incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Patrick Hunt <>
Subject Re: Concerning Sentry: A disagreement over the Apache Way and graduation
Date Wed, 04 Nov 2015 17:02:23 GMT
On Wed, Nov 4, 2015 at 8:32 AM, Joe Brockmeier <> wrote:

> On Wed, Nov 4, 2015, at 10:27 AM, Joe Schaefer wrote:
> > Joe, has any of this conversation put your mind at ease about the
> > podling?
> I'm less concerned than I was, yes. I'm still not in +1 territory. I'm
> not entirely sure I'm out of -1 territory.
> Sentry has made progress in its time in the incubator, but I feel it's
> required a lot of prodding at each step of the way - to reporting on
> time, adding contributors*, making sure its incubation status page is
> filled out, etc. It's also had some problems with release process, but I
> don't hold that against any podling because our release process can be
> hard to get right.
> But I view the podling as one that's concerned with releasing software,
> not growing community. I keep seeing references to "actively preventing"
> contributions - but I don't think that's a very high bar to clear. I
> want to see a podling actively working to make it possible to join and
> contribute.
> I'll note that I may see Sentry differently because I am a
> non-developer. The Jira-focused process may be adequate for folks who
> are primarily only focused on the release of software. It is not a
> particularly inviting or transparent process to anybody who might like
> to participate in Sentry in non-development roles. And I hope we care
> about contributors who will add value to Apache projects in
> non-development roles (documentation, marketing, translation, etc.).
> At any rate - I've said my piece, and I'll just reiterate that I don't
> think additional time is the answer. The signal I get from Sentry is
> that the podling feels it's ready to graduate, and they've indicated
> that they don't feel my suggestions are a "valid ask" - so I don't see
> much value in holding back a DISCUSSION and VOTE.
Personally I don't see this. To be fair, to my eye in each case when you've
brought up issues the podling has done their best to address them. They
even reached out to all the mentors recently and asked for feedback on
whether they are ready or not. The frustration podlings have is that
graduation is a moving target, even the mentors/ipmc can't agree. They are
trying to do their best, but growing a community is hard. They have been in
the incubator for two years, have built a useful tool, multiple releases,
have 30+ committers and 20+ ppmc members. Of course they want to graduate.
What I see in the sentry discussions is that they want to stay true to the
apache way, but don't want to do it artificially so. Just "ticking the
boxes" as has been brought up elsewhere in this thread.

> Note, as I understand it the board "is unlikely" to approve a podling
> where a mentor is voting -1. While I have concerns, I also don't want to
> filibuster the process and just keep Sentry in Limbo. I'd appreciate
> input from other IPMC folks on best decorum (e.g. abstaining from the
> vote, stepping down as mentor) in this situation. If other folks share
> my concerns, the vote wouldn't pass. If I'm wrong, I don't feel I should
> hold it up single-handedly.
If we can't reach consensus then we shouldn't go ahead. Your input is
valuable, that's why I'm spending my personal time on it. ;-)  However we
need to have some clear action items for the podling so that they have
something solid to build off. As it stands now I don't see a path to
graduation given the current IPMC climate.


> * I would invite folks with access to go to Sentry's private list and
> look over discussions about adding new contributors, and discussions
> about the project in general.
> Best,
> jzb
> --
> Joe Brockmeier
> Twitter: @jzb
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message