incubation process and new committers (was: [vote] Aaron Mulder as committer)

[Greg Stein <gs...@lyra.org>]

On Tue, Sep 16, 2003 at 07:41:46PM -0000, geronimo-dev-digest-help@incubator.apache.org wrote:
>...
> From: Richard Monson-Haefel <Ri...@Monson-Haefel.com>
> Date: Tue, 16 Sep 2003 01:30:33 -0500
>...
> There. That is honest feedback on a public list. I debated doing this. The
> fact that I debated speaking my mind at all, convinced me that private
> voting in the PMC is a must. You want honesty in a public forum, then people
> are going to be hurt -- or at least offended.  I'm sorry, Aaron if this has
> offended or embarrassed you in any way. I felt compelled to be honest.
> Apparently David Blevins and others feel you deserve commit access ... I
> respect there opinions and won't object.

Exactly. And even if the person is *not* offended (as Aaron stated), there
is *still* that tiny bit of doubt in the back of the votee's mind. A small
shift in the shading of how future posts are viewed. A new set of glasses
to read future posts. Simply look at Aaron's response and his questioning
of the commentary that Richard made. While Aaron wasn't "offended", he
*is* thinking that Richard's post wasn't entirely fair.

For the most part, we are all very nice people. You have to be in the open
source world because you (generally) have to work with people. And part of
being nice is being *considerate*. To that end, it is very hard to post
your innermost thoughts in a public forum, and especially where the person
can read them.

Adding a committer is giving somebody the right to modify ASF property.
That should not be a light decision. That means that candor is required,
and that can be done most effectively in a private environment.


Jason raises a good point, however: how does the Incubator PMC debate the
merits of the various candidate committers? Are they familiar enough with
the details? And this isn't even the half of it -- the PMC members paying
attention to this list are a pretty strong subset of the entire PMC. So
what is the resolution? Hard to say :-)  *That* discussion is an
interesting one. There are two opposing items here:

1) need for information to make a qualified judgement of committer status
2) need for the PMC to be the responsible party for that decision

This really only applies during Incubation, where the PMC members are not
the incubated project's active committers. One answer might be to
constitute the Geronimo PMC early, and get some of the right people into
that PMC (e.g. geronimo committers + some incubator PMC members). That has
a lot of ramifications, though, in terms of the overall ASF oversight that
has been described elsewhere -- that PMC (and thus, the "new" geronimo
committers) would need to start answering directly to the Board rather
than getting its feet wet via the Incubator PMC. And the Board still needs
the Incubator PMC to be involved, so there is potential for confusing
lines of responsibility...

Yes, the incubation process is "relatively" new if you haven't guessed :-)
There are certain requirements from the ASF side of things, but those are
not always clear, and the way to effect those are unknown in many areas.
As each project goes through incubation, we have learned more and more.
Just don't ask Andy Oliver about incubation: he was the first one, and
thus had *THE* most difficult time of anyone. Ever. He doesn't have good
words about the process :-)

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

Re: incubation process and new committers

[Greg Wilkins <gr...@mortbay.com>]

Greg Stein wrote:
> One answer might be to
> constitute the Geronimo PMC early, and get some of the right people into
> that PMC (e.g. geronimo committers + some incubator PMC members). 
> 

Just as the geronimo project/community needs to be incubated into the apache
way - surely the geronimo PMC also needs to be helped into the apache way.

I think there is a good case for forming the geronimo PMC early - even if
it has limited powers and is subject to the incubator PMC.

The geronimo PMC could start functioning now and give recommendations to
the incubator PMC - who could accept/revise/vito as the case may be.
Thus the geronimo PMC would learn the ropes - just as the geronimo community is.

While I'm here... if we do move discussions of committers to a closed
PMC list - I think that the workings of the process must be made public.
The PMC should reveal:

  + That it is considering and discussing a nomination for person X.
    Inviting any comments from the community.
  + The result of that discussion/vote.
  + If the result is negative, I think the PMC should at give a public
    short reason and a private detailed response as to why the nomination was declined.

cheers

Re: incubation process and new committers (was: [vote] Aaron Mulder as committer)

[Bruce Snyder <fe...@frii.com>]

This one time, at band camp, Greg Stein said:

After reading these threads for quite a while and offering very little
commentary, I'm going to jump in with my two cents.

GS>Jason raises a good point, however: how does the Incubator PMC debate the
GS>merits of the various candidate committers? Are they familiar enough with
GS>the details? And this isn't even the half of it -- the PMC members paying
GS>attention to this list are a pretty strong subset of the entire PMC. So
GS>what is the resolution? Hard to say :-)  *That* discussion is an
GS>interesting one. There are two opposing items here:
GS>
GS>1) need for information to make a qualified judgement of committer status
GS>2) need for the PMC to be the responsible party for that decision
GS>
GS>This really only applies during Incubation, where the PMC members are not
GS>the incubated project's active committers. One answer might be to
GS>constitute the Geronimo PMC early, and get some of the right people into
GS>that PMC (e.g. geronimo committers + some incubator PMC members). That has
GS>a lot of ramifications, though, in terms of the overall ASF oversight that
GS>has been described elsewhere -- that PMC (and thus, the "new" geronimo
GS>committers) would need to start answering directly to the Board rather
GS>than getting its feet wet via the Incubator PMC. And the Board still needs
GS>the Incubator PMC to be involved, so there is potential for confusing
GS>lines of responsibility...

That being said, IMHO, there needs to be a closed mail list for the
discussion of project level processes and possibly even decisions. This
list should include the proper shephards as well as the 'new' Geronimo
committers. This would allow questions to be asked in a closed setting
so that there is no further application of processes deemed to be
incorrect. As stated above this list would only remain intact during the
incubation period because part of the terms of exiting the Incubator
would then be that there is a proper understanding and application of
ASF processes and the ASF way of life.

Bruce
-- 
perl -e 'print unpack("u30","<0G)U8V4\@4VYY9&5R\"F9E<G)E=\$\!F<FEI+F-O;0\`\`");'

The Castor Project 
http://www.castor.org/

Apache Geronimo 
http://incubator.apache.org/projects/geronimo.html