- From: Bullard, Claude L (Len) <clbullar@ingr.com>
- Date: Tue, 2 Dec 2003 13:38:35 -0600
- To: 'Walden Mathews' <waldenm@optonline.net>, "Ian B. Jacobs" <ij@w3.org>
- Cc: www-tag@w3.org
The confusion: one doesn't know what a resource is, so ambiguity creeps in right there, typically confusion of representation and resource. Is a person identified by a mailto a resource or a representation of a resource? To me, the answer is obvious. A person is neither a resource nor a representation. The use of the mailto in that example is colloquial. The architecture does not apply. A database is a resource if identified by a URI. A record in a database is a resource if a representation is available for that record whose identifier is a URI. A field in a database field is a resource if it can be identified by a URI (typically, plus a fragment). A value returned on query is a representation. len From: Walden Mathews [mailto:waldenm@optonline.net] If I had some examples, I'd offer them. But I'm actually still unclear on the issue. I thought "A URI identifies one resource" was axiomatic in the Web. And if that's the case, then the architecture rules out "URI ambiguity", and this Best Practice ends up redundant and/or vacuous. I need some examples to see why the above is not simply the case.
Received on Tuesday, 2 December 2003 14:38:37 UTC