drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Dunning <ted.dunn...@gmail.com>
Subject Re: Drill graduation vote
Date Thu, 02 Oct 2014 20:35:48 GMT
On Thu, Oct 2, 2014 at 2:11 PM, Julian Hyde <julianhyde@gmail.com> wrote:

> My understanding is that when a podling graduates to a TLP, all committers
> automatically become PMC members of the TLP. Correct me if I am mistaken.
>


I think that the project actually decides what to propose on this.

With Storm, for instance, mentors and early committers were offered the
choice.  Some declined, citing their lack of contribution, some opted to
continue involvement.  Graduation is a good time to make this sort of
decision.



>
> Can someone with more experience of this process please comment on how
> projects handle this? Particularly wrt initial committers, who, like
> god-parents, are often chosen for their virtue and wisdom rather than their
> willingness to get their hands dirty changing diapers.
>

Or chosen due to their initial enthusiasm.  Things change.


>
> I think that a bloated PMC prevents effective governance, and that now is
> the time to prune.
>

I think that the evidence is somewhat contrary to this.  A PMC with lots of
active non-contributors can be a problem.  This is very rare in Apache.  A
much more common problem is that a small PMC has trouble getting 3 really
solid votes for releases due to having a lean PMC.  Things happen.  Trips.
Babies.  Illness.  Life.

It is surprisingly easy to find yourself with 10 PMC members, all pretty
active but without 3 release reviews.

Mostly having too many PMC members means that many are silent.  With lazy
consensus, this is not a problem.


>
> Should the initial committers be given a chance to renounce their
> comittership? Some other mechanism to prune?
>

I think that is a nice way to do it.  On the other hand, it is nice to
recognize some early committers who contributed early on but may not be
able to currently.

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message