avalon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Leo Simons <leosim...@apache.org>
Subject The big picture
Date Sat, 29 Mar 2003 13:53:18 GMT
I keep saying it is important we look at the big picture as well. This
is to point ourselves at what we are talking about, and why we are
talking about it. Also mostly for reference.

Here's an important question:

   + What should avalon be 2 years from now?

we, as a PMC, are responsible for avalon, and we need to guide the
project in the right direction. We have, together with our friends from
the board and other helpful ASF members, identified a few things that we
believe avalon should start doing differently. Some of these things have
already started, others will take more time or haven't reached general
agreement yet.

There are many difficult questions with no easy answers. We have
identified some problems with the current containers we have, some
already being worked upon. Some of these problems include:

   + there exist(ed) subcommunities around each of the existing
     containers, and they are(have) hamper(ed) each other in development,
     and they make evolution of other parts of avalon more difficult.
   + the current many-container situation is confusing to users
   + the current many-container situation results in a lot of code to
     care for and maintain. This leads to a thin spread of developers
     across the various projects, resulting is lower code quality and
     less guarantee for continuity
   + the current many-container situation has resulted and will likely
     result again in multiple subcommunities within avalon where noses
     don't really point in the same directions with regard to "how
     do we work together during development" and other important items.
     The friction as a result of this is bad for the community

The Avalon community needs to move forward decisively to tackle these
problems and answer these questions. The ball is in the court of this

To this end, we have been working on fixing the issues which have
plagued the avalon community; things like ignorance of unawareness of
vetoes, the lack of respect between community members, the voting
process to be followed by the PMC, the broken or badly maintained
project infrastructure, the overdue product releases, the bad legal
state of avalon, and the gargantuan size of the avalon codebase.

Recently, we've been talking about the issue of dealing with our
existing containers and their further development. The subquestions at
hand here include:

   + What should happen to ECM?
   + What should happen to Fortress?
   + What should happen to Phoenix?
   + What should happen to Merlin?
   + What should happen to containerkit?
   + What should happen to ...
   + What should be the relationship between Avalon and external
     container developments?
   + What new container efforts should we start, if any?
   + How should we go about that effort?
   + What is and what should be the relationship between these
   + How do we remove (the perception of) little fiefdoms, one-man
     codebases, or inequality of committers?

We have talked about several possible answers so far (see the recent
summary I wrote, and the mailing list archives). We seem to be
converging on the notion that we indeed need to change direction, and
several proposals are on the table.


- Leo

To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org

View raw message