[OAI-implementers] Qualified Dublin Core

[Pete Johnston]

Rachel said:

> I have always liked the approach taken in PRISM spec:
> 
> "PRISM-compliant applications MUST NOT throw an error if they 
> encounter unknown elements or attributes. They are free to 
> delete or preserve such information, although recommended 
> practice is to retain them and pass them along. Retaining the 
> information is an architectural principle which helps new 
> functionality be established in the presence of older 
> versions of software." see 
> http://www.prismstandard.org/Pam_1.0/PRISM_1.2h.pdf

Hmmmmm, but PRISM uses RDF/XML for its serialisation. It seems to me the
couple of sentences previous to the above are crucial:

====
PRISM descriptions are compliant with the RDF constraints on the XML
syntax. Thus, they begin with the
rdf:RDF element. Because PRISM obeys the RDF constraints on XML
structure, implementations are
guaranteed to correctly parse even unknown elements and attributes. 
====

I tend to agree with Jeff:

> My feeling is that if someone wants to mix DCQ with other 
> stuff, the container deserves its own namespace. Otherwise, 
> there is no standard way to recognize the significance of the 
> mix when the record escapes the confines of its peculiar context.

The provision of a container schema for a "qualified DC" metadata format
which (via some XSD wildcard wizardy) permits arbitrary XML child
elements (N.B. XML elements not RDF properties), just as long as one
child element associated with a DC XML namespace is present, implies (to
me, at least) that there is some predictable interpretation (in terms of
the DC property/value/encoding-scheme principles) of what may in
practice turn out to be a completely adhoc XML sub-tree structure.

Pete