Re: datatype()

On Fri, Jul 01, 2005 at 06:53:23PM +0100, Seaborne, Andy wrote:
> 
> Apologies for the fine-grainedness.
> 
> While implementing rq23, I came across:
> 
> DATATYPE(A)  	rdf:literal	N/A	sop:datatype(A)	rdf:uri
> 
> 
> What's the equality operator on rdf:uri?

The only one I've been using is '=' (sop:RDFterm-equal(A, B)).

> How about making datatype() return an RDF term? Or making rdf:uri as 
> subtype of
> RDF term, and then the equality test is the RDF term equality?  It makes 
> the use
> of <http://foo> in a filter like use in a pattern.

I think you're suggesting that we we get to re-use sop:RDFterm-equal,
which I believe we already use (by virtue of using the '=' operator).
Wouldn't returning in RDF term allowed literal or bNode datatypes?

(I picked URI over string so we could use qnames, but that's probably
obvious.)

> [Aside: rdf:bNode is only mentioned once.]

It's used in the list of SPARQL argument/result types. At a glance, it
seemed less confusing to have a term without spaces, and bNlode seemed
like a good choice. Do you think "SOP:Blank Node" or "SOP:Blank_Node"
is better?
-- 
-eric

office: +81.466.49.1170 W3C, Keio Research Institute at SFC,
                        Shonan Fujisawa Campus, Keio University,
                        5322 Endo, Fujisawa, Kanagawa 252-8520
                        JAPAN
        +1.617.258.5741 NE43-344, MIT, Cambridge, MA 02144 USA
cell:   +81.90.6533.3882

(eric@w3.org)
Feel free to forward this message to any list for any purpose other than
email address distribution.