A Prehistory of Cyc
The Cyc project was actually incepted by
Alan Kay at the
Atari Research Centre in the
70's. Kay was planning to build a large
computerized collaborative encyclopedia of
knowledge in an experimental project for
Atari.
Sound familiar?
He approached Dr. Douglas B. Lenat for his prowess in the field of artificial intelligence to see if he could make a contribution to the database. It was then that Lenat suggested "automating the white spaces" within the encyclopedia. What he meant was programatically applying common sense rules to the database such as "water cannot flow uphill" and "two people cannot be in the same place at the one time".
As we all know, Atari hit a brick wall in the early 80's and one of the first departments to be struck off was research, so Lenat moved the project to Microelectronics and Computer Technology Corp.
Read Enth's writeup on the history of and application of Cyc.
What stage is Cyc at today?
Cyc has now become so "intelligent" that it scans it's own texts and adds the information to itself, it only requires the grammatical rules of the language to be applied by the researchers.
However, critics of Cyc such as Hubert Dreyfus have described it as the "last bastion of top down, strong artificial intelligence" in that it will fail to produce humanlike intelligence the same way all other top down AI projects are "failing" ie. that the human mind is just a complex symbol processing machine.
However, this is a different debate.
Much information from Paradigms Regained by John L. Casti, a great read.