- From: Mark Nottingham <mnot@akamai.com>
- Date: Mon, 2 Oct 2000 15:25:01 -0700
- To: David Wagner <dwagner@sa.kevric.com>
- Cc: "'RDF Interest (E-mail)'" <www-rdf-interest@w3.org>
On Mon, Oct 02, 2000 at 04:29:33PM -0500, David Wagner wrote: [...] > I have considered two possibilities for what I am implementing. > Following the aboutEachPrefix model I have considered adding many > attributes including these. > > aboutEachRegEx > aboutEachDomain > aboutEachHost > aboutEachMIMEtype > > This list may get mighty long. Another more promising possibility is > matching on URI syntactic components after tokenizing along the lines > of of http://www.isi.edu/in-notes/rfc2396.txt, 3. URI Syntactic > Components. This is precisely what URISpace does. The external selector hook means that application-specific things like MIME type can be accommodated without trouble. I've been debating whether it should just be a generalized XML rule matching mechanism, or if the link with URIs should be stronger (as it is currently). I'm inclined to make it strong, as URIs are, by definition, universal, while other criteria like Content-Type are often specific to a particular application scope (even if standardized, once again like Content-Type). > I agree there is a much bigger picture and there are many people still > hashing this out (the developers of Apache's Cocoon project just > developed their own URI matching syntax) even as they implement it. Do you have a reference for this? -- Mark Nottingham, Research Scientist Akamai Technologies (San Mateo, CA)
Received on Monday, 2 October 2000 18:25:41 UTC