Mailing-List: contact general-help@db.apache.org; run by ezmlm
Precedence: bulk
Reply-To: general@db.apache.org
Date: Thu, 18 Dec 2003 14:28:08 -0500 (EST)
From: Henri Yandell <bayard@generationjava.com>
To: general@db.apache.org
Subject: Re: mess   Was: [sql] move from commons sandbox to db or db-commons?
In-Reply-To: <1071774935.23243.79.camel@zakopane.forthill.int>
Message-ID: <Pine.LNX.4.44.0312181419460.10650-100000@umbongo.flamefew.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


Answering:

1) I fail to really see this existing. There's very little bureaucracy and
I take my hat off to the creators of the community as with very little I
think things cling together well.

2) Yep. Does J-C get more mail than httpd? One idea I have to help this is
to have a commons-vote mailing list that user-developers can avoid. Unsure
if there's much else though.

3) Standard problem for moving anywhere. You'd have the same issue in
moving to DB Commons if it has similar concepts. As far as I know, you can
vote a project straight into Commons and not the Sandbox, but it has to
have a community and not just be one person's play-thing.

4) Solved now. Any Apache committer can get commit access. The
infrastructure parts of adding someone from outside Jakarta slows things
down a bit I think and people seem to persist in thinking that some kind
of vote is needed.

5) Agreed, and I'm increasingly loud on the subject :)

Hen

On Thu, 18 Dec 2003, Brian McCallister wrote:

> On Thu, 2003-12-18 at 13:43, Henri Yandell wrote:
> > Out of interest, what part of the mess do you find to be a mess?
>
> 1) Jakarta-Commons is *HUGE* and the bureaucracy around it is as huge
> 2) The mailing lists are so high volume that I have trouble following
> individual projects, much less trying support users of a project
> 3) The project already exists, has had a release, and is in use
> 4) Until recently I couldn't get added to jakarta-commons-sandbox as I
> was a DB committer but not a Jakarta committer (oddly enough I got
> commit in Jakarta to work on a sandbox project started by someone else,
> go figure)
> 5) Jakarta shouldn't be a dumping ground for projects without a logical
> home. "Server Side Java" is a bit large a mission statement ;-)
>
> 3 is probably the biggest point. The jk-c-sandbox concept is broken here
> as OS projects thrive on "release small, releases often" but the sandbox
> says "you can graduate when you release (or vice versa)" This creates a
> situation where a project doesn't get used because it has never had a
> release, as it doesn't get used it cannot attract any additional
> contributions or feedback from users, etc.
>
> FWIW - I think that Apache is not a good place to start a small project,
> but I think it is a good place to move a project that already exists and
> is not yet mature.
>
> -Brian
>
>