- From: Ivan Herman <ivan@w3.org>
- Date: Mon, 12 Sep 2011 12:18:44 +0200
- To: Gregg Kellogg <gregg@kellogg-assoc.com>
- Cc: RDFa Community <public-rdfa@w3.org>, W3C RDFWA WG <public-rdfa-wg@w3.org>
On Sep 7, 2011, at 18:05 , Gregg Kellogg wrote: > In my implementation, I added vocabulary caching for some well-known vocabularies. So did I. We are dangerously similar:-) > This seems to follow our earlier recommendations about caching popular profiles. I started with obvious candidates from prefixes in the default profile. Note that in some cases, the vocabulary URI does not result in an RDFS representation of the vocabulary itself. Presumably this will resolve itself over time if the vocabularies become popular, but schema.org is a notable exception: > > http://creativecommons.org/ns# (describes itself) > http://purl.org/dc/terms/ (describes itself) > http://usefulinc.com/ns/doap# (describes itself) > http://xmlns.com/foaf/0.1/ (describes itself, if given application/rdf+xml accept header) > http://purl.org/goodrelations/v1# (describes itself) > http://schema.org/ (does not describe itself, but even if it did, it wouldn't match http://schema.rdfs.org/all.ttl) > http://rdfs.org/sioc/ns# (describes itself) > http://www.w3.org/2004/02/skos/core# (describes itself) > http://www.w3.org/2008/05/skos-xl# (describes itself) > Yes, schema is the odd man out. But, actually, it is not really relevant. The schema.org does not map (eg, with subproperty relationships) to any other vocabulary, ie, it does not really provide any information for the vocabulary expansion (yet?) But caching is good nevertheless, to avoid accessing it all the time. > We should consider recommendations for caching well-known vocabularies and provide guidance for vocabulary authors, particularly that HTML versions of vocabularies SHOULD represent themselves using RDFa. > We did have some text for profiles in the earlier version of the document. I may have a look at that and see what can be moved to the vocabulary part of RDFa Core. Actually, my implementation can take in RDF/XML, Turtle, or RDFa. But yes, the RDFa encoding would be a good idea for documentation purposes anyway... Ivan > Gregg > > ---- Ivan Herman, W3C Semantic Web Activity Lead Home: http://www.w3.org/People/Ivan/ mobile: +31-641044153 PGP Key: http://www.ivan-herman.net/pgpkey.html FOAF: http://www.ivan-herman.net/foaf.rdf
Received on Monday, 12 September 2011 10:18:38 UTC