vendredi 09 novembre 2007 à 21:28 +0100, Joerg Schilling a écrit :
> there is a problem in wodim.
> 
> The GPL and the Urheberrecht both forbid to publish modified versions that
> harm the reputation of the Author. 

There is nothing like that in the GPL. It only forbids misrepresentation
of the Author's work.

> The Debian fork is full of extreme bugs
> and many people are thus completely unable to use the fork at all.
> You should be aware of the possibility that I (as the Author) could disallow 
> publikshing the fork.

Well, to do that, you would have to prove:
  * either that wodim is violating the license - which is not, it is
based on entirely GPL code and follows the GPL to the letter;
  * or that the wodim authors are violating the law - if they do, it
is surely not by misrepresenting your work.

The whole point of free software is to allow forks. If you can't accept
that, you'd better find another hobby.

Anyway, I would *love* to see you do attempting to forbid distribution
of cdrkit. If you lose, we will all have a good laugh. If you win, the
community would have to work on a new, free alternative that would force
us to finally get rid of your crappy code. In all cases, that's a big
win.

> >Debian change their opinion. So would it be so difficult for you to 
> >change the license in other to satisfy Debian (and other Linux 
> >distributor)? I think it is the only way you can hope they reconsider 
> >the inclusion of cdrtool.
> 
> This kind of blackmail attempts have been send to me more than once.

You must deeply want to see cdrtools back in Debian if you think of such
requests as "blackmail"…

> Cdrtools uses a license that is widely accepted as true OpenSource and
> even the official Debian statement is that the CDDL is obviously free.

The Debian project also allows OpenSSL code and GPL code, but doesn't
accept mixing both without correct exceptions from the copyright
holders.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le jeudi 08 novembre 2007 à 11:46 -0500, Steve Langasek a écrit :
> If you want my services as an English teacher, you'll
> have to ask me for a quote; otherwise, finding the errors in your logic is
> your problem, not mine.

Le jeudi 08 novembre 2007 à 17:55 +0100, Joerg Schilling a écrit :
> Is this because you need an English teacher or because you have problems 
> understaning  complex sentences? Anyway, ask someone in your vicinty for help!

A, it's so annoying when you think of something nice to say, just to
notice you could only think of it after reading it elsewhere a few
minutes before…

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le vendredi 09 novembre 2007 à 11:14 +0100, Joerg Schilling a écrit :
> Other code that is not derived from the GPL code is not part of "the work":
> 
> - You do not need to put "non-derived" code under the GPL.

You are basing all of your reasoning on the assumption that a program
that uses a library isn't a derived work of the library. Which is
technically wrong and would never hold in front of a court.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le jeudi 08 novembre 2007 à 11:15 +0100, Joerg Schilling a écrit :
> "John Halton" <[EMAIL PROTECTED]> wrote:
> > As has been said already, the GPL does allow non-GPL code to appear in
> > GPL projects, but it requires that code then to be distributed under
> > the GPL. But to do so may infringe the licence for that non-GPL code.
> 
> This is a false claim! The GPL does not require to change the license of such
> other code and any lawyer would laugh on you as this would be illegal.

The GPL cannot of course change the license of some other code. However,
if that other code isn't under the GPL, you cannot link to it. It is as
simple as that. If the author wanted to allow that, he would have chosen
the LGPL. And the fact that you can *technically* link GPL code to
GPL-incompatible code doesn't make it more legal.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le vendredi 09 novembre 2007 à 11:59 +0100, Joerg Schilling a écrit :
> Please first rething the rest of your text as you did base your claims
> in a way that misses the fact that the GPL makes a clear difference between
> "the work" and "the whole source". GPL licensing only applies to "the work".

Yes, and it considers the work as a whole. That includes any libraries
that are required to make it run and scripts that are required to make
it build.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

RE: The legality of cdrecord

[Josselin Mouette]

Le jeudi 08 novembre 2007 à 11:07 -0800, Yuhong Bao a écrit :
> That is exactly why the code, not just the build scripts, are CDDL in
> current versions of cdrtools. Now the remaining problem is about the
> GPLed library that the CDDL mkisofs links to. Removing HFS support
> would solve this problem. Yuhong Bao

Maybe.

Unfortunately, the truth is that we don't care anymore. I don't know
anyone in the project who would now trust Jörg Schilling when it comes
to licensing or legal issues. And he has proved repeatedly to be
stubborn about technical issues which are important to our users.

Get over it. The cdrtools crap won't make it back in the archive.
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: The legality of cdrecord

[Josselin Mouette]

Le mardi 06 novembre 2007 à 22:10 +0100, Joerg Schilling a écrit :
> Don't belive a site that publishes an incorrect FAQ for their own license.
> Don't believe people who make inappropriate generalisations.
> Don't believe people who do not discuss specific license problems.

And above all, don't believe Joerg Schilling, like when he says:

> The GPL explicitely allows to use code under other licenses from GPL code.

No, it does not. If you think it does, please point the line where it
"explicitly" allows it.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: transitive GPL (exim4, OpenSSL, mySQL and others)

[Josselin Mouette]

Hi,

Le jeudi 08 novembre 2007 à 19:27 +0100, Marc Haber a écrit :
> (1)
> Is it ok to change exim's SSL library to OpenSSL in the current setup
> without violating the GPL for some of the library currently in use

As you said, libmysqlclient and exim are OK with linking with OpenSSL. 

The one problem that could remain is that of libperl. It is OK to link
libperl with libssl because libperl is dual-licensed under the GPL and
the Artistic license, which is compatible with the OpenSSL license, but
that makes another GPL incompatibility.

Fortunately, libmysqlclient also allows linking with code under the
Artistic license. As for Exim's exception, it is so broad that
practically speaking, it is as if it was licensed under the LGPL:

In addition, for the avoidance of any doubt, permission is
granted to link this program with OpenSSL or any other library
package and to (re)distribute the binaries produced as the
result of such linking.

It would be nice to get explicit permission from the Exim developers to
link with Perl code under the Artistic license, but it seems to me that
this whole mess is already legally redistributable.

> (2)
> Will it be a violation of the GPL to link exim to a
> GPL-without-OpenSSL-exemption-clause library in the future?

Yes.

> (3)
> Is this violation maybe already happening by virtue of linking
> indirectly to OpenSSL via libpq?

It would, if exim and libmysqlclient's exceptions weren't so broad.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: is the "lucent public license" DFSG-free?

[Josselin Mouette]

Le vendredi 05 octobre 2007 à 01:10 +0200, Francesco Poli a écrit :
> On Sat, 22 Sep 2007 14:38:56 +0200 Josselin Mouette wrote:
> 
> > Le samedi 22 septembre 2007 à 13:18 +0200, Florian Weimer a écrit :
> > > The whole license is CPL-based.  
> > 
> > Indeed. I guess that settles the issue.
> 
> I have to disagree.

I couldn't expect any different reaction from you.

> Unfortunately I do not have the time to do a detailed license analysis,
> at present.  But you yourself, as well as others on this list, found
> some issues in the license.

And I explicitly used the word "borderline".

> Nonetheless, as soon as a similar license which seems to be accepted
> is pointed out, you seem to be ready to close your eyes and pretend
> the issues have vanished, magically.

I don't think the issues have vanished, and I would certainly not use
such a license myself.

Still, the contributor indemnification clause is very bad in spirit,
but, as I already explained, I don't think it has practical
consequences, so we can pragmatically accept it.

The patent retaliation clause has more real-world implications, but
frankly I couldn't care less of some random company being bitten for
using software patents.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Bacula and OpenSSL

[Josselin Mouette]

Le dimanche 30 septembre 2007 à 20:24 +0200, Kern Sibbald a écrit :
> > > However, the concept of deleting parts of the license don't appeal to me.
> > >  I prefer the following which is a modification of my prior license that
> > > was accepted by Debian.  The modification makes my prior license a bit
> > > more specific -- i.e. it restricts it to OSI licensed libraries.
> >
> > Well, any exception that you add to the GPL can be removed, this is by
> > design of the GPL. This is also true of the other wording you suggested.
> 
> It isn't really important, but I'd be surprised if that is true as it the 
> author of the code can decide anything he/she wants by modifying the GPL.

Yes, it is possible, but in this case I don't think the resulting
license is compatible with the GPL; if these new terms must me
supplemented for all derived versions, that makes them "further
restrictions".

-- 
 .''`.   Josselin Mouette/\./\
: :' :   [EMAIL PROTECTED]
`. `'[EMAIL PROTECTED]
  `-  Debian GNU/Linux -- The power of freedom


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Bacula and OpenSSL

[Josselin Mouette]

Le mardi 25 septembre 2007 à 15:14 +0200, Kern Sibbald a écrit :
> Thanks for looking up the above -- very interesting.
> 
> However, the concept of deleting parts of the license don't appeal to me.  I 
> prefer the following which is a modification of my prior license that was 
> accepted by Debian.  The modification makes my prior license a bit more 
> specific -- i.e. it restricts it to OSI licensed libraries.

Well, any exception that you add to the GPL can be removed, this is by
design of the GPL. This is also true of the other wording you suggested.

Cheers,
-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: is the "lucent public license" DFSG-free?

[Josselin Mouette]

Le samedi 22 septembre 2007 à 12:06 -0400, Joe Smith a écrit :
> >I'm not sure I understand what this clause means. What if there is no
> >jury for the trial?
> 
> All this means is that should a trial arise, neither side will request a 
> jury to decide
> the questions of Fact. If no jury is requested, the default is a bench 
> trial, where the judge
> decides the questions of fact in addition to the questions of law.

Thanks for the explanation.

My point is that this clause doesn't make any sense e.g. in France,
where civil cases are never decided by a jury.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée