Re: A comment on rdfms-reification-required

>>>Brian McBride said:
> I received the following comment on the proposed resolution of
> rdfms-reification -required from Wolfram: 

<snip/>


> In the light of this comment, I've modified the proposed resolution:
> 
> Proposal:
>     The WG resolves that a parser is not required to always emit statements
>     representing a bag of reified statements for each Description
>     element.  A parser SHOULD emit statements describing a bag of reifed
>     statements when an rdf:bagID is present in the description element.
>     It SHOULD be possible to instruct a parser to emit bags of
>     reified statements for all description elements.


No.  This is requiring a processor has an option in handling RDF/XML,
and I don't want to start doing that without a good reason (this is
the first time optional RDF/XML processing has come up).

  Proposal:
     The WG resolves that a parser MUST emit statements describing a
     bag of reifed statements when an rdf:bagID attribute is present
     on the rdf:Description (or typed node) element.  It MUST NOT
     emit bags of reified statements for rdf:Description (or typed
     node) elements without rdf:bagID attributes.

I don't think we can decide this tomorrow since there are no test
cases.  However, the current draft description for this bit of syntax
  http://ilrt.org/discovery/2001/07/rdf-syntax-grammar/#nodeElement
describes the triple-production in the above way.

Dave