RE: "do not occur"



> -----Original Message-----
> From: Alan Ruttenberg [mailto:alanruttenberg@gmail.com]
> Sent: 3 June 2009 15:52
> To: Seaborne, Andy
> Cc: Axel Polleres; Peter F.Patel-Schneider; sandro@w3.org; public-rdf-
> text@w3.org
> Subject: Re: "do not occur"
> 
> On Wed, Jun 3, 2009 at 6:05 AM, Seaborne, Andy <andy.seaborne@hp.com>
> wrote:
> >> Incidentally, the fact that you can filter using the DATATYPE
> function
> >> in sparql is another hint that something is amiss. By my earlier
> >> analysis, the DATATYPE function should never return rdf:PlainLiteral,
> >> according to our spec.
> >>
> >> -Alan
> >
> > Quite - only if the new text is in force and then only if suitable D-
> entailment is being applied or the data path goes through OWL2.
> 
> Are you saying that DATATYPE *could* return rdf:PlainLiteral under
> these circumstances? By my analysis it couldn't, even then.
> -Alan

So far as I can see, yes. It's just the "anyone can write it in RDF now" point.

As things stand today, the rdf:PlainLiteral doc does not apply - it's not REC.

DATATYPE("anything@"^^rdf:PlainLiteral) is rdf:PlainLiteral.  

IRI   DATATYPE (typed literal typedLit)
IRI   DATATYPE (simple literal simpleLit)

It's a typed literal.  I have no idea what it means or whether it's consistent WRT it's own entailments but that does not matter under simple entailment.  

The fact that little if any data (except this mailing list) exists means I see no problem with this nor do I worry about it.  I do worry about (RDF) data that starts out as traditional plain literals but may end up as typed literals by some means.  That will carry forward.

You could appeal to messing around with the rdf: namespace and ownership thereof but the reality is that data does mess with namespace and ownership all the time.

 Andy

Received on Wednesday, 3 June 2009 15:05:39 UTC