- From: Jeremy Carroll <jeremy@topquadrant.com>
- Date: Thu, 21 May 2009 19:05:33 -0700
- To: "'Sandro Hawke'" <sandro@w3.org>, "'Harry Halpin'" <hhalpin@ibiblio.org>
- Cc: "'Pat Hayes'" <phayes@ihmc.us>, "'Boris Motik'" <boris.motik@comlab.ox.ac.uk>, "'Eric Prud'hommeaux'" <eric@w3.org>, "'Andy Seaborne'" <andy.seaborne@hp.com>, "'Alan Ruttenberg'" <alanruttenberg@gmail.com>, <public-rdf-text@w3.org>, "'Semantic Web'" <semantic-web@w3.org>, "'Axel Polleres'" <axel.polleres@deri.org>
> > OK, so my reading of rdf:text was not mistaken then. > > > > Yes, I was very surprised when I read the rdf:text draft that this > > wasn't actually what was done, I just sort of assumed it was, since > it > > seems rather sensible, allowing RIF and OWL2 to be used better over > > RDF data. > > I haven't really been able to follow what you've been saying, but, yes, > of course RIF and OWL2 can (in all designs considerd) be used over RDF > data with plain literals (and the plain literals are interpreted as > rdf:text literals). If the specs were ambiguous about this, I'm sure > it's because it never occured to us that someone might imagine it > otherwise. > > -- Sandro A further issue is whether rdf:text can occur as a datatype in an RDF/XML document. If it can, then there is a further syntactic form semantically equivalent to a plain literal. If rdf:text is explicitly prohibited from being in an RDF/XML document then it forces the users of rdf:text to be aware of the required serialization as a plain literal, and ensures backward compatibility Jeremy
Received on Friday, 22 May 2009 02:06:24 UTC