- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Sat, 28 Feb 2009 18:39:57 +0100
- To: Ben Adida <ben@adida.net>
- CC: Sam Ruby <rubys@intertwingly.net>, noah_mendelsohn@us.ibm.com, Henri Sivonen <hsivonen@iki.fi>, Mark Nottingham <mnot@mnot.net>, HTMLWG WG <public-html@w3.org>, RDFa mailing list <public-rdf-in-xhtml-tf@w3.org>, public-xhtml2@w3.org, "www-tag@w3.org WG" <www-tag@w3.org>
Ben Adida wrote: > Julian Reschke wrote: >> Ben Adida wrote: >>> ... >>> The registry doesn't/cannot change prior specs which specify that >>> @profile affects the value of @rel. >>> ... >> Again: @profile does not help with disambiguation when multiple profiles >> are declared. It just does not work, so pretending it does isn't helpful. > > Assuming this is true, that disambiguation is not possible, what does it > matter? The point is that @profile *must* be considered by your parser > looking at @rel, so you can't just compare @rel values. Even if @profile > is imperfect, it's part of the spec and you can't ignore it. It would be useful to not only consider what @profile may have been intended to do years ago, but what it actually *can* do in practice. (That would also be useful with respect to the discussion in the HTML 5 WG.) If I follow your argument, no new relation names can be used in HTML 4, except by using a profile. We have evidence that this is not true. For instance, consider rel="nofollow". Best regards, Julian
Received on Saturday, 28 February 2009 17:40:39 UTC