Re: seq-gen appears to be non-free
On 03/28/2012 02:25 PM, Andreas Tille wrote:
> Hi Alex,
>
> On Wed, Mar 28, 2012 at 12:00:38PM +0200, Alex Mestiashvili wrote:
>
>> I did that partially:
>> http://lists.debian.org/debian-med/2012/01/msg00241.html
>> Andrew Rambaut answered (ignoring CC to debian-med) that He has
>> permission from Yang to use this code, which is useless for us in
>> licensing question.
>>
> OK, thanks for the clarification.
>
>
>> Zhieng Yang asks to use a forum for questions and bug reports, so there
>> was no easy way to CC to debian-med.
>> Here is the thread:
>> http://www.ucl.ac.uk/discussions/viewtopic.php?f=54&t=9306
>>
> Thanks for the link. Stupid clash of incompatible communication
> channels, but you did right to ask there - thanks for your effort.
>
>
>>>> I would like to upload the new version of seq-gen (current watch file is
>>>> broken) because there is some number of users according popcon statistics.
>>>> How should I proceed now ? simply change section to non-free/science in
>>>> the debian/control and upload the new version?
>>>>
>>>>
>>> Depending on your answer to my question above it is probably
>>> contrib/science and yes if paml can not be freed this would be the way
>>> to go.
>>>
>>>
>> Good, than I will prepare the package for non-free.
>>
> Unfortunately there seems to be no work around. I have two remaining
> remarks:
>
> 1. It seems that parts of PAML should be moved to a library (considering
> the fact that you suspect that several projects might use this code).
>
>
As far as I see it is not correct for seq-gen package.
The header in files mentioned in debian/copyright has statement that the
code is taken from the PAML package, but in the PAML distribution I
couldn't find files with similar names.
It seems that only some parts of code are taken from PAML and I don't
see how making PAML a library can help in this case.
But it can be true for other packages if they exist.
> 2. I'm sad to see the pool of our packages in non-free growing. It will
> increase our maintenance burden and we should really try hard and
> regularly talk to upstream to change their license. I'm a bit sad
> that our PhyLip effort started at SouthPort[1] was basically ignored.
> I wished that people have an opinion on this if they read this list.
> It would help if people who not consider signing (which is fine for
> sure) could explicitely state their reasons for not doing so to
> let those, who are convinced about the petition might make aware about
> flaws in their approach. I would really like to seek with higher
> effort for successful methods to convince upstream authors.
>
> Kind regards
>
> Andreas.
>
>
> [1] http://wiki.debian.org/DebianMed/Meeting/Southport2012/ePetition_Phylip
>
>
Best regards,
Alex
Reply to: