pd-zexy

Jonas Smedegaard dr at jones.dk
Tue Aug 17 16:36:31 UTC 2010


On Tue, Aug 17, 2010 at 03:34:57PM +0200, IOhannes m zmoelnig wrote:
>On 2010-08-17 14:56, Jonas Smedegaard wrote:

>> Would you mind me adding the CDBS copyright-check routine?
>
>i don't object, but:
>
>>
>> I can imagine how that may feel odd since you yourself are upstream 
>> for the code, but it does seem to me that you missed out some 
>> copyright and licensing pieces.  Not licensing conflicts, so nothing 
>> for you to worry about as upstream, but for Debian we care not only 
>> about that but also of properly recognizing all upstreams - not only 
>> main ones.

Please note that above is about redistribution for Debian and derived 
systems, i.e. debian/copyright and semi-automatically maintaining it.


Below is about upstream packaging, i.e. what might cause uncertainty for 
users and redistributors on what exactly thy are permitted to do with 
all the pieces that you offer them.


So I dare ignore your "but" above and (parallel to continuing below 
discussion) I will start working on copyright-check and rewriting 
debian/copyright using DEP5 machine-readable format.

If you disagree with that, please shout, so I can stop wasting time on 
something that you will revert anyway afterwards :-)


>> Upstream you might want to consider improving the licensing headers 
>> to (more clearly) declare copyright years.  Currently you state 
>> copyright like this:
>>
>>  * copyleft (c) IOhannes m zmölnig
>>  *
>>  *   1999:forum::für::umläute:2004
>>
>> There are multiple problems with this (IANAL, so YMMV):
>>
>>   * the term "copyleft" have no legal meaning
>
>but i guess that "(c)" is clear enough.
>
>>   * strings are expressed in latin1 (or latin13 or similar)
>
>this is surely no legal requirement
>
>>   * it is unclear if the second string is copyright years
>>
>> I recommend to instead use the following one-liner:
>>
>>  * copyright 1999-2004, IOhannes m zmölnig
>
>you translated it correctly :-)
>
>the thing is, that this piece of software is also to be seen in an "art 
>context" (phew!).
>the entire copyright headers are in a way a "logo".
>also the encoding confusion has a historic (for me) relevance.

I do not understand how that header is a logo, but that magic word "art" 
make me bow down in respect and not want to discuss any further.

It was just a suggestion (or, really, a recommendation), not a 
requirement (for you as upstream).

We (Debian) do need, however, to make sure that upstream copyright 
claims and licensing permissions are properly identified.

If the file headers do not (in our understanding of legal requirements 
for *redistributors* - not for *upstreams*) contain clear statements of 
both copyright and licensing, then we will need a written statement from 
those authors stating clearly the info we need.

This also, as I see it, implies that if e.g. a README file in the 
topmost directory of the upstream tarball claims that this project is 
Copyright authors A, B and C, but one file claims Copyright author D, 
then file takes presedence over README, and if then that file cannot be 
parsed reliably regarding *both* copyright and licensing, then that file 
is deemed unsatifiably licensed for redistribution.


You are free to not want to improve clarity of upstream copyright and 
licensing, but you risk some distributors then choosing not to want to 
redistribute your fine code.


>i would hate to change these.

Ah, that is a language I understand :-D


  - Jonas

-- 
  * Jonas Smedegaard - idealist & Internet-arkitekt
  * Tlf.: +45 40843136  Website: http://dr.jones.dk/

  [x] quote me freely  [ ] ask before reusing  [ ] keep private
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.alioth.debian.org/pipermail/pkg-multimedia-maintainers/attachments/20100817/18df2d2f/attachment.pgp>


More information about the pkg-multimedia-maintainers mailing list