- From: Nathan <nathan@webr3.org>
- Date: Tue, 12 Oct 2010 09:31:24 +0100
- To: Ivan Herman <ivan@w3.org>
- CC: RDFA Working Group <public-rdfa-wg@w3.org>
Hi Ivan, Cool, I can agree with that as it opens up a whole can of worms, if you make access to Collections easier / custom, then why not the same for BlankNode, and so on and so forth, after implementing it's certainly easier just treating the store as a graph of triples, similar to N-Triples. Will give some feedback specific to your RDFa Collections proposal under separate cover. Many Thanks, Nathan Ivan Herman wrote: > Nathan, > > This is, again, a typical '3rd box' feature, if any. Ie, it should be part of the possible future extensions of a RDF API, but certainly not part of the core RDFa API. In my personal opinion we should not spend energy on that one for now (regardless of the outcome of ISSUE-16), but flag it as a possible future item for RDF API > > Ivan > > On Oct 11, 2010, at 21:10 , Nathan wrote: > >> Hi All, >> >> I've just (about) completed a Turtle parser which has brought the issue of RDF Collections, specifically Lists, to the forefront of my attention. >> >> Currently I've resorted to taking the N-Triples approach adding the appropriate series of rdf:first, rdf:rest, rdf:nil to the store - however I'm very aware that in no way do we mark Collection as being a specific type of RDF Interface, nor do we provide any way to work with Collections/Lists. >> >> So, in terms of the API, do we want to ignore that Collections even exist, or do we want to specifically address them by adding an RDFCollection Interface? (perhaps extending BlankNode and containing a sequence of RDFNode, the list elements) >> >> Somewhat related to ISSUE-16 . >> >> Best, >> >> Nathan > > > ---- > 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 Tuesday, 12 October 2010 08:32:18 UTC