[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: new maintainter process [was: Re: Orphaning my packages]



On Oct 02, Joseph Carter wrote:
> That's not to say that I don't believe James hasn't made bad calls in the
> NM area.  However, I can't rightly say I believe anyone else could do the
> same job any better.  The final stages of the NM process come down to a
> single judgement call.  If there is disagreement, who is right?  Until
> there is some reasonable way to determine that, there will be no sure-fire
> way for everyone to be happy with NM.  I don't believe this problem has a
> complete solution, either.

Well, it seems to me that if the buck needs to stop with someone, it
should be the DPL; after all, he nominally is empowered to replace the
DAM, so he ought to be trusted to overrule his/their judgment.  In
practice, I don't see this as a serious problem, especially since I
don't think James' workload would be reduced much if we added extra
people but had him review each application still (in which case adding
extra people is pointless).

That doesn't solve the issue of widening the trusted set to make these
determinations; I'm unwilling to put forward names of people who might
see it as being volunteered for duty, but it seems there are plenty of
obvious candidates: former DPLs, people trusted with other core
project functions (vote counting, monetary accounting, etc.), other
long-serving delegates of the project, etc.  But I think it's clear
that we often have serious problems where there is a single point of
failure in the project, and we should try to address these situations
(and more importantly, to make sure they don't come about in the first
place--because when they do, it leads to overwork or burnout that
often results in these very problems).

The irony is that we spend a lot of time seeking "co-maintainers" for
packages when we have a procedure to deal with AWOL maintainers
already (NMUs, package hijacking); however, we don't have a fallback
if the ftpadmin group goes on a two-month retreat to neverneverland or
Joe Cabal boycotts updating the LDAP database.


Chris
-- 
Chris Lawrence <cnlawren@phy.olemiss.edu> - http://www.lordsutch.com/chris/

Computer Systems Manager, Physics and Astronomy, Univ. of Mississippi
125B Lewis Hall - 662-915-5765

Attachment: pgpluXH2qoBt5.pgp
Description: PGP signature


Reply to: