On Mon, Jun 28, 1999 at 03:26:02AM +0000, David L. Nicol wrote: > If one was to combine the text analysis tools discussed by Rosie Jones > with the infobots discussed by Kevin Lenzo, you'd get an IRC channel > robot that could do imitations! I did something very similar to this at school several years ago (albeit with QuickBASIC (bletch) and a homebrew chat system that worked by swapping files over a network (more bletch)). My AI engine was of the NIALL type; a reasonable C[++?] implementation can be found for UNIX as gNiall (on Freshmeat). It worked remarkably well, chatting with users in the channel and learning at the same time. The "MegaHAL" AI program uses a modified NIALL engine which makes more sense---quite often it spits out reasonable answers to questions if you've told it them before. Stick that in an IRC channel---like #perl---and see how long it takes people to figure out that it's not a real person. :) At some point I'll reimplement NIALL in Perl or Python; I suspect it could be done as a Perl one-liner. Any takers? -- Adam Sampson azz@gnu.org ==== Want to unsubscribe from Fun With Perl? ==== Well, if you insist... Send mail with body "unsubscribe" to ==== fwp-request@technofile.org