On Sun, 29 Oct 2006 01:10:49 +0800 Tim Post wrote:
> On Sat, 2006-10-28 at 18:14 +0200, Francesco Poli wrote:
> > Hi all! :)
> >
> > I filed a RFP wishlist bug about a very interesting
> > scientific/technical data visualizer called VisIt (see bug #395573
> > [1]).
[...]
> > [1] http://bugs.debian.org/395573
[...]
>
> My concerns aren't really legal .. What I'm worried about is creating
> disappointment in the software from botched installs.
>
> I'm not so sure this will lend well to conventional "single"
> packaging. Most people working on projects serious enough to be
> interested in it would (likely) be working on parallel processing
> architecture.
>
> http://www.llnl.gov/visit/FAQ.html#10 suggests some things that I
> think are a bit out of aptitude's capabilities. Not that aptitude
> could not do it, but it would get .. interesting :) I think it may end
> up deterring people from the software because it didn't work correctly
> after installing (on any platform.) Look here:
>
> http://www.llnl.gov/visit/FAQ.html#6
>
> I'd be a little worried (again) that [binary] packaging it would end
> up giving it a bad rep. Scripting this would be a challenge.
[...]
I don't quite understand your packaging concerns, to be frank (but we
are going off-topic here): I thought that a split into two packages
(visit-server and visit-client) or maybe into three packages
(visit-server, visit-client, and visit-common) could solve situations
like this one (think about dhcp3, for instance).
Anyway, I will drop a message on debian-science@l.d.o and hear if anyone
there has experience in packaging software with similar
parallel-engine/frontend architecture...
P.S.: Please do not Cc: me on replies, as I didn't asked to be Cc:ed.
Thanks.
--
But it is also tradition that times *must* and always
do change, my friend. -- from _Coming to America_
..................................................... Francesco Poli .
GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4
Attachment:
pgpL79Nhhbc1f.pgp
Description: PGP signature