incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mattmann, Chris A (388J)" <chris.a.mattm...@jpl.nasa.gov>
Subject Re: Nomination of Chris Mattman for the IPMC Chair (was: Re: NOMINATIONS for Incubator PMC Chair)
Date Sat, 04 Feb 2012 01:47:40 GMT
Hi Ross,

On Feb 3, 2012, at 4:41 PM, Ross Gardler wrote:

> Sent from my mobile device, please forgive errors and brevity.
> On Feb 3, 2012 4:27 PM, "Mattmann, Chris A (388J)" <
> chris.a.mattmann@jpl.nasa.gov> wrote:
>> 
>> Hi Greg,
>> 
>> On Feb 3, 2012, at 1:44 AM, Greg Stein wrote:
>> 
>>> On Fri, Feb 3, 2012 at 00:58, Mattmann, Chris A (388J)
>>> <chris.a.mattmann@jpl.nasa.gov> wrote:
>>>> ...
> 
> ...
> 
>> "Oh, we have these sets of projects that are related, let's create a
>> meta committee that will [wrangle] them together, and then report
>> out on their status, share MLs, etc. etc."
>> 
>> Each and every time the above is presented, the argument against
>> (besides maintaining the status quo, which I honestly think is being
>> pushed here) is that there is no need for such a meta committee
>> (and by transitivity) a meta VP role. That's what the Incubator VP
>> is. A meta VP. We don't need the role.
> 
> This ignored a great many suggestions. Including my own (which I've not got
> sufficient time to flesh out right now, but you seem to be making
> significant assumptions about what flesh there would be on those bones).

Well what do you expect? Everyone is chiming in, and I'm doing my 
best to reply to all the emails but I can't keep up. 

Also, I will state that in response to "...you seem to be making significant
assumptions about what flesh there would be on those bones.", you are
doing *precisely* the same thing in your responses below.

> 
>> 
>> 
>>> As I
>>> mentioned before, I believe there are aspects to incubation that
>>> require a supportive group which cannot simply be shifted to the
>>> podling-TLP
>> 
>> I don't agree with this. It's shifted to the project TLP. That's OK.
>> Why is this not?
> 
> The incubator had demonstrated that relying on mentors is not always
> sufficient. The incubator has failed in it's guidance rule. It has turned
> to oversight and interference. Your proposal, in it's current form, will
> remove the interference but will not revive the guidance.

First off, the incubator isn't the only committee that's demonstrated
relying on mentors (which I'll substitute ASF members for) is not
always sufficient. *Every* single committee does that. That's why
they have a VP, but beyond that, that's why we have a board, and
the VPs are accountable to the board. And that's why the board
can replace a VP, b/c they don't always get it right; or because
of other priorities, or recommendations; or because they get 
TL;DR after reading Mattmann's emails. Or whatever.

Second, my proposal in its current form does the best job
I know of with the ASF of addressing guidance. It puts it
in the hand of a VP for a project (formerly called a podling) 
and that project's Project Management Committee.
And it also requires 3 ASF members to be there. That's the 
best way I know how to do it. And, if you'll recall Jim's message
to the members in the past 2 years about "internal belly gazing" or 
"navel" gazing or whatever the word was when he was talking
about the foundation as a whole and how even though there
have been little blips over the years, it's still largely been a 
successful model.

Finally, my proposal (despite Marvin's comments about boiling
the ocean -- no worries Marvin :) ) addresses what I think 
are concrete, actionable steps and does *not* try to boil
the ocean. Those concrete, actionable steps are things that
can be monitored and acted upon over the coming months 
(and yes they can even be reverted, though I hope not!).
Sure, I'd love for the Incubator to be dissolved at the next board meeting
because I think all the extra drum beating and "data points" 
and whatever that people want and keep clamoring for 
are already there if you've been active in the Incubator and 
watching which podlings succeed and which don't, and whatever, 
and I don't think the sky would fall, nor would we suddenly become
the wild west . But, I'm realistic. Based on the feedback so far, I 
am willing to concede that this will take time, months. I'm even
willing to concede that during that time, I'll (or someone else) will 
volunteer time to assist in that transition and report to the board.


> 
> The ASF is not just a place to host open source. It is a community
> learning how to do community is hard. The incubator was created to help
> that learning. Your proposal, as it stands will remove the interference but
> will not revive the learning.

The learning is done. Have you seen the 
oodles and oodles of Incubator documentation? Have you seen
the quality of RMs and the peeps that are rolling releases, responding
to feedback and navigating that body of work? Sure, there are snafus, 
but so what. It's a good body of work.

Everyone keeps pissing and moaning that getting rid of the Incubator is
a "punishment". It's *not* that in the least. Go back and read my initial 
response to Roy. I called it a "success". It is a success. It's produced tons
of successful Apache projects; great documentation; lots of good will and
press, and beer, and everything else. 

That being said, I'm a fan of measuring timelines for things and being realistic
about what we need, and what we don't. And I'm sorry; we don't need the 
Incubator PMC anymore, and no we don't need the Incubator.

Incubation, yes; Incubator, no.

> 
>>> The Board has enough to do without trying to
>>> *also* verify release processes, check on podling branding and press,
>>> etc.
>> 
>> You guys don't do that for projects, why would you do it in this case?
> 
> TLPs are self governing, podling TLPs could be, but your proposal assumes
> they alwats will be, from day one. Yet the incubator demonstrates mentors
> often fail in this regard.

And, experience over some recent TLP projects (called "umbrellas") over the past 
5 years at Apache demonstrates that even TLPs have this property too. 
So what?

> I am not involved in a single poddling that can
> mystery three binding votes on a release today.

My proposal doesn't fix that -- your podling and community needs to.
Every single podling I'm involved in *can* muster 3 binding VOTEs,
on everything from releases, to new committers.

> The IPMC fills the gap, but
> it also generates interference. Your proposal, as it stands, will remove
> the interference but will not maintain the necessary oversight.

Yes it will. It'll do the best job that I know of, and that I've heard
at the Foundation today to maintain the oversight. 3 ASF members, 
+ a VP + a direct artery to the board. What's better than that?

> 
>> 
>> To summarize in a sentence my proposal:
>> 
>> "Get rid of the Incubator PMC, its VP, etc and just start treating
> incoming
>> projects like Apache projects, day 1."
> 
> Most incoming projects are not Apache infects on day 1. Your proposal, as
> it stands, will result in a new kind of Apache. One in which the average
> standard of IP management is reduced.

Hrm. I don't think that conclusion follows any of the data I've seen on that one. And if
it does; I'd love to see your data that you're basing that conclusion on? If they're 
following the same Incubation process (just not with an oversight of an "Incubator PMC"
or a VP of that), and reporting directly to the board and really the membership as a whole,

I think not.

> One in which the strength of the
> average community is reduced.

Disagree; as we train more incoming projects to be Apache projects from
day 1 (which mind you, I'll state is really the way the Incubator works 
today, or should work), then we're strengthening the bonds between
incoming projects and the rest of the ASF. We're empowering new projects
to know what an ASF "board" is; what reporting is; what other ASF resources
are. As a TLP, we'll expose them directly to this.


> 
> Despite all this I like you're proposal.

Well shucks, you could have fooled me :) 

In all seriousness, thanks for the VOTE of confidence, Ross.

> It has a great deal of merit, but
> it is incomplete. All that is required for me to like it is to say, this is
> the starting point, lets identify suitable projects that can survive in
> this format. Work with them and address the issues that emerge. Later we
> can move the more complex projects into this format.

Yep that's kind of what I'm saying now by volunteering as the 
VP, Project Incubation in this role. I'm volunteering to make sure
this proposal gets implemented. I'd like it to happen within a definitive
timeline and think the foundation will be better from it.

Thank you for your detailed feedback!

Cheers,
Chris

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


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


Mime
View raw message