What to Do

Define a small, feasible, fun, potentially useful semantic web project. Do this right and you could have the next Flickr or Facebook.

• Small and feasible means that there's some simple examples that can be implemented right now, using the Lisp tools and skills you already have, and Semantic Web data that exists or could be easily generated.
• Fun means something that interests at least you and those two or three other weird people like you.
• Potentially useful means that there's something that people do -- or would like to do -- that could be improved if computers had the kind of knowledge a Semantic Web-based system could provide

This will require a combination of searching the web for what kind of machine-usable knowledge is out there (or almost there), and how such knowledge could make a difference. Start with but do not limit yourself to the links on the Semantic Web page.

What to Submit

Email me a report (plain text, Word file, RTF, or PDF) on a possible class project that you'd like to do. The project should be doable as a series of relatively small exercises, similar to the existing semantic web exercises. I can help you break things down. Your project description should be short, a page or so, but clearly describing:

• What would be built, i.e., what kind of thing is it?
• Examples of how it might be used, e.g., make up some calls to Lisp functions, e.g., the deductive retriever or the XML-RPC functions, that would get some answers, based on knowledge gathered, dynamically or in advance, from the Semantic Web.
• Links (with descriptions) to idea sources, ontologies, rules, whatever on which this project would be based.

Comments? Send mail to Chris Riesbeck. Put EECS 325 in the Subject.