Re: Bug#196367: debian-policy: clarify what to do about priority mismatches
On Fri, 1 Aug 2003 13:41:47 +0200, Josip Rodin <joy@srce.hr> said:
> On Thu, Jul 31, 2003 at 04:51:29PM -0500, Manoj Srivastava wrote:
>> > No, I mean that a complete consistency in the set of 10K packages
>> > is practically impossible to achieve, let alone sustain. And then
>> > there's always situations where it seems wrong to demote all
>> > non-default alternatives to extra just because there has to be a
>> > default.
>>
>> Are you arguing that we should then give up having policy at all,
>> since it is all futile?
> No, I'm not. Read the next sentence. (This is not meant as a flame,
> although your sentence certainly seems like it was.)
Well, no, I did not mean it to be a flame either; since it was
a logical consequence of what seemed to me your stance.
>> > We can and should strive towards the goal, but insisting on that
>> > this must be done is not a particularly productive use of
>> > anyone's time and makes the Policy Manual more an idealist rather
>> > than a prudent document, and that is not in its scope.
>>
>> Ah. But policy never insists on anything. Policy just is. It
>> defines the rules that allow invariants to be maintained by
>> packages in Debian that allow for closer integration or improve end
>> user usage patterns. Policy does not have an enforcement arm.
> None of that means that Policy should have 'must' rules that are so
> idealist that they are pointless (barring the current case where
> Santiago says he worked it all out -- I've taken his word for it and
> didn't reply further).
Why is having proper priority on a package a goal that is so
hard to realize that it is idealistic? I mean, the fix to a wrong
priority is not hard; you change the control file, and petition ftp
admin with a mail stating the rationale for the priority change.
> "All software in our distribution must not have too many bugs." --
> that would certainly "allow for closer integration or improve end
> user usage patterns", but we're not writing that. Not because we
> can't enforce it, but because it would be pointless.
The difference here is that fixing all bugs, known and
unknown, is indeed harder and less deterministic than changing a line
in the control file and sending a mail message.
It may require a modicum of though, and perhaps some
communication with the maintainers of related packages -- and if a
maintainer things that is too much effort to ask for, should that
maintainer really be in Debian?
manoj
--
Quod licet Iovi non licet bovi. (What Jove may do, is not permitted to
a cow.)
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Reply to: