[rdfweb-dev] Re: FOAF 0.1: Organized, Explained, Collated, Part One

Dave Beckett dave.beckett at b...
Mon Jul 15 13:32:00 UTC 2002

>>>Aaron Swartz said:
> It doesn't work because the RDF/XML syntax was designed by a committee. 

A working group. And you are on the successor Aaron ;)

> I agree that something like this should be allowed. Since this will 
> never happen, you will have to use the more verbose:
> <foaf:nick foaf:domain="http://www.icq.com/" rdf:value="2927491" />

Despite what it looks like here, the values of attributes in RDF/XML
are always literals, not URIs, so that gives, in N-Triples:

_:a <http://xmlns.com/foaf/0.1/nick> _:b .
_:b <http://xmlns.com/foaf/0.1/domain> "http://www.icq.com/" .
_:b <http://www.w3.org/1999/02/22-rdf-syntax-ns#value> "2927491" .


_:a <http://xmlns.com/foaf/0.1/nick> _:b .
_:b <http://xmlns.com/foaf/0.1/domain> <http://www.icq.com/> .
_:b <http://www.w3.org/1999/02/22-rdf-syntax-ns#value> "2927491" .

I don't know the actual reasoning why, but basically you can't
annotate an XML attribute to distingish it without an extra
schema of some sort (DTD, W3C XML Schema, ...)


More information about the foaf-dev mailing list