> Interestingly, some of Doug Lenat's early work was maybe more relevant to that problem than Cyc is.
Yeah, Eurisko was really impressive, I often wondered why people don't work on that kind of stuff anymore.