Re: ISSUE: Inconsistency in number of properties

Oops.  I think you're right.  That would certainly explain all the "wait
a minute, which message is the inbound message" moments that seem to
come up in discussing the req-resp MEP.

If everyone agrees this is correct, I withdraw the issue, in which case
please read  "Message" as "the OutboundMessage at the sender or the
InboundMessage at the receiver, as the case may be." in places where
I've previously used the term.


noah_mendelsohn@us.ibm.com wrote:
> David Hull writes:
>
>   
>> Hang on.  The req-rep MEP has two properties for two messages. 
>> InboundMessage is the request and OutboundMessage is the response.
>>     
>
> This may explain why we're going in circles, because my reading of the 
> SOAP 1.2 Recommendation is clearly different than yours.  Table 6 says [1] 
> of the requesting node:
>
> "Initiate transmission of request message abstracted in 
> http://www.w3.org/2003/05/soap/mep/OutboundMessage ."
>
> Table 7 says [2] of the responding node:
>
> "Start making an abstraction of the request message available in 
> http://www.w3.org/2003/05/soap/mep/InboundMessage"  and "Initiate 
> transmission of response message abstracted in 
> http://www.w3.org/2003/05/soap/mep/OutboundMessage."
>
> So, I believe my suggestion for what we should do in one-way is entirely 
> consistent with the SOAP 1.2 precedent:  we should have a table with 
> sender properties that contains ImmediateDestination and OutboundMessage; 
> we should have a separate table for the receiver with ImmediateSender and 
> InboundMessage.  As with SOAP 1.2 Req/Resp, inbound and outbound are 
> always relative to the local node at which the work is being done. 
>
> Noah
>
> [1] http://www.w3.org/TR/soap12-part2/#tabreqstatetrans
> [2] http://www.w3.org/TR/soap12-part2/#tabresstatetrans
>
> --------------------------------------
> Noah Mendelsohn 
> IBM Corporation
> One Rogers Street
> Cambridge, MA 02142
> 1-617-693-4036
> --------------------------------------
>
>
>
>
>
>