Re: Finding RDFa content on the web

Mark Birbeck wrote:
> ... As you can see, RDFa tries to gently build on HTML best practice, but ramps
> up to full-blown RDF (we've sneaked reification and bnodes in there!). But
> at what point you say you are using RDFa, I don't know. ...

The 'gentle ramp-up' sounds like a good idea for people.
Does this presuppose a use case where people are entering
the markup manually?  Will it be important if most pages
with RDFa content are generated by special editors or from
databases or programs?

Our work on Swoogle is focused, I suppose, on the use case
where the semantic web is being used by programs and
software agents.  We imagine that software agents will need
to look up information on the Semantic Web just like people
find information on the conventional Web, e.g.  using search
engines.  Programs and agents will want to find web pages
that are both likely to have the data that they seek and
also to have it encoded in a form they can understand
(e.g., RDF).

If you have a way to extract any RDFa content on a given
page and are willing to do an exhaustive Web crawl then I
suppose there is no problem.  Google could extend its
metadata model and indexing approach to support searches
that included RDF elements.

But it would be nice to be able to do it using today's
search engines and/or using the approach Swoogle takes.

Tim

-- 
  Tim Finin, Computer Science & Electrical Engineering, Univ of Maryland
  Baltimore County, 1000 Hilltop Cir, Baltimore MD 21250. finin@umbc.edu
  410-455-3522 fax:-3969 http://umbc.edu/~finin http://ebiquity.umbc.edu

Received on Thursday, 8 June 2006 19:02:03 UTC