W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > January 2008

Changes to syntax document for @rel and @rev

From: Mark Birbeck <mark.birbeck@formsPlayer.com>
Date: Wed, 23 Jan 2008 00:00:07 +0000
Message-ID: <a707f8300801221600q7a9dfcafh52b6a0d6291d6a54@mail.gmail.com>
To: "W3C RDFa task force" <public-rdf-in-xhtml-tf@w3.org>

Hello all,

I've still got more to do on the syntax document, but I thought I'd
save the changes that I've made to the document to handle @rel and
@rev processing, as we've been discussing.

I've added a new section in the general processing section, called
"CURIE Processing". If you can see Shane's server you'll find it here:

  <http://htmlwg.mn.aptest.com/htmlwg/rdfa-syntax/#sec_5.4.>

If you can't see that, the prose goes like this:

--- STARTS ---

Many of the attributes used in RDFa are defined to carry a list of CURIEs. Some
attributes support values of an additional type, such as a URI, and in
this case the
CURIEs used must be [safe CURIE]s. For example, @about can contain a URI:

      <div about="http://dbpedia.org/resource/Albert_Einstein">
        <div rel="dbp:citizenship"
resource="http://dbpedia.org/resource/Germany"></div>
        <div rel="dbp:citizenship"
resource="http://dbpedia.org/resource/United_States"></div>
      </div>

or a [safe CURIE]:

      <div about="[dbr:Albert_Einstein]">
        <div rel="dbp:citizenship"
resource="http://dbpedia.org/resource/Germany"></div>
        <div rel="dbp:citizenship"
resource="http://dbpedia.org/resource/United_States"></div>
      </div>

In general, all non-CURIE values MUST be ignored. For example, the following
value in @about would not set a new subject, since the CURIE has no prefix:

      <div about="[Albert_Einstein]">
        <div rel="dbp:citizenship"
resource="http://dbpedia.org/resource/Germany"></div>
        <div rel="dbp:citizenship"
resource="http://dbpedia.org/resource/United_States"></div>
      </div>

There is one exception to this, and that is that in @rel and @rev, any
value in the list
in the section on 'The rel attribute', MUST be treated as if it was a
full CURIE with the
XHTML vocabulary as its prefix mapping. For example, the following are exactly
equivalent:

      <link rel="next" href="http://example.org/page2.html" />
      <link rel="xhv:next" href="http://example.org/page2.html" />

(Note that a [safe CURIE] is not required here, since @rel and @rev do not take
non-CURIE values.)

However, the following will be completely ignored by the processor,
since it is not a
valid CURIE and is not included in the list of values in the section
on The rel attribute:

        <link rel="foobar" href="http://example.org/page7.html" />

<issue title="rel and rev values">This has not yet been voted on, but there
is a general desire that legacy values like @rel="next", should cause a triple
to be added to the [default graph].</issue>

--- ENDS ---

The document already includes the part that causes non-prefixed values
to be ignored, so I won't put that here. So just so that the approach
is clear, what I'm doing is saying in one part of the document 'this
must be a valid CURIE', and then in another part of the document, 'a
valid CURIE must have a colon'.

Comments are, as (nearly) always ;) gratefully received. However, I
don't plan to edit this much more for now, since it's main purpose is
for us to discuss. Of course, if there are no objections to it, it
wouldn't hurt to vote on it and get it out of the way -- one less
issue to worry about.

But I'll leave that side of it to Ben to sort out.

Regards,

Mark

-- 
  Mark Birbeck, formsPlayer

  mark.birbeck@formsPlayer.com | +44 (0) 20 7689 9232
  http://www.formsPlayer.com | http://internet-apps.blogspot.com

  standards. innovation.
Received on Wednesday, 23 January 2008 00:00:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 23 January 2008 00:00:24 GMT