91.520-Internet and Web Systems Project
The Research Engine Project

What would be the component of such a helper?

1. A "registry," where one could register keywords for topics to be monitored
2. A search engine, which goes out and searches those topics
3. A "monitoring agent," which, once information sources have been found by the search engine, keeps monitoring them for updates and changes (a VERY crude example of such a monitor is URL Minder, crude because it only tells you that the contents of the monitored site have changed, but nothing more)
4. A downloader, which downloads information that is found
5. A report manager, which allows you to specify the frequency and depth of reporting results to you
6. A report producing agent
7. A follow-up agent, which seeks additional information on selected items
8. ...and more

This is such an obvious idea that you are going to find several systems attempting to provide these services for you.

The closest I have seen so far is Compernic, which I have just started using (an in-depth report and review to follow later).

Infoseek's Express is a far weeker attempt, to the (very limited) extent I have been able to use it in the last few day (more later).

There is also Altavista's Discovery, which is more of a search engine on the client-side machine. My limited experience with it has been quite disappointing so far.

So, there is still an opportunity to do it right. And there is room for multiple projects towards this goal.

Copyright © 1998 Dr. Haim Levkowitz Associate Professor, UMass Lowell Department of Computer Science