- From: Manu Sporny <msporny@digitalbazaar.com>
- Date: Fri, 19 Mar 2010 23:08:10 -0400
- To: RDFa WG <public-rdfa-wg@w3.org>
There were two issues that came to light during the discussion last Thursday during the telecon... documenting them here to see if any solutions or alternative viewpoints pop up. So with the current @token proposal, it is asserted that the following will happen: * @token declarations, when specified in RDFa Profile documents, will be pulled into author documents via @profile. * @token declarations are processed differently in RDFa Profile documents than they are in regular documents. Here are the concerns with the previous points: 1. Are we promoting the use of hidden data in RDFa Profile documents? Specifically, if one must do this to have their @token declarations seen via @profile: <html token="Person: http://xmlns.com/foaf/0.1/Person; name: http://xmlns.com/foaf/0.1/name; depiction: http://xmlns.com/foaf/0.1/depiction;"> ... </html> Is that a good thing? Hidden data is commonly viewed as an anti-pattern in the semantics-embedded-in-HTML world - we do it only if there is no other alternative available. 2. Are we comfortable with @token not following the standard all-RDFa-attributes-are-scoped design pattern that has served us well up to this point? So, if the RDFa Profile document contains something like this: <html> ... <body> <p token="Person: http://xmlns.com/foaf/0.1/Person"> You can use "Person" in the typeof attribute to specify that the subject is a human being. </p> ... </body> </html> If we want to allow that markup, which one could argue is more desirable than the markup outlined in #1, it would mean that we must view @token as scope-less when processed as a part of an RDFa Profile. It's an exception that makes understanding RDFa slightly more complicated. Not a deal-killer by any means, but it's concerning none-the-less. -- manu -- Manu Sporny (skype: msporny, twitter: manusporny) President/CEO - Digital Bazaar, Inc. blog: PaySwarming Goes Open Source http://blog.digitalbazaar.com/2010/02/01/bitmunk-payswarming/
Received on Saturday, 20 March 2010 03:08:39 UTC