Re: pkg-science, again
On Mon, Feb 25, 2008 at 2:16 PM, Andreas Tille <tillea@rki.de> wrote:
> On Mon, 25 Feb 2008, Sylvestre Ledru wrote:
>
> > This group could take inspiration from pkg-scicomp or debian-med to
> > create his own policy (goals are very similar to pkg-scicomp).
>
> I'm a little bit concerned about the nomen est omen _pkg_-science
> and _pkg_scicomp. IMHO we need more than just packaging random
> scientific software. What Debian-Med does and what a CDD really
> makes is forming a community around the software in the focused area,
> try to teach them, building a network of cooperatioen etc. It is not
> only a technical thing like pure packaging. It is about making
> Debian the distribution of choice for scientists.
Well said. I have been suggesting this for years. I believe the
scientific method is generally thought to include "verifiability" and
particularly "reproducibility" as criteria for a well-performed
experiment. Experimental computer science, algorithmic, or machine
learning results are fairly baseless unless accompanied by open source
software. Therefore, I have claimed for years that all honest
experimental computer scientists should make a point to determine the
most standard way to offer their source code to the world, and follow
that convention. I think at first this meant autoconf and automake
for me, but as I tried Slackware, RedHat, and Debian, I found out that
Debian performed the best (in terms of maintainability and reliability
with controlled circumstances) in the longer (multi-year) term. So I
moved from spec files to debian directories and have never regretted
it.
I now write emails encouraging other scientists along these same
lines, to remember our scientific credo that is ostensibly supposed to
include the scientific method. I see it as part of the 'open science'
movement [1].
Cheers,
Rudi
[1]: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=886906
>
> Kind regards
>
> Andreas,
>
> --
> http://fam-tille.de
>
>
>
>
> --
> To UNSUBSCRIBE, email to debian-science-request@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
>
>
--
Which is worse, ignorance or apathy? Who knows? Who cares? --Erich Schubert
Reply to: