From: Eliezer Yudkowsky (sentience@pobox.com)
Date: Sun Aug 01 2004 - 11:52:18 MDT
I'm not nervous about being mistaken for a cult by anyone who can tell the
difference without a checklist.
Statistical notes: A characteristic such as (1) "thinks he's a genius"
that correlates with both crackpots and geniuses is not useful for
distinguishing between crackpots and geniuses unless you can demonstrate a
differential conditional probability that would make the signifier Bayesian
evidence. Elementary statistical fallacy, but a very common and tenacious one.
(And yes, many geniuses know damn well they're geniuses. They're not
stupid, though those who are not strong in the Way may be dishonest to
themselves or to others.)
In all seriousness, I don't think this sort of checklist is valuable.
Gardner should have known better, devising a weapon that someone might as
easily turn *as an accusation* against any number of non-crackpot theories.
Martin Gardner presumably had some way to distinguish who was and wasn't
a crackpot *before* he came up with this handy checklist, and that's what
he needed to teach his readers. The checklist exalts the signifier over
the substance, and arguing over checklist items is the sort of thing that
pops up all the time in, ahem, unmoderated usenet groups.
And as a practical public relations issue, if you focus on this sort of
thing, it just makes people think that there's something to argue over.
Focus on genuinely being a noncrackpot, not on not looking like a crackpot,
and anyone who can tell the difference without a checklist will understand.
-- Eliezer S. Yudkowsky http://intelligence.org/ Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Tue Feb 21 2006 - 04:22:43 MST