SOAP 1.2 Last Call comments

Primer
-------
http://www.w3.org/TR/2002/WD-soap12-part0-20020626

2.4 SOAP Processing Model:

(this section would better with more subsections)
[...]

Example 7b
[...]
     <r:aThirdBlock xmlns:r="http://example.com">
     ::::
     </r:aThirdBlock>
[...]
In Example 7b, the ultimate recipient of the message - the SOAP
processor which plays the "ultimatereceiver" role - must process both
the Body as well as the header block aThirdBlock. 
[...]
          Node intermediary ultimate receiver
   mustUnderstand     
   "true"         must process   must process
   "false"        may process     may process
   absent         may process     may process
[...]

Does the ultimate receiver must process the block aThirdBlock as
described in the text or may process it, as described in the table?

Part 1
http://www.w3.org/TR/2002/WD-soap12-part1-20020626

General comment: as a user, this specification is difficult to read. The
Infoset-like notation would be better with a BNF notation. I wish
the editors didn't follow the example of the XML Schema specification
before rewriting the SOAP spec. The style used plays against the
technology described in the specification imho.

It is unfortunate that SOAP and WSDL are not used the same conventions
for markup. SOAP uses capitalized-word and WSDL lowercases. I noticed
that SOAP Part 1 contains several occurences of "mustUnderstand" and
"MustUnderstand" btw.

Section 5.4:
How does an intermediary do when receiving a fault? Is it guarantee to
be forwarded to the original Sender (or previous intermediary) if known?

Section 5.4.7:

The envelope (lowercase, without namespace name) element is used to
indicate a mismatch SOAP version in the Envelope (capitalized, with
namespace name) element. How does a SOAP Node know the SOAP version to
use?

Thank you,
Philippe

Received on Monday, 15 July 2002 10:58:45 UTC