Completely off-topic but interesting:
http://www.20q.net/
While funny as a game it's a total mess if one thinks this method as an
appropriate method of computer learning. ;-) Tried it with mykorrhiza
and abondoned after the 28th step when the prog suggested fungus as a
solution... I don't know if it would have asked a while later what I
had in mind. Strictly spoken I didn't teach it mykorrhiza *and* taught
it some wrong facts on fungi. Although having no bad intension's at all
I therefore managed to worsen the decision engine twice in only one try.
What could be done with a (truly) ill natured mind. <BEG>
Learning should *always* involve revision by experiment and/or authority.
Sure I should have read the FAQ and all the fine manuals, though... ;-)
BeAr