W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > November 2004

rdf-html interlinking (was: use case question)

From: Benjamin Nowack <bnowack@appmosphere.com>
Date: Thu, 18 Nov 2004 21:20:07 +0000
To: Mark Birbeck <mark.birbeck@x-port.net>
Cc: public-rdf-in-xhtml-tf@w3.org
Message-id: <PM-EV.20041118212007.96174.1.1D@bodkin.nuigalway.ie>


Thank you very much for the reply, Mark.
I've spent some hours reading the RDF/A doc [1] now, and
I think I worked out a way to markup my use case. Could
someone please check, if I got the RDF/A basics right?

original html snippet
[[
<p>
  John Doe wrote in his
  <a href="http://jd.com/blog/">weblog</a> how to build a
  Semblog. Here is the abstract of his
   <a href="http://jd.com/blog/4711.html">post</a>:
   <blockquote>
      A semblog can be built by ...
   </blockquote>
</p>
]]

In order to be able uniquely identify which John Doe I'm
talking about, I'd need some triples such as:

[[
_:p1 foaf:name "John Doe".
_:p1 foaf:weblog <http://jd.com/blog>.
]]

To provide a rich rdf description of my page,
I'd also like to generate:

[[
<http://jd.com/blog/4711.html> dc:creator _:p1.
<http://jd.com/blog/4711.html> dc:description: "A Semblog can be...".
]]


If there was a URI for John Doe, I could easily use that,
but in many cases (especially for persons), people tend
to use bnodes and inverse functional properties to
identify resources. [The RDF/A doc's examples (mis-)use
mailbox URIs (mailto:daniel.brickley@...) or weblog profile
URLs (http://www.blogger.com/...) as URIs for persons.
I know, it's a draft, and sorry for being pedantic, but
the examples really confused me as I first thought there
was some mechanism that could create the missing bnode
indirection. As the RDF/A doc addresses both html and rdf
folks, it'd surely help to fix these bits.]

But back to my basic problem: I know I can create bnodes
by using a nodeID:

[[
<meta nodeID="p1" property="foaf:name" content="John Doe" />
<link about="bnode(p1)" rel="foaf:weblog" href="http://jd.com/blog/"/>
]],
(I can't simply repeat <link nodeID="p1" ../> for the weblog
here, can I? that would be more intuitive for RDFers, but I
guess things are different in xml..)

but I actually don't want to retype the whole information
and I want to have the RDF directly associated with my html
sections. If I understood the doc correctly, I can do:
[[
...
   <meta nodeID="p1" property="foaf:name">John Doe</meta>
...
]]

This will type "John Doe" as an XMLLiteral which isn't
really an issue for me, I guess.

The weblog part can't be encoded that way, as
http://jd.com/blog/ is a URI, not a Literal. But using the
bnode construct, is it possible to say
[[
...in his
<a about="bnode(p1)" rel="foaf:weblog"
                  href="http://jd.com/blog/">weblog</a>
]]
?

That would be cool, cause then I could probably also enhance
the blog item anchor with
[[
...
of his
<a about="bnode(p1)" rev="dc:creator"
                 href="http://jd.com/blog/4711.html">post</a>:
...
]]

The only redundant part would then be the blog item URI in the
blockquote, but I could live with that:
[[
<blockquote about="http://jd.com/blog/4711.html"
         rel="dc:description">
   A semblog can be built by ...
</blockquote>
]]

Are my examples correct?

Ouch, sorry for the long mail, just trying to
understand that stuff. Am I right, that RDF/A is for
xhtml2 documents only, and that there is not a subset
which could be applied to xhtml1 docs without turning
them invalid?


Thanks a lot,
benjamin

[1] http://www.formsplayer.com/notes/rdf-a.html


--
Benjamin Nowack

Kruppstr. 100
45145 Essen, Germany
http://www.bnode.org/



On 18.11.2004 11:56:14, Mark Birbeck wrote:
>
>Benjamin,
>
>> My idea was to add a click event to the
>> span tag which could e.g. display a div with a  "find out
>> more about this person" link.
>>
>> Is that a use case RDF/A covers?
>
>That's definitely the type of thing we've been trying to get to. A very old
>draft of RDF/A has a slightly more complicated example (since it uses a
>level of indirection), but as you can see from this little screen-shot:
>
>  <http://www.formsplayer.com/notes/xhtml-meta-data-02.html#div353433704>
>
>the principles are much the same as the use case you describe.
>
>Regards,
>
>Mark
>
>
>Mark Birbeck
>CEO
>x-port.net Ltd.
>
>e: Mark.Birbeck@x-port.net
>t: +44 (0) 20 7689 9232
>w: http://www.formsPlayer.com/

>
>Download our XForms processor from
>http://www.formsPlayer.com/



Received on Thursday, 18 November 2004 21:18:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:14:59 GMT