J Aaron Farr wrote on Wednesday, March 10, 2004 9:53 AM:
>     1. Why do I use Avalon?

Well, I used and switched. But initial interest was because of the availability of a 
lot of components combined with a clear guideline to develop own ones. Hoped to gain 
maximum reusability without a big overhead from the framework itself.

>     2. What do I feel Avalon's mission to be?

Fulfill its promise! Currently Avalon has a lot of components, but you never know, for 
which container they were designed and what you have really to do to run them in 
another context. Avalons mission should be to define the contracts for reusable 
components and deliver one container reference that can be easily tailored 
(size/features).

>     3. Where do I see Avalon by the end of 2004?

What I expect or what I would like to see? Looking at the past I am really doubtfull 
that the community-building works this time, but hope dies last. So , Avalon needs 
(IMHO):

- one reference container (that can be tailored form enterprise to single desktop app)
- developers that identify themselves with this container
- a clear specification about the contract between component and container
- a tck ensuring the functionality of the reference container and other 
implementations and ensures component compatibility

>     4. How do I feel about Avalon as an umbrella project vs. a single
> product? 

A unified effort. Umbrella is currently too big.

>     5. Should there be a formal framework specification?

Yes.

>     6. If so, what should it consist of?

Basic contract between container and component. Definition for container "concerns" 
and influence on the components. Definition of behaviour for missing unsupported 
concerns on either container or component side. Definition of the meta data, but not 
its format (attributes vs. XML should be supported by Interface).

Regards,
J�rg

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

Reply via email to