incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <wr...@rowe-clan.net>
Subject Re: Incubation Process and PPMCs
Date Tue, 14 Mar 2006 02:32:37 GMT
Rodent of Unusual Size wrote:
> 
> William A. Rowe, Jr. wrote:
> 
>>>-- We should require 3+ Mentors for each project
>>
>>-1.  +1 to encourage 3+ to mentor, but it's not worth having folks
>>step up to mentor a project simply because 'they need three' -
>>quoting Justin's thoughts - with which I wholeheartedly agree.  We
>>don't need body counts, we need dedicated mentors.
> 
> And we need enough for some objectivity, too.  One mentor is
> not enough, IMHO -- insufficient oversight.

Agreed that it's always better to have 2 or 3, if only to react more quickly
if a project goes off in the wrong direction.  It's better to let people know
within hours of an idea why it doesn't work at the ASF, than let it build up
steam on it's own before shutting down the wrong direction.

As far as one mentor without a vested interest, that's a worthwhile thought,
let's see how often we can find uninvested voulenteers - it's something of
an oxymoron.  We voulenteer for what interests us personally.

>>>-- Upon acceptance, we should establish the initial PPMC as
>>>consisting of the Mentors.
>>
>>--1 - worst idea of the thread.  You can't teach new project members
>>to become project managers if you exclude them from the oversight and
>>management of their own project.  You graduate them to TLP status and
>>they look around without a -clue- how to solve problems.
> 
> Um, I don't think Noel meant that the PPMC *remains*
> consisting of just the mentors, but that they're the
> bootstrap code for it.

Presuming there is a community coming in - that entire community must begin
as part of the ppmc.  That said, the PPMC has no 'final authority' anyways,
it always rests with the incubator PMC.  Think of this as a PMC's training
wheels.  Why do the ASF members who start by mentoring the project need the
training wheels?  They are for the benefit of that future TLP/subproj PMC.

>>>-- Upon project acceptance, we should immediately create the
>>>$podling-ppmc@incubator.apache.org mailing list.  Or -private if
>>>there is a consensus to switch to that nomenclature.
>>
>>That's what we do today, AUIU.
> 
> Well, there are podlings with no PPMC, much less a PPMC mailing list.
> (Or were; that may have been corrected by now.)

That is a serious problem - in two years I never helped bootstrapped a future
TLP w/o one.  In the case of code intake/subproject efforts like cli or ftp
modules for httpd - a ppmc is actually an awkward thing... the code request
came from a TLP, and control will revert to that TLP, and the ppmc is a blip
on the radar with no expectation that those ppmc members will automatically
be considered for pmc status back at the home pmc once graduated.

For subproject efforts, I think this needs more hashing out with the submitting
PMC .. that the graduated subproject comes back with code -and- people, and
those ppmc members will be strongly considered to be folded into the TLP's PMC.
Anything else and the ppmc is just for show.

>>>-- vote on other PPMC members (only those on the Incubator PMC have
>>>binding votes)
>>
>>NO.  The entire Incubator PMC have binding votes, and ratify the
>>decisions made by the podling ppmc.
> 
> Bleah.  I think you've misinterpreted Noel again.  Only those
> *on the PPMC* who are also on the IPMC have binding votes.
> So more IPMC members join the PPMC, more binding votes.

Sure the ppmc can -grow-, but I'm arguing this little homogenous ASF
incubator pmc subcommittee is a bad idea.  Start from the ENTIRE incoming
community, existing authors and contributors.  Not some small cabal of
'us' picking and choosing.  Bleh :)

> Ever hear of delegation?  The IPMC can obviously (well,
> obvious to me) override PPMC decisions, but should *not* be
> involved in them day-to-day.

I think that's what is working where ppmc's are established.  They figure
out when they are ready to 1) add committers, 2) roll out a release, 3) ask
to be graduated... and we just ratify (or sometimes with good reason, turn
down) their decision at the incubator PMC.

I guess I'm a little surprised, having been around for a year or two since
this was all concieved, how there are projects not operating with one already.

Bill

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message