About Debian Science Policy Manual

[Frederic Lehobey]

Hi,

  I have not been involved in policy discussion so far as I am still
very new to git (my experience until last week was mainly with arch
and mercurial and a little of darcs).

  Thanks to debcamp I have now played quite a bit with git. This is a
great tool (though more different from mercurial than what I thought).

Manuel Prinz <debian at pinguinkiste.de> (2008-06-23 10:01:46) :
> Am Montag, den 23.06.2008, 08:57 +0900 schrieb Charles Plessy:

> >   Would a separate debian branch, that would only contain the debian
> >   directory, make any sense?
> 
> Well, if it makes sense for you, then yes. ;)
> 
> Seriously, it is of course possible. If you like things that way, you
> should go for it. But if you want to use git-buildpackage (I recommend
> that) you need to have an integration branch. This is not problem. You

  I would say yes because it is closer to the maintenance practice
described here:
https://penta.debconf.org/dc8_schedule/events/233.en.html (video
should be available soon). Even though the maintenance described by
Martin F. Krafft might be seen a bit complicated, I advocate being as
close to it or as compatible as possible with it, as its main goal is
being able to share patches and repositories with other distributions.

  I think debian-science is a future target user of the vcs-pkg tool
that is currently being designed there http://vcs-pkg.org.

  The communities around scientific tools are small and I think it
would be awesome to share patches with other packagers of scientific
software like FreeBSD, Fedora (I am thinking of Scientific Linux
distribution for example) or Mandriva.

  Just food for thought. I think we should have such a goal in mind
when setting up our own policy.

Best regards,
Frédéric