ACTION-240: Read thread on RDFa, CURIEs and profile and summarize http://lists.w3.org/Archives/Public/www-tag/2009Feb/0295.html

CC+ spec. allows:

<a xmlns:cc="http://creativecommons.org/ns#" rel="cc:morePermissions"
href="http://somecompany.com/revenue_sharing_agreement">somecompany.com</a>

Which is (roughly) allowed by:

RDFa in XHTML: http://www.w3.org/TR/rdfa-syntax/#relValues (text is slightly above this anchor)

BUT:

XHTML itself allows only tokens in @rel values: http://www.w3.org/TR/2008/REC-xhtml-modularization-20081008/abstraction.html#dt_LinkTypes but says it will "one day" allow QNames

(all well-summarized in http://lists.w3.org/Archives/Public/www-tag/2009Feb/0270.html by mnot)

Issues:

i) Link draft (http://tools.ietf.org/id/draft-nottingham-http-link-header-04.txt) doesn't cover the CURIE case at all (offers URIs and tokens only - with IANA registry URL being used to "qualify" tokens.

ii) XHTML spec. does not itself provide the extension point assumed by RDFa in XHTML

iii) CC spec. doesn't use the HTML4-provided 'profile' extension mechanism for @rel (ie. uFormats - http://microformats.org/wiki/rel-faq) so can't "properly" be delivered in text/html and likely isn't compatible with HTML5 either.

So:

=> What really, do we (TAG) think about CURIES - there seems a large amount of consensus on simple tokens (+ WKL for registry) and "complete" URIs, but not much around QName/CURIE solution where ns is described "separately" from identifier value itself.
=> status of the Link draft (ie. should QNames be allowed there - answer mnot's question?)
=> discrepancy between XHTML doc. and RDFa in XHTML

Possible actions:

Ask XHTML group what their plans are to use either a URI or a CURIE in @REL
Recommend something to mnot as a result of the above discussion
Document something general about use of CURIEs

Also relates to http://www.w3.org/2001/tag/doc/selfDescribingDocuments.html