incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Roy T. Fielding" <field...@gbiv.com>
Subject Re: [DISCUSS] incubator voting process
Date Fri, 10 Nov 2006 19:39:00 GMT
On Nov 10, 2006, at 7:18 AM, Sam Ruby wrote:

> There seems to be a persistent delusion that  
> general@incubator.apache.or is where incubation happens.  The  
> reality is that all the real incubating happens on the PPMC private  
> and dev lists.
>
> We can correct this in one of two ways: recognize what actually is  
> happening, or make what we say is happening actually work.
>
> If we go the first route, then we should treat this list as an  
> announcements list.  Results of votes for releases and graduation  
> are posted here, and incubator PMC members will be given 72 hours  
> to raise an issue.
>
> If we go the second route, then we need to set the expectation that  
> most PMC members participate in most votes.

Sorry Sam, that is absolute nonsense.  The reason why we have a
minimal quorum requirement of three +1 votes is so that the vast
majority of PART-TIME, VOLUNTEER open source developers can focus
on what is important for them today.  In particular, a +1 vote to
release should never be given by ANYONE unless they are reasonably
sure that the release is good for Apache.  It is the responsibility
of the people who want to make a release (on any project) to
attract enough real review from responsible PMC members to generate
the three +1 votes.  If the product cannot attract that many votes
with relative ease, then there is obviously something wrong with
the release (socially or technically) and it should not be forced out.
It should be left on the pile of almost-done releases, with a note
to the effect of "needs more review".

If a podling has enough committers to vote for a release, the IP
constraints have been cleared (a prerequisite), the mentors
have voted for the release, and they *still* don't have enough
binding +1 votes to actually make the release even when the rest of
the Incubator PMC is off in the woods navel-gazing, then that
podling SHOULD HAVE ALREADY GRADUATED.

Apache Jackrabbit did only one release in incubator.  It is not
necessary for podlings to generate PUBLIC releases every few weeks
just to hit a milestone.  Just do one.  A podling with an active
community and proven IP-completion should prepare one release,
seek out individual reviews if necessary, and then graduate.

The last thing we need is for Incubator PMC members to be guilted
into voting for a release that they care nothing about.

....Roy


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


Mime
View raw message