Re: [regrep] Core Components and XML Serialization - WhichApproach?

Thanks David. What do you believe are the advantages and disadvantages
of the hypothetical example I posted?

Joe

David RR Webber - XML ebusiness wrote:
> 
> Joe,
> 
> I've created a ZIP file of the CRI work carried out under CEFACT.
> 
> The ZIP contains the draft spec', sample schemas and XML
> documents.
> 
> The download link is here:
> 
>  http://cam.swiki.net/11
> 
> I believe that this work can be simplified and edited down
> to give a first draft "CRI for Registry" - particularly now that
> the CCTS is in a final form.
> 
> I snipped the following out of the document as I think this is
> instructive (see bottom).
> 
> I'd suggest that we first tackle the conceptual core component
> representation, then move to the logical and physical layers.
> 
> Obviously this early CRI work led to the work on CAM, so
> there are several pieces in the old CRI that make it
> significantly more complex - because it was having
> to do both jobs - (and yes - it used recursion to capture
> the assembly details).
> 
> Anyway - I will undertake to post a revised OASIS CRI draft
> for Monday next week - that is a first brush on taking the
> earlier work and updating it for March 2003, instead of
> Nov, 2001.
> 
> This should then give the team a logical startpoint that
> incorporates the earlier work - and should shortcut alot
> of the simpler basic questions that were starting to crop
> up here.
> 
> Thanks, DW.
> 
> <snip>
> 
> Each item is annotated to describe the specific individual function.
> It should be noted that the schema is designed to fulfill a
> flexible set of uses, including:
> 
> ·       Logical core components
> ·       Physical component details
> ·       Assembly documents
> ·       Compound items
> ·       Atomic nouns
> ·       Atomic and compound parts of nouns
> ·       Process Verbs
> ·       Reference tables
> 
> As a guide for implementers, users should be encouraged to first
> create assembly document references based off their existing DTD,
> Schema or EDI transaction structures.  From this naturally flows
> the associated component details, and then above those the logical
> core component models can be derived.  The associated noun and
> part details can then be derived along with verb and reference
> table details.
> 
> <snip>