Re: Rethinking ISSUE-12 with lang datatypes

On 27/05/11 19:43, Richard Cyganiak wrote:
> On 27 May 2011, at 11:33, Andy Seaborne wrote:
>> Why not declare a class rdf:String that is the features we want and leave rdf:PlainLiteral, with all it's datatype features that are being used, alone? (make it a subclass).
> ...
>> A reasonable expectation of users (whether technically right or not - people have intuitions about strings)
>>
>>     skos:prefLabel rdfs:range<datatype>
>>
>> is that they can write
>>
>>     <s>  skos:prefLabel "foo"^^<datatype>  .
>
> Huh?
>
> You still can't say
>
>     skos:prefLabel rdfs:range rdf:String .
>     <s>  skos:prefLabel "foo"^^rdf:String .
>
> unless you define a lexical space, and this would take us right back to the rdf:PlainLiteral "foo@en" mess.

rdf:String isn't a datatype.  It's a class only.
Only the base of the subclass hierarchies are datatypes: 
rdf:LangTaggedString and xsd:string

But rdf:PlainLiteral is a datatype already, which I see as odd.

> Also, I don't like having both xsd:string and rdf:String with different meaning.

What name would work for you?
My point about partial use of rdf:PlainLiteral still stands.
>
> Richard

	Andy

Received on Friday, 27 May 2011 20:44:16 UTC