Re: Proposed resolution on using schemas to default itemType and nodeClass (subissue of 231)

noah_mendelsohn@us.ibm.com wrote:

>This note is in fulfillment of an action that I took on todays WG call.
>  
>
Thanks.

[...]

>Does this seem like an acceptable approach?  I think it is appropriately
>symmetric with the way we've been applying schema to other aspects of the
>encoding, and it provides a standard interpretation for attribute defaults
>in situations where schema validation is desired.  Thanks.
>  
>
I accept this proposal.  It does have the desired overall effect of 
unifying the self-describing approaches with the schema-describing 
approaches, especially in this newly-proposed device for array detection.

At first I had a bit of trouble understanding how one part of the 
specification says "The values associated with element and attribute 
information items defined in this specification MUST be carried 
explicitly in the transmitted SOAP message except where stated 
otherwise" and the other section says "the values of such defaulted 
attributes affect the deserialized graph in the same manner as if the 
attributes had been explicitly supplied in the message".  Using the 
schema information not only associates type information, but it also 
enhances the class (array/struct/simple/size) information of nodes in 
the graph, in apparent violation of the spirit, if not the rules of the 
normative section.  You get a significantly different graph if you use 
the schema, which from the beginning has been a valid usage mode for the 
SOAP encoding -- the non-self-describing encoding that relies on schema 
to fill in all the type and class details since clients and servers all 
look at the same WSDL file containing the relevant schema definitions.

For some this might appear to be a simple contradiction within the spec 
with a number of arguments possible for justifying or disregarding the 
non-normative section.  If there were a simple statement that could be 
added in the non-normative section justifying it against the normative 
statement, it would be nice, but I failed to produce one.

Also:  Is it enough in the appendix to refer to the attributes by local 
names, or should we add " of the encoding namespace" after we name them?

Thanks,

Ray Whitmer
rayw@netscape.com

Received on Thursday, 26 September 2002 07:34:15 UTC