incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benson Margulies <>
Subject Re: Meta-question: How many committers on a proposal are enough?
Date Fri, 03 Jun 2011 12:45:51 GMT
Here's the dilemma of 'question 2' as I see it.

Many podlings launch here with a very small group. If they do a good
job of marketing, if the podling is an itch that a lot of people want
to scratch, then more and more people show up, and all is well.

However, and it's a pretty big however, most podlings are starting
with a much smaller and more tractable quantity of code.

So I can see why people are diffident about approving the podling
without some forward visibility to critical mass.

At the same time, it's very hard to predict the situation. We know it
won't be 'what if we threw a party and no one came,' because we expect
IBM to show up. We don't know who else will turn up. The decisions
people have made over the last year or so were conditioned by the
particulars, and an Apache podling would be a rather significant
change to the particulars.

On Fri, Jun 3, 2011 at 8:35 AM,  <> wrote:
> Benson Margulies <> wrote on 06/03/2011 08:02:25 AM:
>> There is a meta-question here: what are the criteria by which the IPMC
>> should evaluate a proposal?
>> 1. "Are there enough people on the proposal to plausibly start out?"
>> I think everyone agrees on this as a legitimate criterion.
>> 2. "Given the vast size of the codebase, is there any chance of
>> building a large enough group to maintain and enhance it."
>> I fear that this involves the application of a crystal ball, but
>> others may disagree.
>> 3. "How many people are detectable on the two existing projects, as
>> this will teach us something about (2)"
>> No. It won't. Others on this thread of perfectly eloquently explained
> why.
>> So, please make some new threads with some new subjects if you want to
>> argue my view here or any of the substantive questions.
> Done.
> I think these are good questions.  But can you recommend a plausible way
> to answer your question #1 without at least estimating an answer for
> question #2?
> And I'd alter your question #3.  The better question, IMHO, is not "how
> many people are detectable".  I don't think anyone has seriously advocated
> that.  But "how many people are active" or "how many people are
> responsible for 90% of the contributions" or similar questions are
> indicative.
> Regards,
> -Rob
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message