Greetings all, Just a reminder (for those of you in Duluth :) that we'll be having our first group meeting at 2pm in HH 321 (CS Conference Room). The main item on the agenda is Eliza. We'll be discussing the original paper on Eliza (from 1966) by Joseph Weizenbaum, and also discussing your experiences with Eliza and your efforts to reverse engineer what she must be doing based on the input and output you've seen in your dialogues with her (and of course you can also use the info in Weizenbaum's paper to help you understand).
One of the remarkable things about the Eliza paper is that Weizenbaum provides his source code, and it's remarkably easy to figure out (I think). It's a little cryptic, but overall once you figure out a few syntax rules in the language you can really follow it, and in fact I think this is why there are so many Eliza implementations out there, even though the original Eliza was programming in SLIP (a lisp like language) on an IBM 7094. So, yet another good argument for releasing code. :) There is another very famous chatbot called Parry that we'll talk about next week, and unlike Eliza it's very very hard to find a working implementation of Parry (I've tried but haven't so far). I suspect that if Parry's source code had been published that situation might have been very different. I've put a copy of the original Eliza paper in the files section of the Yahoo group, and please bring a printed copy of the paper with you. I'll provide copies of any interactions you may have sent, plus my own, so that we can discuss those as well. See you soon, Ted -- Ted Pedersen http://www.d.umn.edu/~tpederse
