avalon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Noel J. Bergman" <n...@devtech.com>
Subject RE: On Multiple Containers
Date Sat, 23 Nov 2002 05:31:41 GMT
> The problem is that we don't know what works - we just know what doesn't
work.

That's fine as a statement in favor of rapid prototyping and proofs of
concept.  But I don't see that as not a valid justification for failing to
build a consensus, and consensus is not defined as everyone agreeing to do
their own thing.  If a concept looks like it may be either fragile or
diversely attacked, that is just a stronger impetus for the community to do
what ought to be considered best practice anyway: make a best effort to
figure out the proper abstraction, and isolate it from the rest of the
common infrastructure.  If there is concern about a given direction then
design for change and warn potential clients that a particular area is in
flux.  We understand and appreciate such candor, and it gives us a chance to
isolate our own use of that area.

I value your perspective on the history of attempts, failures and changes.
However, I do notice that your recital appears to consist of incidents where
people ignored your advice, and you now feel vindicated.  Now, I might agree
with you that some things should have been recognized as problems earlier,
but I would like you to consider the possibility that just perhaps the real
failure was not a failure of technology, but of leadership and community
process.

Peter, for the sake of argument, let's stipulate that you are right in every
case.  But if you cannot present your case in such fashion as to induce
agreement from your peers, then whose failure is it?

Disagreement is a normal part of the process.  The issue is how we resolve
them.  We had a thankfully rare and major (at least for us) disagreement in
James last month.  It was vocal, contentious and, yes, it was disruptive.
But in the end, we took the best ideas from both competing "camps" until we
ended up with something everyone could accept, and which really is better
than either of the starting points.  Anyhow, that's just an example.  We'll
have to see how the solution works out in hindsight.  But we did it as a
community, we support it as a community, and if necessary we'll fix it as a
community.

The point here is that leadership is not about having the best ideas.  I
would say that, especially in the ASF model, leadership is about working
within a community to build a  constructive consensus, and trusting the
process to produce quality products.

You have some great sounding ideas that you feel will bear fruit.  You've
also pointed out that Cocoon might be a concern in a couple of cases.  Well,
that consumer of the techology is here to give his good guidance and
feedback, and to help manipulate the design.  Please express your ideas, and
explain *why* they are right, but also listen for other ideas upon which to
build a *consensus*, and trust the process to work.  Give it a try.  It
might just work this time.

	--- Noel


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


Mime
View raw message