Brian Dennis at the New Media Hack has an idea to help with (personal) Knowledge Aggregation
I'd like to write a script that daily pulls down those tagged items, marks them as read, and in a blue sky world, fed them to a personal focused crawler. The crawler would do some clustering, context analysis, and automated search to build a compendium report for me. I'd pick it up the next day in my aggregator or stash it somewhere easily accessible. Call it knowledge aggregation.
Sharpreader and RSS Bandit do this from a different perspective. As they aggregate new items, they look for trackbacks and links in each post and build nested threads of those related materials. The default Newsgator view provides a "related" link that goes to their own search, but it can be customized to do other things. Bloglines provides a link to all items that reference the item you are viewing as well as providing a partial count, so you can see right away how many people have been referencing a given item (similar to the Technorati cosmos. It would be interesting to take this to the next level as Brian suggests and link to related-by-search items from one of the blog search tools (Feedster, PubSub, Waypath, Blogdigger, ...). And of course, these should look beyond web feeds to the web as a whole and bring back related materials. Brian suggests doing this only for specifically marked articles to save the spin cycles. And he suggests grouping all the requests into one cluster to find articles that are related to the topics I've highlighted.
How is this related to personal knoweldge management? In my world of research and reading, having my tools be smart enough to bring content to me that could be relevant without overwhelming me would be wonderful.