Re: Doubts about packaging of OpenBUGS
On Thu, Jun 05, 2014 at 06:17:07AM +0200, Dylan wrote:
> Hi all,
> Andreas, sorry for CCing you, I thought you were interested in this topic
> since I have proposed to package it in my previous mail [1].
>
> I think it could be interesting in Debian Med since this software is used
> for in biomedical research (e.g. some examples of current course/workshop
> presenting OpenBUGS [2,3])
Ahhh, thanks for these pointers. I simply failed to see the relevance,
but that's fine.
> Moreover it seem to have some interest [4,5]. The paper [6] published in
> "Statistics in Medicine" seem to be cited more than 560 times on Google
> scholar and 80 times on PubMed and the paper [7] about the original version
> (only available on MS Windows) is cited more than 2680 times on Google
> scholar.
>
> As OpenBUGS is a statistical software, maybe it is better if I propose this
> package to Debian Science?
Well, finally it does not matter what team will care for a package in
the end as long anybody cares to make our users happy. We are
maintaining some quite general packages in Debian Med team simply as
predependencies for pure medical packages or simply because the
maintainer of a package feels stronger connected to Debian Med. So
from my point of view it is fine and if you would go on with the
packaging feel free to do so. Just aks here if you need any help.
Many thanks for your input
Andreas.
PS: I'll do my "paperwork" for the new contributors game today - I just
had some past-vacation backlog ...
> [1] https://lists.debian.org/debian-med/2014/03/msg00222.html
> [2] http://www.crfr.ac.uk/bayesian-disease-mapping/
> [3] http://academicdepartments.musc.edu/phs/news/SummerInstitute_2014.pdf
> [4] https://bugs.launchpad.net/ubuntu/+bug/191101
> [5] https://www.jiscmail.ac.uk/cgi-bin/webadmin?A0=BUGS
> [6] Lunn, D., Spiegelhalter, D., Thomas, A. and Best, N. (2009) The BUGS
> project: Evolution, critique and future directions (with discussion),
> Statistics in Medicine 28: 3049--3082.
> [7] Lunn, D.J., Thomas, A., Best, N., and Spiegelhalter, D. (2000) WinBUGS
> -- a Bayesian modelling framework: concepts, structure, and extensibility.
> Statistics and Computing, 10:325--337.
>
>
> 2014-05-19 8:14 GMT+02:00 Andreas Tille <andreas@fam-tille.de>:
>
> > Hi Dylan,
> >
> > On Sun, May 18, 2014 at 06:36:59PM +0200, Dylan wrote:
> > > Hi Andreas,
> > > I have some doubts about the packaging of OpenBUGS
> >
> > I have not even thought about this software. Why do you think this is a
> > topic for Debian Med?
> >
> > Kind regards
> >
> > Andreas.
> >
> > PS: There is no reason in CCing me. :-)
> >
> > > which are already
> > > mentioned in the Fedora mailing list [1] and quickly in the Debian
> > Science
> > > mailing list [2].
> > >
> > > To summarize, the main issue is about the core of OpenBUGS
> > (libOpenBUGS.so)
> > > because it is written in Object Pascal which requires a compiler
> > framework
> > > only available on MS Windows (Black Box). In the "source code" available
> > on
> > > their website, "libOpenBUGS.so" is given as binary already compiled.
> > >
> > > I don't know Debian opinion about this issue. Maybe it is possible to
> > > package libOpenBUGS.so as "non-free" and others OpenBUGS components as
> > > "contrib"? Or it's unimaginable to have this software in Debian
> > repository?
> > >
> > > Best regards,
> > > Dylan
> > >
> > >
> > > [1]
> > https://lists.fedoraproject.org/pipermail/devel/2011-June/152747.html
> > > [2] https://lists.debian.org/debian-science/2008/12/msg00107.html
> >
> > --
> > http://fam-tille.de
> >
--
http://fam-tille.de
Reply to: