Rule Based Information Aggregation and Filtering

An email on RuleML 2008 triggered some thoughts. If I have to submit a paper or participate in the challenge what would be some cool applications? It is easier to start with my own needs, as I do in every project we start with.

1. Need a way to classify my email so that I can decide what to read after I finish the most essential work items
2. Need a way to filter news to a very fine grained level of my interest
3. Need to track specific sub segments of the industry, some times up to a single company or even division or a product group in a company
4. Keep an eye out for emerging technologies (I became aware of the Semantic Web in 1999 and XML in 1998 by tracking W3C website)
5. Find podcasts, videos, blogs and articles on big ideas (TED Talks and HBR)
6. Track discussions and ideas on innovation, learning
7. Talk trends in software development, programming languages, web scale applications and parallel programming

I want to do this without being overwhelmed by the huge volume of information that comes from a wide variety of sources.

I tried several things from recurring searches to Google alerts. Even built our own information trackers (InfoMinder) and a Feed aggregation and filtering system (NewsMinder). I think a rule driven Information Aggregation and Filtering system that can be fine tuned constantly may be a really useful system.

In this system each user would build his own set of rules in a language as close to English as possible. I know it is a dream, but we can inch towards it. I need to think through a lot of issues before presenting a paper or building an application. This blog post is the first attempt to put down my thoughts.