RE: Proposed replacement text for Section 1.6

The way I understood Hao's statement is different. In my view it refers to the fact that in a given SOA system (e.g. a set of Web services participating in some kind of choreography/orchestration) the relevant interfaces remain the same (e.g. in BPEL you define the WSDL interfaces once, even though at run time you may associate different Web services to those interfaces, even dynamically).

My clue to that interpretation is the reference to a "small set of interfaces", which I understand to mean that we are talking about a single SOA system and its associated interfaces.

So the word "ubiquitous" might be confusing, but something else like "consistent" might be appropriate to illustrate the concept I mentioned before.

Ugo

> -----Original Message-----
> From: www-ws-arch-request@w3.org [mailto:www-ws-arch-request@w3.org]On
> Behalf Of David Booth
> Sent: Wednesday, January 14, 2004 7:50 AM
> To: He, Hao
> Cc: 'Champion, Mike'; 'www-ws-arch@w3.org '
> Subject: RE: Proposed replacement text for Section 1.6
> 
> 
> 
> At 03:25 PM 1/12/2004 +1100, He, Hao wrote:
> >. . .
> >2. Two main architectural constraints of SOA: 1) A small set 
> of simple and 
> >ubiquitous interfaces to all participating software agents.
> 
> I think we better NOT say "ubiquitous" here, because that 
> makes it sounds 
> like Web services offer a "uniform interface" (a la REST).  
> According to 
> REST folks, one of the problems of WS is that they do NOT 
> obey the uniform 
> interface requirement.  Instead, each WS defines its own interfaces.
> 
> I think we can just drop the word "ubiquitous" and the above 
> would be fine.
> 
> 
> -- 
> David Booth
> W3C Fellow / Hewlett-Packard
> Telephone: +1.617.253.1273
> 
>

Received on Wednesday, 14 January 2004 11:41:42 UTC