- From: Laurian Gridinoc <laur@gd.ro>
- Date: Thu, 26 Aug 2004 09:27:05 +0000
- To: Eric Jain <Eric.Jain@isb-sib.ch>
- CC: Lars Marius Garshol <larsga@ontopia.net>, rdf-interest <www-rdf-interest@w3.org>
Hello, I thought that not using contexts was a good point, because I can make or verify statements out of any context... Of course, reification is redundant because there is no way to assign an URI to a triple unless the underlying storage supports this, but out of the storage context won't work. Eric Jain wrote: > [...] > It would indeed be nice if there was a way to avoid global identifiers > for statements, but all syntaxes that I have seen so far that do so > end up being rather verbose, i.e. using even more disk space than > standard RDF/XML (unless perhaps you need to reify every single > statement). Maybe one solution (global identifiers), would be to have an URI schema for triples, something like urn:rdf:sha1:BASE32ofSHA1SumOfNTriple to address a statement in an uniform way (decoupled of any storage internal identifiers). I know it's wrong, but what if? Cheers, -- Laurian Gridinoc Chief Developer GRAPEFRUIT DESIGN mobile: +40.724.363656 e-mail: laur@gd.ro www.grapefruitdesign.com www.gd.ro
Received on Thursday, 26 August 2004 09:27:04 UTC