- From: Graham Klyne <gk@ninebynine.org>
- Date: Fri, 11 Apr 2003 12:54:10 +0100
- To: w3c-rdfcore-wg@w3.org, Jeremy Carroll <jjc@hpl.hp.com>
Responding to: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Apr/0190.html which hasn't yet made it to my mailbox. >Jeremy: > >>The LC text does not claim that the RDF representation is equivalent, > merely > >>that it is an expression of, an n-place predicate or a n-column table. >??? > >>The primer does not deal with this problem, which is why I thought we had > >>this text in the first place. > >Graham: > >I'm not sure, now, what "this problem" is, that is not covered by the > >proposed revised text. > >As I saw it the LC text was intended to indicate one way to migrate legacy >relational data into RDF. I had considered proposing to cut this text >completely when I reworked the example a couple of releases ago. I decided >against on the basis that relational data is the most important type of >legacy data, yet the topic is slightly beyond the primer treatment. Nothing >Peter says seems relevant to that judgement. > >While I am not particularly attached to this baby see your proposed >changes as >throwing it out with the bathwater. Well, I surely don't think this section is trying to promote a "migration strategy". I think it is appropriate here to indicate that the simple RDF binary predicate form can be used to express the same information (if not the exact same facts) that may be contained in a relational database, which I think the text does. I remain satisfied with the currenbt text. #g ------------------- Graham Klyne <GK@NineByNine.org> PGP: 0FAA 69FF C083 000B A2E9 A131 01B9 1C7A DBCA CB5E
Received on Friday, 11 April 2003 09:33:21 UTC