W3C home > Mailing lists > Public > public-rdf-wg@w3.org > November 2012

Re: B-scopes

From: Richard Cyganiak <richard@cyganiak.de>
Date: Thu, 15 Nov 2012 18:59:24 +0000
Cc: RDF Working Group WG <public-rdf-wg@w3.org>
Message-Id: <2B3E55CC-866B-401E-BC8B-FF4DB484FDF1@cyganiak.de>
To: nathan@webr3.org
On 14 Nov 2012, at 11:58, Nathan wrote:
> It would be really, really, nice if a named-g-box could act as a b-scope.

Well, in a SPARQL graph store, you can move blank nodes between named-g-boxes, so either you have to say that the entire store is within the same b-scope, or allow blank nodes from different b-scopes in the same g-box (which is unpleasant from an implementation point of view).

In your own named-g-box specs or implementation, you can of course outlaw this and manage scopes in whatever way you like, assuming you don't need to be compatible with SPARQL Update.

The more important thing IMO is that if you serialize a snapshot of the named-g-box as a g-text, then that g-text is its own scope, so you get a fresh set of blank nodes, and it is clear that you don't need to worry about how the blank node identifiers in the g-text relate back to blank nodes in the g-box.

> Regardless, I like this proposal, it adds clarity to b-nodes and pre answers many bnode related faqs, kudos.

Thanks.

Richard
Received on Thursday, 15 November 2012 18:59:52 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:52 GMT