Manu Sporny wrote:
This is the crux of the argument, as far as I see it, and I have to
strongly agree with Mark's position. RDFa is not a vocabulary, thus
placing something in @profile doesn't make much sense as @profile is
meant for specifying definitions of vocabularies[1].
I concur. However, *part* of RDFa is a vocabulary. We define values
for @rel/@rev. So.... it *could* be argued that we the thing at the end
of the profile URI should define that vocabulary. And then RDFa parsers
could use that vocabulary AND know that when that vocabulary is in use
the document is definitely RDFa.
One approach would be to create a vocabulary that specifies the type of
metadata that is embedded in the document, so, we could have something like:
<head profile="http://www.w3.org/profiles/html-metadata";>
<meta name="metadata-flavor" content="xhtml+rdf">
<meta name="metadata-flavor" content="microformats">
</head>
Interesting idea - add a layer of abstraction. Its a classic. Also
happens to work in HTML 4 and XHTML.
Another approach could have us using @class to hint to the user agent
that there is RDFa data in the document, something like:
<div class="xhtml-rdf">... RDFa goes here ...</div>
Can I say "ick"? I really think this is an overloading of @class
(despite the fact that @class can contain anything). Plus, we are
really talking about announcement, and that should happen in the head IMHO.
--
Shane P. McCarron Phone: +1 763 786-8160 x120
Managing Director Fax: +1 763 786-8180
ApTest Minnesota Inet: [EMAIL PROTECTED]
