Re: Science linguistics and debian-science SVN repository at alioth
On Tue, 6 May 2008, Ross Boylan wrote:
Well, I perfectly understand your intend but finally I don't care
about a name but about work getting done.
I understand the sentiment, but in this case think the opposition is a
false one. If people with scientific, but not computer-scientific,
software to package think pkg-scicomp is not for them, they'll go
somewhere else.
I think your remark is very sensible and as I said I'm beaten myself
hardly with a bad name "Custom Debian Distributions" [1] I regard your
input as important even if Rudi Cilibrasi does not share your point
from a linguistic point of view.
That will impede coordinated effort. I don't know how
likely that is to happen, but it seems enough of a chance to be a little
careful with the name.
So I try to come back to my original proposal debian-science. I also
do not really like the 'pkg' in the name (even if I could live with it).
IMHO Debian and thus Debian Science is more than just packaging stuff -
it is about supporting users by providing some good infrastructure
including documentation web pages etc. Well, packaging is the most
important part of this - but it is not everything and in so far the
name part 'pkg' might be missleading as well. That's the reason for
the Debian Med SVN layout featuring "packages" and "community"[2].
I wouldn't call Debian Med SVN as a good example to adopt because
we are thinking about a restructuring - but the issue to include
packaging and other things into the SVN reflects the reality better
and should be applied in any form.
So back to the main issue: IMHO we need a well structured common
repository for the work of the Debian Science Team. I'd regard
taking over pkg-scicomp as the fastest way to approach this because
there is some content that would need a change which needs acceptance
of the former providers of this content (who do not seem to be very
active in this discussion unfortunately). But itaking over pkgscicomp
is actually the "quick-n-dirty" approach and if you think that this
is the wrong way I would be happy if David Bremner would start
on realising what he suggested in his last mail (and leave him the
decision to continue with pkg-science or just start from scratch with
debian-science).
I also have a question about WordNet. I know in the past there was an
issue with package what were essentially large databases (I think that
may have come up for astronomical and geographic data). Would that be
an issue for Wordnet?
Not that I know of. The term "large" is relative and WordNet is not
THAT large that it should be an issue. There are other package inside
Debian that ship ten times larger data packages.
Kind regards
Andreas.
[1] http://wiki.debian.org/CDDNamingProposals
[2] http://svn.debian.org/wsvn/debian-med/trunk/?rev=0&sc=0
--
http://fam-tille.de
Reply to: