W3C home > Mailing lists > Public > www-ws-arch@w3.org > October 2002

RE: Label for Top Node of "triangle diagram"

From: Sedukhin, Igor <Igor.Sedukhin@ca.com>
Date: Tue, 1 Oct 2002 15:40:17 -0400
Message-ID: <87527035FDD42A428221FA578D4A9A5B928300@usilms24.ca.com>
To: "Francis McCabe" <fgm@fla.fujitsu.com>
Cc: <www-ws-arch@w3.org>

Frank,

I guess it is not the triangle diagram, but the terminology that is
getting a lot of people confused here.

The triangle depicts those logical roles that you've nicely described in
your message below. I don't see it as a bottom-up really (even though
I'm sure some would like it to be). It's conceptual roles, based on
which we could start to roll other stuff in. It does not necessarily
have to map exactly into an implementation. For example an actual
service can be a client to some other service as well.

I like your classification approach that you've presented in your
tabular diagram. I'd favor to use it in WSA document when describing
meaning of different applicable technologies such as orhestration. The
exact content/names of the cells and tabs could be decided by the group
consensus.

"To support larger-scale interactions it must be clear -- to a computer 
-- what the roles, rights and expectations of the parties are."

It feels to me that it would be too much detail too quick for the basic
simple concept. When two people meet they don't necessarily present
photo IDs and liability disclosure documents upfront :). So, may be in
WSA we could similarily define and agree upon some fundamental conepts
first.

-- Igor Sedukhin .. (igor.sedukhin@ca.com)
-- (631) 342-4325 .. 1 CA Plaza, Islandia, NY 11788


-----Original Message-----
From: Francis McCabe [mailto:fgm@fla.fujitsu.com] 
Sent: Tuesday, October 01, 2002 1:11 PM
To: Sedukhin, Igor
Cc: www-ws-arch@w3.org
Subject: Re: Label for Top Node of "triangle diagram"


That all depends to whom one is trying to explain WSA to. I don't claim 
its the only diagram you need.

On the other hand, if a bunch of smart people can get confused over the 
triangle diagram, guess how the rest of the world will feel?

I would start to explain SOA by addressing the different kinds of 
things are of concern. Perhaps a bottom-up approach is easier for 
technologists to get their teeth into; but knowing where you are going 
is also useful.

I.e., something like the following:

The heart of WSA is enabling two computer systems to fulfill their 
roles in a common task; be it purchasing a book or logging an expenses 
report, or whatever.

In order for that to happen the parties must find each other.

To support larger-scale interactions it must be clear -- to a computer 
-- what the roles, rights and expectations of the parties are.

In order for the transaction to be valid it must be clear whom is doing 
what to whom; and why.

And so on.


I guess that, in this scenario, I wouldn't be over-emphasising 
individual MEPs but more staking out the territory the WSA covers. Then 
you go in and `fill-in'.

I would observe here, that the WSA WG's remit in `filling-in' is 
somewhat fuzzy: we don't do SOAP itself, we don't do descriptions 
themselves, we won't be doing choreography itself, etc. etc. We MUST 
try to get that global POV that helps the other groups relate to each 
other and to Web Services as a whole.

Frank


On Tuesday, October 1, 2002, at 09:33  AM, Sedukhin, Igor wrote:

>
> This diagram is a good classification that may help, but I would get 
> no idea what SOA is about by looking at it. Would you start describing

> Web Services and SOA from a diagram like that?
>
> -- Igor Sedukhin .. (igor.sedukhin@ca.com)
> -- (631) 342-4325 .. 1 CA Plaza, Islandia, NY 11788
>
>
> -----Original Message-----
> From: Francis McCabe [mailto:fgm@fla.fujitsu.com]
> Sent: Tuesday, October 01, 2002 12:15 PM
> To: Doug Bunting
> Cc: www-ws-arch@w3.org
> Subject: Re: Label for Top Node of "triangle diagram"
>
>
> This is a resend of a diagram I sent out earlier.
>
> It is different kind of diagram to the triangle as it doesn't focus on

> the players in the game so much. However,  see my previous comments 
> ;-> Frank
>
>
Received on Tuesday, 1 October 2002 15:40:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:09 GMT