Re: PROPOSAL: Split RDFa into two pieces--core attributes, plus language-specific 'interpretations'

Hi José,

> I agree with your proposal, but I was thinking more in depth about it
> and it seems to me that the RDFaCore spec is going to be something like
> the spec of some attributes without anymore.
> I mean, that, without the host language, you are not going to be able to
> provide "semantics" to the RDF-A attributes.

The idea is that RDFa-core still requires a host language--after all,
it's only a collection of attributes--but the question is whether the
host language provides any additional 'mappings', or not.

For example, I might define a really simple XML language that contains
only two elements, <a> and <b>, with <a> being the root element,
containing any number of occurrences of <b>. I could then add
RDFa-core to this language, and be able to produce mark-up like this:

  <a>
    <b
     about="http://blah"
     property="http://xmlns.com/foaf/0.1/name"
     content="Mark"
    />
    <b
     about="http://blah"
     property="http://xmlns.com/foaf/0.1/homepage"
     resource="http://internet-apps.blogspot.com/"
    />
  </a>

In my view, we should clearly define the processing rules for
RDFa-core, so that the attributes work in all situations--which as it
happens, we have pretty much already done. So the example above could
be abbreviated to this:

  <a about="http://blah">
    <b
     property="http://xmlns.com/foaf/0.1/name"
     content="Mark"
    />
    <b
     property="http://xmlns.com/foaf/0.1/homepage"
     resource="http://internet-apps.blogspot.com/"
    />
  </a>

Now, in my simple language, there are no other semantics, so RDFa-core
gives me everything. But that isn't the case with HTML 4 or XHTML 1.x.

Does that explain it better? Or have I missed your point?

Regards,

Mark

-- 
  Mark Birbeck, formsPlayer

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

  standards. innovation.