Leo,

Thank you for the summary.  More comments below:

> -----Original Message-----
> From: Leo Simons [mailto:[EMAIL PROTECTED]
> 
> background
> ----------
> our initial plan was simple:
> 
> 1 - move utility code inside excalibur out to commons
> 2 - move container code inside excalibur to the main avalon cvs
> 3 - move components inside excalibur to the components cvs

Sounds good.

> Reasons for that plan were simple as well: our organisation is in a
> rather permanent structural mess, and flux. That led to various problems
> in the past.

I'll agree with the structural mess part, too.  I'm hoping to help fix that.
Part of my problem is I don't want to overstep my bounds and start
re-organizing or refactoring things without a solid plan that (most)
everyone agrees on.

<plug type="JIRA">
Yet another reason why I think issue tracking and roadmaps would help.
</plug>

> now what?
> ---------
> You tell me. Whatever needs to happen, its boring work :/. The first
> step seems getting our website, which is just as structurally in a mess
> (only periodically fixed), up-to-date with information about what
> components we have and the essentials surrounding those components.

Yes, it can be boring work -- but it _really_ needs done.  Particularly the
web site.  So how about I start working on that and once the site is more
presentable, I'll start bugging people about excalibur and cornerstone
re-organization again.  :)

> > [package structure]?
> 
> I would recommend against anything new. It will just add to confusion. I
> would suggest
> 
> org.apache.avalon.cornerstone.{blah} (api and/or spi)
> org.apache.avalon.cornerstone.{blah}.impl (implementation)
> 
> or
> 
> org.apache.excalibur.{blah}
> org.apache.excalibur.{blah}.impl

I brought this up because the new exporter component I put together for
Merlin is a much needed feature (easy JNDI exporting of components.  I at
least wanted something in place.  I'll refactor it to the cornerstone
package so as to maintain some consistency.

> Sorry to be so cynical, but I really think we should focus on getting
> our 'older' stuff into the same quality range as recent releases like
> merlin before focussing on 'new'. That, or just accept many components
> are 'stale' and treat them as such. Which has other downsides.
> 
> cheers!
> 
> - Leo

I don't think you're being cynical.  More like realistic. ;)

J. Aaron Farr
  SONY ELECTRONICS
  DDP-CIM
  (724) 696-7653

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

Reply via email to