oodt-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Justin Erenkrantz <jus...@erenkrantz.com>
Subject Re: [DISCUSS] OODT Podling Incubator Experiment (was Re: Radical revamp (was: an experiment))
Date Tue, 17 Aug 2010 04:56:00 GMT
[ CCing general@incubator as I think I can now place my finger a bit
as to why I'm discomforted with Greg's proposal in the OODT context ;
and more importantly, another potential experiment at the end; leaving
context in for those on general@incubator ]

On Mon, Aug 16, 2010 at 9:21 PM, Mattmann, Chris A (388J)
<chris.a.mattmann@jpl.nasa.gov> wrote:
> (moving to oodt-dev@incubator.a.o, context coming in separate email FWD)
> Hey Justin,
>>> +1 from me with my OODT hat on.
>>> Also, I like Greg's proposal b/c it puts the onus on those (proposed)
>>> $podling.apache.org PMC members who are active, without external "peanut
>>> gallery" oversight.
>> However, I think we should probably have a discussion on the OODT list
>> as we should think through what this means and how it'd affect the
>> nascent community.  With Subversion, it already had a very vibrant,
>> diverse, and self-governing community - OODT isn't quite there so
>> there's a bit of a risk there.  Perhaps this will act as a prod to
>> promote the self-governance - which is ideally what we want anyway.
> +1
>> At the moment, I probably don't have the time necessary to sit down
>> and lead the conversation within OODT.  That alone does give me a bit
>> of a reservation about what exactly we're signing up for.  =)  --
> To me, all we are signing up for with Greg's proposal is basically to have
> something like:
> 1. oodt.apache.org exists today
> 2. Ian, Chris, Justin and Jean Frederic are OODT PMC members + committers
> 3. OODT committers continue as-is
> 5. There is no more IPMC oversight
> 5. VOTEs on releases are approved by 3 +1s of OODT PMC members
>   - OODT committers weigh in on releases and their weigh in is taken into
> consideration by OODT PMC members (as is done today even with PPMC and IPMC)
> 6. VOTEs on new committers are approved by 3 +1s of OODT PMC members
>   - OODT committers weigh in on new committers and their weigh in is taken
> into consideration by OODT PMC members (as is done today even with PPMC and
> 7. When we're ready (we can even keep the same Incubator checklist), we put
> up a board resolution to "graduate" into *true* oodt.apache.org TLP. To me,
> ready =
>   - we've made at least 1 release (we're close!)
>   - we've VOTE'd in a couple new committers (keep those patches coming
> people!) hopefully with some diversity in mind, but if we don't get there,
> and the committers are still vibrant and healthy, then we move forward.
> OODT already has a pretty vast user community and healthy community that I'm
> slowing working to get signed up over here in the Incubator. We've had
> contributions from folks from Children's Hospital (thanks guys!), interest
> from other NASA centers (welcome Mark and others!), and some new folks from
> JPL stepping up and earning merit (welcome Cameron, and thanks for popping
> up Rishi!).
> Is that your take too?

Yes, I think that roughly outlines what Greg proposed.

See, here's where I get a bit discomforted by this entire process: I
honestly don't feel that I deserve a "vote" on OODT releases.  I've
known you and Dave for long enough that I have no concerns advising
the OODT community and trying to help out - but...giving me a binding

I want to encourage a process where the people doing the work get to
have the power.  At the core, that is what Apache is about - and
having doofus's like me casting a vote for a release seems like
straying from that.  I'm *totally* fine turning on "cranky" mode and
keeping the peanut gallery away so ya'll on oodt-dev@ get real work

For Subversion, I was already a full committer and earned my merit.
So, I had zero qualms about giving my $.02 there whether they wanted
it or not.  =)

Given your (Chris) experience with other ASF projects (and, heck,
being a PMC Chair), I can see exactly how the Subversion analogy (in
my head) applies to you.  You're a member, you know how things work,
you have merit within OODT - so, yah, perfect sense.  Smucks like me
who get confuzzled reading Maven build scripts?  Nah, not right that I
should have a binding vote.

Now, could we say that I would act as a "certifier"/"observer" that
all of the major processes were followed?  Heck yah.  No qualms there.
 Here's an analogy I'm coming around to: in a lot of new democracies,
there are "observers" who are sent in to monitor elections.  They
witness the elections, poke around, and make sure nothing unseemly is
going on.  They don't vote, but they do "observe".  They then issue a
certification or report to be filed with the vote.  (I'm catching up
on my backlog of issues of The Economist; just read their article
about nascent democracies in Africa on the plane...)

Hmm, maybe there's something to this "observer" model as this
reconciles my qualms and could provide the basis for an oversight
model.  Does this analogy move the needle for anyone else?  Could a
combination of "mentor" and "observer" be sufficient?  I think so...
-- justin

View raw message