Re: Proposal for resolving #171 - attributes on references

 Rich,
 we decided to propose ignoring rather than forbidding additional
attributes for the following reasons:
 1) smoother transition path from SOAP 1.1
 2) attributes like env:mustUnderstand and env:actor, which are
out-of-scope for Encoding, must be allowed to appear and be
effective on referencing accessors. These particular two
attributes will not be a good example if the proposed resolution
to #170 presented in [1], point 2, option b, passes. But Encoding
may be used in various situations where out-of-scope attributes
that do not affect the encoded data may appear.
 The proposal allows that. And yes, by referencing accessor I
mean b. 8-)
 Best regards,

                   Jacek Kopecky

                   Senior Architect, Systinet (formerly Idoox)
                   http://www.systinet.com/

[1] http://lists.w3.org/Archives/Public/xml-dist-app/2001Oct/0231.html



On Mon, 19 Nov 2001, Rich Salz wrote:

 > >  An example for illustration of the problem:
 > >         <a xsi:type="string" id="1"    >3</a>
 > >         <b xsi:type="int"    href="#1"/>
 > >  The problem is: what is the type of b?
 > >
 > >  The proposal is: "SOAP Encoding processors MUST ignore
 > > attributes on the referencing accessor except for the following
 > > list of attributes: href, enc:position, xml:base."
 >
 > The referencing accessor is b, right?  And the referred-to is a :)
 >
 > Can we reword it so that it says processors MUST NOT put any
 > attributes?  That way a receiver could send a fault back if it saw
 > one.    That seems more robust.
 > 	/r$
 >

Received on Monday, 19 November 2001 22:14:04 UTC