incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "robert burrell donkin" <robertburrelldon...@gmail.com>
Subject Re: Getting the Incubator PMC to be more responsive (Re: [VOTE] Incubator PMC to approve the 4.0-RC2 release of ActiveMQ)
Date Mon, 10 Apr 2006 12:28:45 GMT
On 4/10/06, James Strachan <james.strachan@gmail.com> wrote:
>
> On 4/10/06, Jochen Wiedmann <jochen.wiedmann@gmail.com> wrote:
> > On 4/10/06, James Strachan <james.strachan@gmail.com> wrote:
> >
> > > incubator? Then at least if the Incubator PMC are too busy to get
> > > involved on a particular project, folks from the sponsoring PMC can on
> > > their behalf.
> >
> > I would think that to be contrary to the incubators intentions. In
> > effect, it would lead to a situation where the incubating PMC would in
> > fact do the incubation. If that would work, then I'd believe that the
> > Incubator would never have been invented.
>
> Note I'm only talking about getting releases out here; the Incubator
> PMC members would still have their veto on any release vote and carry
> out the same duties they do during the incubation process.


IMO we should be convinced that the current system cannot be fixed before
adding new rules.

Maybe this new rule should only apply once one release of the podling
> has been performed? e.g. in ActiveMQ's case we had lots of Incubator
> PMC involvement during the first release, then there's been a natural
> tail-off of Incubator PMC interest since.


perhaps the move to increase the number of mentors should help. three
mentors should give enough +1's to proceed in the absence of any -1's. not
that this solves the current issues :-/

Anyone got any better ideas for how to increase the involvement of the
> Incubator PMC in the release process? The offer of free beer maybe? :
> )


i'm out of no silver bullets, i'm afraid :)

are the podling and incubator pmc votes run concurrently ATM?

FWIW I personally like projects to release early and release often, so
> I can see this problem getting worse soon.


 +1

if i'm going to rigourously check a release for quality then it'll take at
least an hour and much longer for an unfamilar code base. this time adds up.

but is this really what the incubator PMC should be doing when checking a
release?

it takes me around 15 minutes to check that a release satisfies the minimum
standards for apache releases: licenses, notices, signatures, MD5s and so
on.

is it clear what standards the vote should be applying and what a +1 from
the PMC means?

- robert

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