- From: Dan Connolly <connolly@w3.org>
- Date: Mon, 01 Oct 2007 07:47:17 -0500
- To: Alan Ruttenberg <alanruttenberg@gmail.com>
- Cc: Richard Cyganiak <richard@cyganiak.de>, "Henry S. Thompson" <ht@inf.ed.ac.uk>, www-tag <www-tag@w3.org>
On Sun, 2007-09-30 at 01:47 -0400, Alan Ruttenberg wrote: > On Sep 28, 2007, at 6:29 PM, Dan Connolly wrote: > > > I'd really rather not put 303 redirect configuration in the > > critical path to deployment of garden variety Semantic Web data. > > The <doc#term> technique is available to ordinary authors who can > > only use ftp to upload data to the web. > > Dan, > > This is a critical point, and one of the basic problems with aligning > the RDF world with the web and particularly http. On the Semantic > Web, we need to see everything visible in the RDF. Anything that > encodes some aspect of the meaning of a thing in the http protocol > and doesn't surface it in the RDF will lose when made available by > any other protocol (like from a triple store). Not necessarily; for example, the tabulator (http://www.w3.org/2005/ajar/tab ) copies http info into its triple store. > It's one of the > reasons that content negotiation is a loser for the Semantic Web, You lost me there. Care to elaborate, step-by-step? > and, similarly, solutions that rely on response codes such as 303, > since both of these mechanisms are http dependent. > > -Alan -- Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Monday, 1 October 2007 12:47:12 UTC