I have started working on an RSS search engine to add to SproutSearch (http://www.sproutsearch.com). I used SimpleXml to parse the RSS. This makes it really easy to parse them. The general idea I have is to store items from the feeds in a database, along with their tag information. Then I can generate a bunch of tag pages once I have gathered enough items.

The tags will have their own pages on SproutSearch, but I am also thinking of combining all the different sources into one page. So, on a topic page, it might have some listings from Blogger, some from LiveJournal, and some from RSS. As I add more data sources, those can be integrated as well. Then it would have links to view a more detailed listing of a particular data source.