XML vs. RDF ==> see infoset domains

I was in a meeting the other day where TimBL was trying to explain the
difference between RDF and XML.  He said something vaguely like "XML
only lets you say things about parts of documents; RDF lets you say
things about anything."

I though that was an interesting angle.  It led me to this little
command line:

$ GET http://www.w3.org/2001/04/infoset | grep rdfs:domain | sort | uniq -c
      8   <rdfs:domain resource="#Attribute"/>
      3   <rdfs:domain resource="#Character"/>
      2   <rdfs:domain resource="#Comment"/>
      9   <rdfs:domain resource="#Document"/>
      4   <rdfs:domain resource="#DocumentTypeDeclaration"/>
      9   <rdfs:domain resource="#Element"/>
      2   <rdfs:domain resource="#Namespace"/>
      4   <rdfs:domain resource="#Notation"/>
      4   <rdfs:domain resource="#ProcessingInstruction"/>
      5   <rdfs:domain resource="#UnexpandedEntityReference"/>
      6   <rdfs:domain resource="#UnparsedEntity"/>

which I think supports the argument rather nicely.

Is anyone doing XML processing by turning XML instances into infoset
RDF graphs, processing the RDF graphs?  I'm not sure when/where/if all
that extra processing would pay off.

    -- sandro