W3C home > Mailing lists > Public > xmlp-comments@w3.org > February 2007

WSDL WG review of SOAP1.2 Part 3 Oneway MEP

From: Jonathan Marsh <jonathan@wso2.com>
Date: Thu, 8 Feb 2007 09:52:06 -0800
To: <xmlp-comments@w3.org>
Message-ID: <025201c74ba9$da321520$3501a8c0@DELLICIOUS>

The WSDL WG agreed to forward the following comments, collected by
Jean-Jacques, on its behalf.

Note: The comments are in reading order.

Section 2.2 Description
1. What does "nearly identical" mean? A later sentence (starting with 
"Typically") seems to indicate that the difference between two "nearly 
identical" messages could be more than just "their destinations".

2. s/meps/MEPs/

3. s/is not/is NOT/

4. "is not required to detect [...] but should fault": the sentence is 
somewhat confusing.

5. "A receiving node MUST determine whether a given message is 
successfully received": this is likely outside the scope of SOAP. The 
SOAP 'middleware' may not even be able to detect the message was not 
successfully received. Suggestion: "When a message is successfully 
received by a SOAP node, the node MUST populate etc.".

6. "Determination of success by a receiver": do you mean 'successful 
processing as per SOAP's processing model', or that + "successfully 
received"? I suspect the former, but the current text is a little ambiguous.

7. "at either or all of": should this not be instead "at any of"? Or do 
you really mean that if any one of the multiple receivers fail, they all 
fail? Using a longer description would help.

8. At this point (last paragraph of 2.2), it would be worth indicating 
that multiple faults may be received in the case of errors at multiple 
receivers (and also link to Section 2.4).

9. What happens in the presence of intermediairies? Are they supported?

Section 2.3

10. The properties defined by the spec are very similar to the ones in 
the SOAP 1.2 spec. It would be interesting to reference SOAP 1.2/Part 
2/Table 3 and point out the differences.

11. "There may be other properties related to the operation of the 
message exchange and are processed according to their own feature 
specifications.": this is a little vague. Suggestion: "This MEP may be 
extended by later specifications. Such extensions can add additional 
properties to the ones listed above. The behaviour of the MEP will then 
be changed according to these specifications, when these properties are 
in effect."
Received on Thursday, 8 February 2007 17:51:46 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:17:01 UTC