RSS soundbite

(OK, this isn't a soundbite ;)

re http://tbray.org/ongoing/When/200x/2003/06/04/Evangelizing 
  "RSS is so personalized, though, that your set of feeds, 
   or mine, are not apt to impress the person you're 
   talking to unless they're a lot like you."

my take is that RSS and related technology will be an easier sell
once RSS feeds can be filtered by various more interesting criteria 
than simple categories.

I guess I'm mostly thinking of feeds representing weblogs, and wanting 
to filter on characteristics of the author (hence FOAF etc) rather than 
characteristics of the feed itself.

Weekend fiddling produced a demo of RSS/weblog filtering based on 
CPAN contributions. I'd like to do the same with W3C TR authors/editors, 
ie. scenarios along lines of 'Find me weblog items by contributors to 
W3C specs' (or IETF specs, or CPAN modules, or ...).

For example, http://rdfweb.org/2003/06/cpan/lookups.rb matches the following
RDF query template:

	SELECT ?name, ?homepage, ?weblog, ?x, ?feed, 
	WHERE 	(foaf::name ?x ?name)
		(dc::contributor ?x http://www.cpan.org/)
		(foaf::homepage ?x ?homepage)
		(foaf::weblog ?x ?weblog)
		(foaf::rssfeed ?weblog ?feed)
	USING 	dc for http://purl.org/dc/elements/1.1/
		foaf for http://xmlns.com/foaf/0.1/

...against a merge of CPAN contributor data with a list of people, their weblogs,
and the RSS feed URIs for those weblogs. It gives tabular results, looking like:

	Earle Martin homepage: http://downlode.org/ weblog: http://downlode.org/blog.pl rss: http://downlode.org/blog.pl/index.rss10
	Jo Walsh homepage: http://www.zooleika.org.uk weblog: http://www.zooleika.org.uk/blog.html rss: http://www.zooleika.org.uk/blog.rss
	Norm Walsh homepage: http://nwalsh.com/ weblog: http://norman.walsh.name/ rss: http://norman.walsh.name/index.rss


And there's a problem. There are only 3 hits, because I hand-coded the person 
descriptions in http://rdfweb.org/2003/06/cpan/perl-blog-dir.rdf for this demo. 

My working hypothesis is that finding/filtering RSS feeds, weblogs etc. is mostly going 
to be done via descriptions of things other than those documents. Usually it'll be 
characteristics of the author that are of interest. But we don't have enough of the 
latter descriptions yet to make compelling demos, except to the early-adopters who 
enjoy seeing the point of obscure-ish technologies.

Maybe it's just a matter of bootstrapping enough data to impress folks, by converting 
CPAN lists, some weblog directories, etc?

Dan