- From: Giosue Vitaglione <giosue@umich.edu>
- Date: Mon, 9 Apr 2001 09:52:15 -0400 (EDT)
- To: "Dan Connolly" <connolly@w3.org>
- Cc: <www-rdf-interest@w3.org>, <www-rdf-logic@w3.org>
> Huh? Each URI identifies exactly one resource. > (URL is an informal synonym for URI, no?) > I want to believe that here there are some ironies. Sorry if I enter in your discussion. Please let me give my two cents. Of course the last letter in the two acronyms (URL, URI) has a different meaning, Locator and Identifier. For something that never moves, let's say a building, these two concepts are equivalent. Can you identify a car with a locator ? Not, of course. You need an identifier (plate number) and maybe if you are looking for that car, you need a huge database with a the couple (ID, Position) contantly updated. For the web, can you use URL as identifiers ? We all did that way, so far. There are some clear issues in doing that, because documents move, and domain name can be transferred. If we could assign wisely the URL strings, could we use them as identifiers ? The big limitation is the host name included in the URL. Any way to use URL as identifier is as good as PURL (www.purl.org) and cannot be much better than that. Best Regards, Giosue' Vitaglione
Received on Wednesday, 11 April 2001 07:34:24 UTC