- From: <noah_mendelsohn@us.ibm.com>
- Date: Fri, 30 Sep 2005 10:24:43 -0400
- To: Tim Berners-Lee <timbl@w3.org>
- Cc: Dan Connolly <connolly@w3.org>, W3C TAG <www-tag@w3.org>
I asked earlier in this thread whether there was a reason that "#" should be preferred to "/" for members of namespaces in particular. Tim Berners-Lee responded: > Er... without any deployed representations what does / have over #? > And when you deploy stuff, the # builds on very basic web > architecture which has been tested for ages and the / builds on HTML > redirects which have just been dreamed up by the TAG and not > seriously deployed at all yet as far as I know. I thought we agreed that namespaces can be information resources, and therefore can appropriately return representations with status code 200. So, I wasn't intending to talk about the 302 redirect case. Rather, I was presuming that, at least in the '/' case, there were representations both for the namespace as a whole and for the individual members of the namespace, each served with status code 200. So, N+1 representations served from N+1 URIs. Whatever the other pros and cons, I don't think this depends on the admittedly new proposal to use redirects for non-information resources. Again, I'm not expressing a preference for "/" over "#", just asking whether there's any reason why "/" should be prohibited or strongly discouraged in the case of namespaces. There may be such a reason, but I'm not convinced that a need for redirects is it. Noah -------------------------------------- Noah Mendelsohn IBM Corporation One Rogers Street Cambridge, MA 02142 1-617-693-4036 --------------------------------------
Received on Friday, 30 September 2005 14:25:48 UTC