- From: Martin Hepp <martin.hepp@ebusiness-unibw.org>
- Date: Thu, 13 Oct 2011 09:31:03 +0200
- To: Gregg Kellogg <gregg@kellogg-assoc.com>
- Cc: Bob Ferris <zazi@smiy.org>, "public-vocabs@w3.org" <public-vocabs@w3.org>
Hi Gregg, all: > > We discussed doing datatype entailment in RDFa, but the problem is that we can't require that all parsers would do this, due to problems accessing external resources. Better to lobby for an @itemdatatype, which has been suggested. This would have to go through Hixie, but please send use cases to the HTML data TF. Actually, I think that the Microdata approach of taking the datatype from the schema and not from the data is much better than the RDFa approach of repeating the datatype over and over again - why should millions of triples repeat that e.g. gr:hasCurrencyValue is xsd:float? So I cannot support @itemdatatype in the MD spec ... The main reason for my above suggestion is that IMO, a big chance for saving the work of the SW community is in providing an environment for *consuming* structured data in semi-controlled environments. At the publishing site, I think the SW movement is dead, simply because the impact on the actual Web developers' communities is minimal, for a variety of reasons. But for consuming the data, there is some hope that RDF and SPARQL are really useful for innovative developments. So the need for the above is solely to be backwards-compatible with older RDF work. Martin
Received on Thursday, 13 October 2011 07:31:38 UTC