Envisioning a Future With Intelligent Software Agents

The notion of Intelligent Software Agents (ISA) capable of finding, organizing, and analyzing information on behalf of their owners is the most important concept regarding the Internet today. This new generation of agent technology promises to help reduce the cost, effort, and time of filtering information on the Web.

Although some of the ISA concepts will never fly, other applications are sure to create the next wave of Internet buzz.

Most of the buzz so far has been around using ISAs to produce a kind of “local assistant” that runs on the user’s computer and interacts with information or other users on behalf of its owner. Another idea for using ISAs is to somehow optimize user searches for relevance.

In theory, this sort of “search assistant” agent could help you develop a report by filtering out non-useful references from the flood of stuff a search engine generates -- for example, making sure you don’t get fast food restaurant references when you need a poet’s name.

I don’t believe the search assistant ISA is going to happen because it faces three insurmountable problems. First, current search engines do so much that it’s hard to create search ISAs with enough incremental value.

Second, the issue of more relevant searching will probably be better handled through the evolution to the “Semantic Web,” which explicitly addresses data relationships. Third, “relevance” means picking results that match the users’ needs. This conflicts with search engine operators that are being paid by advertisers to do the opposite.