Re: tests for qualified RDFa attributes [was: ODF and semantic web]

Hi Dan,

> > Ivan has implemented a parser for XHTML+RDFa - and in that language
> > thare are NO namespace-qualified attribute names.
>
> Oh. Now I'm totally confused.
>
> Maybe it'll become clear to me in due course.

:)

The idea is that we have this 'generic' way of adding metadata to
mark-up, consisting essentially of a set of attributes that can be
used in all languages. As you know these attributes are things like
@about, @property, @datatype, @resource, @instanceof, and so on.

However, these attributes can only be used in a host language, and
since some host languages have metadata features of their own, the
definition of 'RDFa in language Y' includes an 'interpretation' of
these features from an RDF perspective.

The reason this is especially confusing is that in the last couple of
years we've been working on 'RDFa in general' and 'RDFa in XHTML' at
the same time, feeding things backwards and forwards between the two.

To illustrate how the generic and the language-specific work together,
consider a relationship between two resources, such as a foaf:Person
having a home-page; in RDFa one could write this:

  <any_old_element
   about="http://...dan#me"
   rel="foaf:homePage"
   resource="http://...dan"
  />

However, RDFa doesn't really exist on its own, so we should really
look at this in the context of some language. Assuming that we go for
'RDFa in XHTML', this might be expressed like this:

  <div
   about="http://...dan#me"
   rel="foaf:homePage"
   resource="http://...dan"
  />

But almost certainly we are writing this mark-up for a 'purpose', so I
could put on my web-page this:

  Go check out
  <a
   about="http://...dan#me"
   rel="foaf:homePage"
   href="http://www.w3.org/People/Connolly/"
  >Dan's home-page</a>.

As you can see, it feels quite 'natural' to give an RDF interpretation
to this mark-up.

I don't know if that helps or not. :)

But essentially Shane is saying that in 'RDFa in XHTML' there is no
need to qualify the attributes, but in some other language, it may be
necessary to do so.

Of course then we're in the realm of defining 'RDFa in language Z',
which means we'd also have to define the 'RDF interpretation' for that
particular language combination. For example:

  <svg:text>
    Go check out
    <svg:a
     xh:about="http://...dan#me"
     xh:rel="foaf:homePage"
     xlink:href="http://www.w3.org/People/Connolly/"
    >Dan's home-page</svg:a>
  </svg:text>.

Or whatever.

The key thing is that the RDFa extension module already allows its
attributes to be used by other languages in this way--i.e., with or
without namespaces.

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 Monday, 15 October 2007 20:10:51 UTC