W3C home > Mailing lists > Public > public-ws-addressing@w3.org > January 2005

i022 - Relationship to the SOAP binding framework (summary)

From: Greg Truty <gtruty@us.ibm.com>
Date: Sun, 16 Jan 2005 13:03:18 -0700
To: public-ws-addressing@w3.org
Message-ID: <OFB1FBAF0E.13B71C16-ON86256F8B.003FB22E-86256F8B.006E2A71@us.ibm.com>
It's been asked to write up a summary of where we stand wrt. i022.  Here 
it is...

Issue i022 - The rules of the SOAP binding framework need to be reconciled 
with WS-Addressing (for example, precedence)

Summary: David Orchard has written up an initial proposal (located at 
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0159.html
).   In it, he provided a WS-Addressing Adjuncts document w/a SOAP request 
MEP and a one-way SOAP HTTP binding (to allow the SOAP request/response 
MEP to be layered on 2 one-way SOAP HTTP bindings).  It leveraged the SOAP 
1.2 adjuncts MEP and binding section.  I refer to the original post (vs. 
duplicate the information that he laid out).

Open questions left (based on commentary by folks and unanswerd): 

Mark Baker:

- How do responses using this MEP differ from that of the SOAP 1.2 default 
binding and req/resp mep?  In addition, shouldn't a 202 be returned vs. a 
200?
- Wrt. the inability to send SOAPFaults via the HTTP response, why can't 
SOAPFaults be used as nacks?
- How does optional binding specific response relate to the forced 200 
(since the recipient has to handle non-200 response codes anyhow)?

Jonathon Marsh/Umit Yalcinalp

- similar to Mark's question about what cases should you know to use this 
MEP vs. SOAP 1.2 req/resp mep?  This lead to ambiguity between WSDL WG vs. 
XMLP on whose responsibility it was to define/allow the scenario where 
there is wsdl in-out, soap req-resp mep, and 2 soap req-resp bindings 
(where it's been precluded by the WSDL 2.0 WG).

Marc Hadley:

- Section 3.5.2.2 Receiving allows for a SOA message in the response HTTP 
entity body.  Does this allow a fault to come back over the HTTP response 
like the WSDL robust-in-only MEP?
- There is a statement about the fact that the bidning can allow an empty 
body, especially in cases where the action is sufficient.  This is an 
empty HTTP entity body, not an empty SOAP body - right?

Links to the discussion of the issue:
David Orchard           
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0159.html 
(original David Post)
Jonathon Marsh          
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0189.html
David Orchard           
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0190.html
Umit Yalcinalp          
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0191.html
David Orchard           
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0192.html
Mark Baker              
http://lists.w3.org/Archives/Public/public-ws-addressing/2004Dec/0186.html
Marc Hadley             
http://lists.w3.org/Archives/Public/public-ws-addressing/2005Jan/0024.html

Regards... Greg

Greg Truty
WebSphere Architecture/Development,  IBM Austin
EMail:     gtruty@us.ibm.com 
Phone:   (ext)  (512) 838-2828
                (Tie)  678-2828
Received on Sunday, 16 January 2005 20:03:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:35:01 GMT