RE: Under Merlin's hood (WAS: RE: [avalon] roadmap - library crit eria )

["Farr, Aaron" <Aa...@am.sony.com>]

> -----Original Message-----
> From: Hamilton Verissimo de Oliveira (Engenharia - SPO)
> >
> > Can we dig into this a little further.  In particular - I'll like to
> > known what the underlying (real) objective is:
> >
> >     1. a "sealed deliverable" (i.e. something that the client
> >        cannot tinker with) ?
> >     2. simple installation without network access ?
> >     3. something else ?
> 
> Here it comes again.
> 
> - I don't want X
> - Why? Whats the underlying (real) objective to not want X?

I don't think Stephen's trying to be difficult at all.  He has particular
views about what makes a good container and is trying hard to convince
everyone else. :)

Back to the question:  Why would I like a single jar?  Well, sometimes it's
not _just_ the ability to have a single jar, sometimes it's having the
option to use something other than repository format, such as:

  * one large single jar
  * one single jar directory (/lib)
  * using an already existing classloader (perhaps from the CLASSPATH)

Why?

  * Simplicity
  * Webapps (using WEB-INF/lib)
  * Simplicity
  * Developers and users are "used" to these sort of layouts
  * Sealed jars (sometimes needed)
  * Simplicity

I certainly see it could be done and the balls as much in my court to write
a new implementation as any one else.

On a slightly different subject, I'm going to try to pull together the
various threads on roadmaps, meta, and containers and try to get us back on
track to come to a conclusion.  :)

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

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