- From: Wing C Yung <wingyung@us.ibm.com>
- Date: Sat, 10 Feb 2007 13:12:47 -0500
- To: public-rdf-in-xhtml-tf@w3.org
If I understand this correctly, I want to point out that the syntax document [1] has an error in 6.2. <section id="person"> ... <h1 property="foaf:name">Dan Brickley</h1> </section> will not generate <#person> foaf:name "Dan Brickley"^^rdf:XMLLiteral . Instead, it will generate <> foaf:name "Dan Brickley"^^rdf:XMLLiteral . I will be including several tests in the test suite to make sure that the id attribute behavior is properly tested. Wing Internet Technology wingyung@us.ibm.com 617.693.3763 Ben Adida <ben@adida.net> Sent by: To public-rdf-in-xht bob@snee.com ml-tf-request@w3. cc org public-rdf-in-xhtml-tf@w3.org Subject Re: question on RDFa subject 02/09/2007 10:50 resolution AM Bob DuCharme wrote: > When should it pick up an id attribute as a subject? If I make a mistake here (and this is why we need to finish the Syntax document ASAP), I hope Mark can come in and correct me :) In the current specification, and ID attribute becomes the subject in two specific cases: 1) the triple you're considering was generated by a LINK or META, and there is an ID on the immediate parent. 2) you can encounter, as you go up the ancestor hierarchy and BEFORE you hit an ABOUT, an element with a rel="", no href="", and an id="". This case is *not yet* reflected in the RDFa Syntax document, and is meant to handle the following striping situation: ======== This paper was written by <div rel="dc:creator" id="me"> <span property="foaf:name">Ben Adida</span>, <a rel="foaf:mbox" href="mailto:ben@adida.net">ben@adida.net</a>. </div> ======== which yields: ======== <> dc:creator <#me> . #me foaf:name "Ben Adida" ; foaf:mbox <mailto:ben@adida.net> . ======== Note how, in both of these cases, the absence of an ID would yield a bnode. That's the way you can remember where ID comes in: if you would otherwise get a bnode, but you actually want it addressable, you can use ID. Note also how, in both cases, an ABOUT on the *same* element will trump the ID. -Ben
Received on Saturday, 10 February 2007 18:12:59 UTC