incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ant elder" <antel...@apache.org>
Subject Re: Podling Committer policy
Date Thu, 07 Jun 2007 10:40:47 GMT
On 6/7/07, Paul Fremantle <pzfreo@gmail.com> wrote:
>
> My apologies if I'm resurrecting something that was discussed extensively.
>
> It has been pointed out to me that:
> https://issues.apache.org/jira/browse/INCUBATOR-57
>
> The actual votes are held on the private lists. I'm not sure I
> understand that. I would have thought that - after having the pre-vote
> discussion on the private lists, the main vote would happen on the
> podling-dev and general@incubator lists.
>
> Can someone please explain the logic to me?


How come there hasn't been a formal [VOTE] on this as it's a policy change?
Its quite hard to keep track of what the current incubator policy's are, at
least [VOTE]s are easier to spot, with this its required just watching the
guide history in SVN:
http://svn.apache.org/viewvc/incubator/public/trunk/site-author/guides/ppmc.xml?view=log

Why did it need to change so that poddling votes are no longer binding when
voting in new committers? When a poddling votes in a new committer the new
person doesn't really get any power, their vote doesn't count for a release
or anything, all they get is write access to the poddling SVN which isn't
really even a 'real' apache project svn, and they will have submitted a
CCLA, its almost like having to get IPMC approval just to attach code to a
JIRA? This is also just for a new committer not a new PPMC member. The
current guide says voting in a new PPMC member _doesn't_ need any binding
IPMC votes, surely at least this should be swapped so its the new PPMC
member that needs the binding IPMC votes not the new committer?

The new way gives two ways of voting in a committer,  the private way, and
the public way (which is described as inferior).

The public way seems _really_ inferior because those IPMCers with binding
votes haven't been involved in any of the private discussion yet so there's
no assurances at all the the vote will pass which would be embarrassing for
all those involved.

The private way doesn't seem great either. If its anything like release
votes it can take weeks and some off list prodding to get the votes but the
poddling can't see what goes on on incubator private so they don't know if;
its almost there; or any votes at all; or ongoing discussion; or if their
email has even made it through moderation to the private list yet.

How about changing it so;

(1) incubator-private is notified that discussion of a new committer is
starting on the poddling's private list so IPMCers can participate in that
discussion;

(2) when the actual vote happens the incubator is notified again so they can
participate on the poddlings mailing list (ie not voting /discussing on
incubator-private where the poddling PPMC can't see whats going on);

(3) the IPMC peoples votes are done by lazy consensus so as long as there
are enough poddling PPMC votes the vote passes whether or not there are 3
binding IPMC votes

(4) the same (1) (2) (3) for a new PPMC member (maybe (3) should be dropped
for a new PPMC member?)

   ...ant

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