Re: Licsensing scientific data
On Fri, Aug 12, 2011 at 3:10 PM, Peter Cock <p.j.a.cock@googlemail.com> wrote:
> On Fri, Aug 12, 2011 at 2:41 PM, Andreas Tille <andreas@an3as.eu> wrote:
>>
>> Could anybody contact UniProt whether this whole issue is not intended
>> but just an effect of not (yet) updated web page?
>>
>> Kind regards
>>
>> Andreas.
>
> I've emailed a contact at UniProt to try and find out more, and in particular
> who we should be speaking to.
>
> Peter
>
> CC'ing Peter Rice from EMBOSS in case he isn't on this list, given
> the recent discussion stemmed from packaging EMBOSS:
>
> http://lists.open-bio.org/pipermail/emboss/2011-July/004161.html
> http://lists.debian.org/debian-med/2011/07/msg00146.html
>
Dear all,
I have now emailed UniProt regarding the recent discussion on the
Debian-Med list about the UniProt CC BY-ND 3.0 licence stemming
from the inclusion of example data from UniProt in EMBOSS (and
potentially other open source tools like BioPerl and Biopython).
My email has been received by the UniProt help desk system,
which assigned reference [help #61792] which should be
included in the email subject of any follow up correspondence.
The automated reply also warned that during the summer holiday
period (until end of August) there may be some delays in dealing
with queries.
Regards,
Peter
---------- Forwarded message ----------
From: Peter Cock <p.j.a.cock@googlemail.com>
Date: Mon, Aug 15, 2011 at 11:19 AM
Subject: UniProt CC BY-ND 3.0 licence
To: help@uniprot.org
Cc: jerven.bolleman@isb-sib.ch, Jerven Bolleman <me@jerven.eu>
Dear UniProt,
I'm contacting this address as suggested by Jerven Bolleman
(who works on the UniProt website). My query concerns the
UniProt CC BY-ND 3.0 licence.
Note that the Creative Commons has revised its advice and
now does not recommend Creative Commons licenses for
scientific databases:
http://sciencecommons.org/resources/faq/databases/
I contribute to Biopython and other open source projects too,
such as EMBOSS. An issue was recently raised regarding the
packaging of EMBOSS in Debian and the problem of EMBOSS
including some UniProt files as test data:
http://lists.open-bio.org/pipermail/emboss/2011-July/004161.html
http://lists.debian.org/debian-med/2011/07/msg00146.html
According to http://www.uniprot.org/help/license sample records
would be licensed under the CC BY-ND 3.0 licence. which has a
number of problems - first of which is it is incompatible with
Debian policy. This likely affects other biological libraries which
bundle sample data, including BioPerl and Biopython.
Second, the no derivatives part of the current license appears
to prevent simple tasks like format shifting:
http://lists.debian.org/debian-med/2011/08/msg00101.html
http://lists.debian.org/debian-med/2011/08/msg00102.html
Another concern is "attribution stacking" (which worries me
for practical usage of semantic reasoning and cross-database
searches), as discussed by the Creative Commons here:
http://sciencecommons.org/projects/publishing/open-access-data-protocol/
Please feel free to forward this message to anyone at UniProt
who can discuss this issue - ideally openly on the debian-med
mailing list since that is where most of the debate has
occurred to date, http://lists.debian.org/debian-med/
Thank you,
Peter
Reply to: