Re: HTML in RDF

The HTML/RDF transformation may look like this (first attempt)

  <html>
  <head>
   <title>unification</title>
   <script>alert('unified, at last')</script>
  </head>
  <body><p>It has to be possible!</p><p>It just <a href="imperative.html"><em>has</em></a> to be!</p></body>
  </html>

maps to (something along the lines of)

  @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
  @prefix h: <urn:ns:html#> .

  h:html
    h:head (
      h:title rdf:value "unification"
      h:script rdf:value "alert('unified, at last')"
    ) ;
    h:body (
      h:p rdf:value (
        h:text rdf:value "It has to be possible!"
      )
      h:p rdf:value (
        h:text rdf:value "It just "
        h:a h:href "imperative.html"
          rdf:value (
            h:em rdf:value "has"
          )
        h:text rdf:value " to be!"
      )
    ) .

Ordering is preserved where necessary.

That doesn't really address the need for overlapping annotations, but then again, XML/XHTML wasn't designed for that, being strictly hierarchical. There'd need to be something along the lines of Danny's proposal for an RDF word processor with character-level representation.

Regards

Bent


 
  ----- Original Message ----- 
  From: Karl Dubost 
  To: Mark Birbeck 
  Cc: Hausenblas, Michael ; Bent Rasmussen ; public-rdf-in-xhtml-tf@w3.org 
  Sent: Friday, October 26, 2007 8:03 AM
  Subject: Re: HTML in RDF


  Hi Mark, Bent,

  Sorry for the misunderstanding. The question is very interesting  and  
  I mentionned something similar recently, in the thread about "title  
  element, meta, h1 and rdfa"
  http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2007Aug/0200

  Mark Birbeck (25 oct. 2007 - 17:55) :
  > To convert HTML to RDF would not simply be to take out the content in
  > the way that we have done with RDFa, or that SIMILE does via XSLT.

  It is interesting to come with an RDF schema for HTML. But not  
  obvious, it depends on which point of view you have on the document.

  * Meaningful view
     You could extract the meaning information for example from a  
  blockquote with a cite attribute
      <#anaisnin> a <quote>;
         h:text "… text of the quote…";
         dc:source "http://example.org/blah";
                  dc:language "fr" .
  * Or you could extract a structural view of the document what are the  
  p, the list structure, etc.
  * Or a functional view, what are the links and references inside a  
  document, the possible form values
  * Maybe more

  I wonder which one Bent is interested by.


  -- 
  Karl Dubost - http://www.w3.org/People/karl/
  W3C Conformance Manager, QA Activity Lead
     QA Weblog - http://www.w3.org/QA/
        *** Be Strict To Be Cool ***

Received on Friday, 26 October 2007 18:38:41 UTC