- From: Dan Brickley <danbri@w3.org>
- Date: Thu, 23 May 2002 17:24:48 -0400 (EDT)
- To: Aaron Swartz <me@aaronsw.com>
- cc: "Seaborne, Andy" <Andy_Seaborne@hplb.hpl.hp.com>, <www-rdf-interest@w3.org>
On Thu, 23 May 2002, Aaron Swartz wrote: > On Thursday, May 23, 2002, at 07:36 AM, Seaborne, Andy wrote: > > > A precursor to better modelling is more bNodes - and a general > > enthusiasm to > > use them. I think people shy away from them at present which hurts data > > integration (amongst other things). > > Could you elaborate? I've always found bNodes a bad idea, since, among > other things, you can't refer to them and so I strongly recommend > against them. A bNode is a sub-part of an RDF description. Wanting to refer to the bNode (as against the thing it represents) is like wanting to refer to the " or ' characters around some XML attribute. I thought this thread had been done to death. It seems not :( Can't we talk about code and implementations and test suites instead? Dan -- mailto:danbri@w3.org http://www.w3.org/People/DanBri/
Received on Thursday, 23 May 2002 17:25:50 UTC