- From: Sandro Hawke <sandro@w3.org>
- Date: Tue, 24 Dec 2002 13:11:15 -0500
- To: www-tag@w3.org
Let's just use the existing "rel" and "type" attributes on a-href elements. New values for rel are allowed if you specify an HTML profile [1] detailing the syntax and semantics of new link types [2]. It seems to me that that rel corresponds exactly to both RDDL purpose and RDF predicates, while type corresponds exactly to RDDL nature and (URI-style) media type. Here's the example: <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"> <head profile="http://www.w3.org/2002/12/RelRDF"> <title>The "L" Namespace</title> </head><body> <h1>The Example-L Namespace</h1> <p>This document addresses the <a href="http://lists.w3.org/Archives/Public/www-tag/2002Nov/0147">RDDL Challenge</a></p> <p>If you are using the Example-L namespace, you may be interested in <a rel="http://www.rddl.org/purposes#validation" type="http://relaxng.org/ns/structure/1.0" href="http://example.com/schemas/L.rng">a RelaxNG schema</a> and <a rel="http://www.rddl.org/purposes#render" type="text/css" href="http://example.org/style/L.css">a very terse style sheet</a> </p> </body></html> Which is also at http://www.w3.org/2002/12/RelRDF-extra/rddl-challenge Pros: - valid in HTML and XHTML - in the spirit of HTML; existing HTML works for namespace documents, but without REL= values, the information will be a little vague. (We're left with something like rdfs:seeAlso: an untyped link.) (Missing TYPE= values just mean you need to use the web to find out the type, as usual.) - machine parsing (eg conversion to RDF triples) is fairly easy. (But that's probably true of all candidates.) - can be served in parallel (via content negotiation) with an RDF/XML version giving the same information with more of it machine-readable. (But that's also probably true of all candidates.) Cons: - too Simple and Obvious? :-) - I decided to consider the "nature" information a media-type, because that's what it seems like to me. This means we need to handle URIs for media-types, which I think the TAG supports anyway. If someone can explain how nature is criticially different from media type, I have some ideas how to handle that. - raises some hard questions about httpRange-14 [4] and RDF identifiers. Or is that a "Pro", since it forced me to solve them (to my own satisfaction at least!) [5]. Thoughts? -- sandro http://www.w3.org/People/Sandro/ [1] http://www.w3.org/TR/html401/struct/global.html#adef-profile [2] http://www.w3.org/TR/html401/types.html#type-links [4] http://www.w3.org/2001/tag/ilist#httpRange-14 [5] http://lists.w3.org/Archives/Public/www-rdf-interest/2002Dec/0125
Received on Tuesday, 24 December 2002 13:14:04 UTC