On 1/4/12 2:35 PM, Pete Johnston wrote:
> Hi Antoine,
>
>> If you're not bothered by my rumbling on all this, I'd happily give some
>> feedback (please tell me if I should stop ;-) ) . But there is one thing that
>> prevents me from getting my head around the problem: why was DC-Text
>> making a distinction between ValueStrings in the first place
>> (LiteralValueString, ValueString)?
>> Reading http://dublincore.org/documents/dc-text/ it seems that the two are
>> associated to the same production rules, i.e., they lead to the same kind of
>> (RDF) literals. The fact that this literal is appearing in a non-literal value
>> surrogate or a literal value surrogate apparently does not impact what the
>> literal may be...
>
> The literal itself doesn't change, but its "position" in the RDF graph is different for the two cases. i.e. the production rules are different. See Mikael's document:
>
> http://dublincore.org/documents/dc-rdf/
>
> First, the "ValueString" syntax, representing the "value string in non-literal value surrogate" (!!) case (section 4.3):
>
> The DC-Text description set (a stripped down version of the example in section 4.3)
>
> @prefix dcterms:<http://purl.org/dc/terms/> .
>
> DescriptionSet (
> Description (
> ResourceURI (<http://example.org/123> )
> Statement (
> PropertyURI ( dcterms:subject )
> ValueString ( "Biology"
> Language ( "en" )
> )
> )
> )
> )
>
> corresponds to the following RDF graph (in Turtle):
>
> @prefix dcterms:<http://purl.org/dc/terms/> .
>
> <http://example.org/123> dcterms:subject [ rdf:value "Biology"@en] .
>
> i.e. two triples, with introduction of blank node, rdf:value predicate in 2nd triple.
>
> Second, the "LiteralValueString" syntax, representing the "value string in literal value surrogate" case (section 4.7):
>
> The DC-Text description set (the example in section 4.7)
>
> @prefix dcterms:<http://purl.org/dc/terms/> .
>
> DescriptionSet (
> Description (
> ResourceURI (<http://example.org/123> )
> Statement (
> PropertyURI ( dcterms:title )
> LiteralValueString ( "Learning Biology"
> Language ( "en" )
> )
> )
> )
> )
>
> corresponds to the following RDF graph (in Turtle):
>
> @prefix dcterms:<http://purl.org/dc/terms/> .
>
> <http://example.org/123> dcterms:title "Learning Biology"@en .
>
> i.e. single triple with literal object.
>
> Pete
Thanks, Pete! I think it is clearer to me, now.
Though it may look weirder, in a way. The production rules are the same for DC-Text, aren't they? I is just the mapping to RDF that changes.
In that case, I would have rather expected the distinction to be made strongly in DCAM (where is matters, if one wants a good alignment with RDF there) and hidden in DC-Text (where it does not really matter, unless DC-Text is meant as a complete implementation of DCAM and all the distinctions it could make).
Cheers,
Antoine
|