incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marvin Humphrey <>
Subject Re: July 2013 report
Date Fri, 19 Jul 2013 18:53:37 GMT
On Thu, Jul 18, 2013 at 11:15 PM, Till Westmann <> wrote:
> Yes, release votes are one point, another point is that we could probably
> avoid asking for votes for more than one RC on general@i.a.o if we had more
> ASF-experienced eyes looking at it on the dev list.
> Wrt to the other points you mentioned, collaboration seems generally
> effective (albeit not always on the list), technical infrastructure is not a
> big problem, and most things I learn about ASF culture I learn from
> discussions on general@i.a.o.
> I think that the mentor activity that we could benefit from is the
> occasional benevolent hint on how to make things work more smoothly within
> the ASF.

OK, I'll indulge, then[1]. :)

I think you're right to prioritize getting a release out the door.  Releases
are important not just for the features, but also for regenerating the energy
around a product in the wider community.  "Real artists ship." -- Steve Jobs

Since you've made it as far as creating two release candidates, presumably
there is not anything structural (such as a stalled code grant) holding things
up.  Here's where the two previous release candidate vote threads ended:

Getting a first incubating release out can be labor-intensive and the process
is sometimes frustrating.  Wrangling IPMC votes is a hassle, and it gets
harder the longer a podling is incubation because Mentors, like all
open-source contributors, come and go.  Even for podling contributors who are
truly voracious consumers of Apache documentation, it's hard to produce a
release candidate which encounters no objections from IPMC members.

Nevertheless, dozens of other podlings have bulled their way through this
phase.  The IPMC may not be as efficient or as coherent as we would like it to
be; things may not go "smoothly".  However, if there is sufficient energy
behind VXQuery (and no legal blockers with regards to the code base), your
release *will* get through eventually.  And while occasionally a podling will
have a Mentor who brings that kind of energy, that's not our expectation for
those who fill the Mentor role -- most often, it is the podling's core
contributors who must provide the push and sustain the momentum.

Marvin Humphrey

[1] We sometimes have problems where IPMC members who are not Mentors assigned
    to a specific podling provide guidance which is at odds with what the
    podling's Mentors have advised.  Let's hope that the content of this mail
    is suitably general and non-controversial.

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

View raw message