Mark Boddy, Honeywell
Sheila Banks, AFIT
Synopsis: Now that the field has had several years and several major systems (both successful and unsuccessful) under its belt, what have we learned from all this? Have we identified the right requirements ahead of time or are we still blindly fumbling around in trying to build these systems? Are the "standard" approaches to knowledge engineering still valid in the face of large-scale systems? Or a more fundamental question: Are these techniques valid for the real world? Do we even have the correct knowledge representation for this problem? We will address these questions and attempt to determine what changes must be made in knowledge engineering from the lessons we have learned.