Re: Section 5 vs Schema

 Gudge,
 I think we have four options for answering your question:

 1) Stick with schema, in which case I think we might want to
define a normative binding of our implied data schema language
(for describing structures in our data model) into XML Schema 
(and IMHO this would be helpful for instance in WSDL).
 2) Remove schema, use some well-known programming language like 
C to describe the structures, in which case I think we might want 
to define a normative binding of our implied data schema language 
into that programming language (but I think it is unnecessary).
 3) Remove schema, define our (so far implied) data schema 
language in XML.
 4) Remove any data schemas from our examples and define the data 
in narrative.

 The option 3 would IMHO be the best solution in terms of clarity
of the spec, completeness of the Encoding/Data model spec tuple,
it would be useful in the RPC section, it could be used in WSDL
etc.
 On the other hand it would be some work and YAXL (yet another 
XML language) etc.

 For practical reasons I prefer option 2 without any normative
binding, closely followed by 4. Remember, this is only an issue
with the examples of encoded data, not with the normative parts
of the Encoding section (assuming of course that examples are
non-normative).
 We might move the examples into the Primer, but then the 
Encoding section would be left example-less.

 Best regards,

                   Jacek Kopecky

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



On Sat, 26 Jan 2002, Martin Gudgin wrote:

 > Interesting though this discussion is, it is not really addressing the issue
 > against the spec[1]. The discussion I tried to start with[2] was what to do
 > with the schemas in Part II Section 4 ( was Section 5 ). It is confusing to
 > people that Section 5, which is about encoding things when you don't start
 > from a schema then has schemas for all the examples.
 > 
 > Note, I don't want to stop this interesting discussion, but I would like to
 > get an answer to the question I was trying to pose.
 > 
 > Gudge
 > 
 > [1] http://www.w3.org/2000/xp/Group/xmlp-issues.html#x17
 > [2] http://lists.w3.org/Archives/Public/xml-dist-app/2001Dec/0019.html
 >

Received on Sunday, 27 January 2002 13:56:27 UTC