[AMG] Figure 2.1 suggested revision.

Folks,

I've been playing with Fig-2.1 and thought I'd see what you folks think.
Attached are .PNG, .PDF and .PPT versions.

Basically, I've been trying to match the picture with our eveloving glossary
definitions. What I've done is to 'pull' the handlers back inside the XML
protocol layer and dispatch them directly from the XML protocol processors. 

From an operational/service model point of view outbound messages pass
through the handlers before being sent and on the inbound side the message
passes through the handlers before being delivered (or relayed in the case
of the intermediary).

I rather like the identification of XMLP sender/receiver in this version of
the diagram... it certainly fits with my notion of something (a user/client
of the layer) that has a message to send and the thing to which it wishes to
send it. I have left the oblique initiator/sender and responder/receiver
because sender/receiver does *NOT* fit with a request/response operation -
the removal of which I have yet to see a reasoned arguement for.

I have left the intermediary and intermediary operation in the picture, but
actually I think that they can go provided that we feel the handlers
attached to the intermediary XML protocol processor (Node III) cover all the
functionality of the intermediary.

Thoughts? Comments? Useful improvement or stick with what we have?

Stuart
 <<2001-03-11-Fig2-1.png>>  <<2001-03-11-Fig2-1.pdf>>  
<<2001-03-11-Fig2-1.ppt>>