RE: [xml-dev] XML information modeling best practices

One can say that the XML HAS-A (the element can/shall 
contain attributes or elements of this name) is not 
the same semantic as what RDF means by HAS-A unless 
the application makes this so.  XML Doesn't Care.

Also as I think Manos will agree and Jonathan points out, 
RDF is demonstrably useful for declaring these semantics 
and can be used to annotate/document the intent such that 
a private agreement becomes formal and can then be made 
public in a processable form.  So the application implementation 
must conform to semantics stated in different languages. 
That isn't too terrible until one starts trying to do 
this without something like RDDL to state which document 
pertains to what type of rules.

Argggg.  Xo much for wall-to-wall XML.  It's Infoset 
of A related to Infoset of B related to....  Xo much for 
simple.  Modeling practices will be hard to come by 
even though in practice, it may be that RDF design 
could precede the XML design and keep it on track.  
There's a weird idea:  RDF initialization of XML 
editors/software.

The SemWeb can strand on the MoreMetaThanThou layers of 
abstractions that have to be completely understood 
before a single bit changes state:  Dragons everywhere 
and not a Maiden to be had.

len

-----Original Message-----
From: Robin Cover [mailto:]

Len's clearer statement about XML's fundamental
indifference to HAS-A and IS-A is adequate; his
clarification that agreement is outside the scope
of XML (and must be made explicit by humans) is
likewise to the point.