Re: NMUs for Python Policy -- please hold them a little
Roland Mas <lolando@debian.org> wrote:
> Loïc Minier, 2006-06-23 16:40:06 +0200 :
>
> [...]
>
>> Probably the biggest reason why I feel this extra time would be
>> useful is because the new Python Policy and the tools supporting it
>> saw non-negligible changes in the last days. All of this only
>> settled very recently. This explains why maintainers have been
>> reluctant in moving packages to the new policy immediately.
>
> And here I was, thinking a policy was a collection of tried and true
> best practices turned into official status after they've been in use
> by most concerned packages for some time.
I think this is what should be expected most of the time. However, you
can't always achieve that: In this case, a change was to be made that
required updating application packages, "basic packages" of some area
(like the python* packages here) and helper packages (debhelper) at the
same time. This could /maybe/ be tested in experimental or some local
testbed. On the other hand, it also required to cater for different
packaging styles, maintainer flavors etc., and this cannot be tested.
Similarly, when we established the TeX sub-Policy, we also started to
write up what we already had agreed upon. Some things resulted in basic
TeX and add-on packages being in violation of must-clauses from the
start, just because the conforming basic TeX packages were only in
experimental, and hardly any add-on maintainer new about it. Of course
it took some time until things settled.
Regards, Frank
--
Frank Küster
Single Molecule Spectroscopy, Protein Folding @ Inst. f. Biochemie, Univ. Zürich
Debian Developer (teTeX)
Reply to: