- From: Leif Halvard Silli <lhs@malform.no>
- Date: Fri, 06 Mar 2009 03:36:44 +0100
- To: Julian Reschke <julian.reschke@gmx.de>
- CC: Manu Sporny <msporny@digitalbazaar.com>, Maciej Stachowiak <mjs@apple.com>, Mark Nottingham <mnot@mnot.net>, "www-tag@w3.org WG" <www-tag@w3.org>, HTMLWG WG <public-html@w3.org>, RDFa mailing list <public-rdf-in-xhtml-tf@w3.org>, public-xhtml2@w3.org
Julian Reschke 2009-03-05 16.44:
> Manu Sporny wrote:
[...]
>> However, I don't think that's Julian's point. I believe that his point
>> is that, because of CURIEs, you have a two-stage process instead of a
>> one-stage process.
>>
>> For CURIES, you must:
>>
>> 1. Read the value in @rel.
>> 2. Lookup the prefix mapping and append the reference to the prefix.
[...]
>> For Microformats, non-CURIEs, etc., you must:
>>
>> 1. Read the value in @rel.
>>
>> Is that your issue, Julian? That because of CURIEs/RDFa, you have to do
>> more now to determine what @rel really means?
>
> Yes. Using a safe-CURIE wouldn't have prevented that, but at least it
> wouldn't break URIs in rel values.
But if the user agent would be reading the profile attribute value
and - as HTML 4 says - "perform some activity based on known
conventions for that profile", then one could have a two step
solution in text/html as well.
The problem is establishing the convention. May be the convention
could be to link unknown profile IRI to the CURIE via a <link>
element.
<LINK rel="http://creativecommons.org/ns#" href="[cc]" >
<a rel="[cc:morePermissions]" href="link">text.</a>
--
leif halvard silli
Received on Friday, 6 March 2009 02:48:20 UTC