The big question so far has been should Lifecycle Extensions exist, but
should it be located in the central Avalon CVS--not to be confused with
Avalon Framework which is a separate and distinct work.

As I expected before Stephen's proposal there was a lot of opposition to
that step.  Therefore, I counter-propose that we leave Lifecycle in
Excalibur (it is already there), and focus on releasing the next set
of excalibur components.

There have been many points made both pro and con regarding interceptors
and extensions.  They take radically different paths to the same end
result.  In the end, I believe that interceptions will win--but we don't
have the time to rush an implementation out.  Nor do I wish to rush such
an implementation.

The lifecycle extensions package can still be supported in a pure
interceptor environment, so our user's interests will still be served.
Of course when they see how easy it is to remove the requirement for
lifecycle extensions they may simply do that.

Fortress has a need for them--it is how we are addressing issues like
supporting the instrumentation package, and my GUIApp framework at
D-Haven.sf.net uses it.  It is not as half-baked or hacked as some would
make it seem.  It is a simple solution to a simple problem.  It is
not intended to be the end all and be all of lifecycle management.
It is a tool for a specific purpose that suits the needs of several
people right now.

Therefore, I suggest it remains in Excalibur, and we continue on with
more productive things like finishing the releases.


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to