what do linked data clients do with content of 200 responses?

Looking for some empirical information here:

What do linked data clients such as Tabulator and Syndice do with the
content of 200 responses? Do they scrape it to get RDF? Or do they
just ignore it?  Or do they filter it somehow, e.g. scrape it and then
ignore the content if it doesn't mention the target URI?

E.g. if the URI is U, and GET U -> 200 Z, and Z is text/html
containing RDFa, and the RDFa says <> dc:title "title", does the
client learn that <U> dc:title "title" ? If not, what about Z of type
application/rdf+xml ?

I guess I can ask the linked data list this question, but thought I'd
try here first.

Thanks
Jonathan

Received on Friday, 17 February 2012 12:25:39 UTC