Am 26.04.2010 16:16, schrieb Jonas Smedegaard:
What I meant to say was that I seem to recall other _in_this_team_ happy
to use CDBS as well. As Reinhard pointed out too.


What I questioned was a wording by Benjamin Drung that could only (to
me) be interpreted as "in this team we are phasing out CDBS - new
packages must use dh7 while older ones need not be converted right now".
I still question that.

I do not find that the wording is an attack on me or my personal style
of packaging, but rather that it is narrowing the options of packaging.

I agree with you. We should neither restrict the use of packaging tools too much nor should we declare one of them as deprecated and plan to replace perfectly working solutions with other ones just because they do not fit into the scheme.

Yes, I agree that new contributors are helped by a set of best packaging
practices. But I disagree that mandating specific tools are all helpful.



* we do code review, so please commit in sensible chunks
* most of us use short-form dh7, some use CDBS
* we use git-buildpackage with separate DEP3-hinted patches

With the above, I bet new contributors would choose short-form dh7
unless already decided on CDBS, simply because we clearly describe how
likely it is to get help using either style. Similarly a newcomer would
probably think twice before insisting on using e.g. Darcs since that
would be alien to the team (no matter if some in the team use Darcs in
some other contexts).

I find this idea really, really good. We should document what most of us do already use and not dictate others what they should use. I, personally, have no problem with a new contributor adding a package using darcs, but I will happily ignore any request for comments on this package. Exactly this should be documented.

Thanks Jonas.

Dipl.-Phys. Fabian Greffrath

Ruhr-Universität Bochum
Lehrstuhl für Energieanlagen und Energieprozesstechnik (LEAT)
Universitätsstr. 150, IB 3/134
D-44780 Bochum

Telefon: +49 (0)234 / 32-26334
Fax:     +49 (0)234 / 32-14227

pkg-multimedia-maintainers mailing list

Reply via email to