RE: AFTF requirements, pre-2003/01/31 telcon

Gudge wrote on 02/05/2003 08:00:02 AM:

> 
> 
> 
> > -----Original Message-----
> > From: John J. Barton [mailto:John_Barton@hpl.hp.com] 
> > Sent: 04 February 2003 17:39
> > To: Martin Gudgin; noah_mendelsohn@us.ibm.com
> > Cc: jones@research.att.com; Rich Salz; Sanjiva Weerawarana; 
> > xml-dist-app@w3.org
> > Subject: RE: AFTF requirements, pre-2003/01/31 telcon
> > 
> > 
> > Thanks, I think that these comments make your requirement 
> > clearer.  It seems to me that what you want is for all the 
> > data in a package to be "reachable" from the SOAP message by 
> > XML defined means.  I think this is a reasonable requirement 
> > but not one that can be formulated with the W3C infoset definition.
> 
> Oh, I think that qualified attributes would work just fine ( XInclude
> uses qualified elements after all ).
> 
> > 
> > We have an object--the SOAP XML--with some pointers inside. 
> > Some of the pointers will resolve to entities in the package; 
> > some will not.  As far as I can tell the infoset, being 
> > designed without consideration of our case, treats all 
> > pointers the same and it considers all referenced entities to 
> > be outside of the infoset. 
> 
> The Infoset doesn't say anything about references. It knows nothing
> about URIs. Therefore it knows nothing about 'referenced entities', not
> even the fact that they exist.
> 
> > This is the only sensible answer 
> > for normal hypertext documents. What you seem to want is a 
> > "package-set" that has the properties of an infoset extended 
> > to the pointers that resolve to within the package of data 
> > sent with the SOAP.
> 
> Yes, I want some or all of the 'referenced entities' to actually appear
> at the Infoset level. The infoset still doesn't know they were
> 'referenced entities'. As far as the infoset is concerned the data is
> just inline.

Back to the question I raised previously; how does an infoset "in line"
a document II within another document II??? I don't think that loosing its
"documentness" is necessarily such a good thing. Sometimes, you want it
to be a document, not just a collection of EIIs.

I'm still quite skeptical of this.

<snip/>

Cheers,

Christopher Ferris
Architect, Emerging e-business Industry Architecture
email: chrisfer@us.ibm.com
phone: +1 508 234 3624