Yikes

[Ted Leung <tw...@sauria.com>]

Pardon my absence, but I've been occupied with having a baby.

Stefano, you can consider this my "active developer" response.

Here's my view of the current situation on xerces-j:

1.  There has been a huge amount of bad blood between the developers
at Sun and IBM.  Both sides feel that the other is not to be trusted.  Both
can point to actual historical events as evidence that their mistrust is
warranted.
Each side believes that they are in the right and are doing what is right
for
the Apache community.   Neither of them are.  People's motives are
constantly
being called into question.  As long as that continues to happen, we aren't
going to go anywhere.   The specter of someone's corporate affiliation
hangs over every word that someone posts.  If Sun the corporation and
IBM the corporation (not the 2 dev teams) want Xerces to be a community
developed product, they can make it happen.  Some executive somewhere
in each corp can issue an order to "make it happen by working together - no
matter what it takes -- and that includes busting product schedules."  I
have
no idea how motivated the appropriate execs at Sun and IBM are.  But this
would take a lot of pressure off the dev teams to "do the right thing" for
their
companies.   Sam, Duncan - go tell your bosses to tell the engineers that
the
right thing for IBM and Sun is whatever it takes to make Xerces-J *the*
open development poster child.   Make it more important for the teams
to talk to the world than for them to hit deadlines.   Set performance
milestones
based on how many new external committers there are in 6 months or a year.
Call me a dreamer.  But if it was important enough, it would get done.  This
needs
to get fixed -- it hangs over the list like nuclear winter.

2.  The development of Xerces-2 or whatever it's called is proceeding along
2
different tracks.  One track is trying to gather some consensus on the
requirements,
and another track is building  prototype architecture.  There has been some
concern
that the group working on prototype architecture is ignoring the group
working on
requirements.  There has been some concern that the group working on
prototype
architecture not be held back by the requirements gathering and voting.
There needs
to be some clarification on how these these two tracks are going to
interact, or we're
going to have another train wreck.

3.  The committers list for Xerces-J is composed of people who at one time
worked
at IBM or Sun.  The only exception I am aware of is Ed Staub who became a
committer
by virtue of his work on the requirements.   I think there are a number of
reasons for this:
a) The Xerces-1 code base is fairly inaccessible to a typical open source
developer -- one
who is not getting paid full time to work on XML parsers.
b) It is difficult for a typical open source developer to keep up with the
development rate
of a team of full-time developers who reside in a single physical location.
It is also difficult
for these developers to hit development deadlines -- witness the Linux 2.4
kernel and Tomcat
3.2 as examples.
c) The "negative energy" of the list.

I don't think that these 3 problems can be fixed by a rename, a new mailing
list and a new module,
unless we are going to get rid of all the people -- which I think would be a
huge mistake.

I'm happy to have some more neutral third party people participate -- that
can only help.
I 've definitely appreciated Ed Staub's presence here.   As far as an ASF
member joining
to help -- I'd like to see that happen, but given some of the political
water under the bridge,
I want to see a person that was respected by the "active developers".   I
don't know if that
person is Stefano or not (Jeff Rodriguez's reaction earlier today suggests
to me that maybe
Stefano is not).

I work for a small company which is taking up large amounts of my time and
energy to manage.
The best that I can do for Xerces-J at the moment is to help Ed edit the
requirements  - if that
means I have to leave the project because I'm not active enough, well,
there's not much I can do.

Ted

----- Original Message -----
From: "Stefano Mazzocchi" <st...@apache.org>
To: "Apache Xerces-J" <xe...@xml.apache.org>
Cc: "Apache XML PMC" <pm...@xml.apache.org>
Sent: Sunday, October 08, 2000 3:25 PM
Subject: Xerces-J active developers


> I hereby officially request the list of active developers for the Apache
> Xerces-J project.
>
> It's time to see who is in charge, why and judge their community
> involvement as individuals.
>
> --
> Stefano Mazzocchi      One must still have chaos in oneself to be
>                           able to give birth to a dancing star.
> <st...@apache.org>                             Friedrich Nietzsche
> --------------------------------------------------------------------
>  Missed us in Orlando? Make it up with ApacheCON Europe in London!
> ------------------------- http://ApacheCon.Com ---------------------
>
>