About Debian Science Policy Manual

Teemu Ikonen tpikonen at gmail.com
Fri Aug 29 10:03:28 UTC 2008


On Fri, Aug 29, 2008 at 11:22 AM, Andreas Tille <tillea at rki.de> wrote:
> On Wed, 27 Aug 2008, Manuel Prinz wrote:
>
>> I'd like to know if people here think we should push this workflow by
>> our policy or encourage people to come up with there own ones. I'm very
>> interested in the opinion of people who have little or no VCS experience
>> so far (and are willing to give them a chance).
>
> I would be in favour of putting int to policy.  If people would completely
> disagree they could raise their problem with this policy here.  But it would
> be really helpful if we would settle down with something written soon.

The packaging style demonstrated in Martin Krafft's presentation is
nice (although a little branch-happy, e.g. why do we need a build
branch?), but I would be against putting anything like this into
policy for the simple reason that there are (AFAIK) no packages in
debian-science that use it yet.

Also, the tools are not really there yet. TopGit has the right idea,
but is still missing essential features. And can one even upload
version 3.0 source packages to the archive yet?

I converted the Imview source package to use TopGit branches lately,
but I'm not really happy with the way the patch series is generated in
TopGit. I'm slowly working on it though, so I may have a package which
actually builds after this weekend...

Teemu



More information about the debian-science-maintainers mailing list