[PREV - ESCAPE_TUNNEL] [TOP]
ROUND_AND_ROUND
January 2-18, 2005
good, bad, good, evil, better, worse, best, worst...
The opposite of good is evil. E
o
The opposite of good is bad. G o
o
But evil is worse than bad. B
The opposite of bad is evil.
Fanaticism is bad.
So good and bad are better G-B
than good and evil >
G-E
There are gradations,
shades of light and dark:
b-w
So better or worse is >
better than good or bad. G-B
We must go beyond good and evil.
It is better if we do so.
A uniaxial good/evil is too simple.
It's better to consider multiple qualities.
G-E b3 w2
< | /
|/
b1 ----o---- w1
/|
/ |
b2 |
w3
Which model is better?
Which model is best for
evaluating which is
better?
To evaluate methods
of evaluations, one
could --
Assume a given standard of
meta-evaluation
Use an existing standard
to choose a new standard. (Which would be
an apriori meta-
Try all candidates in evaluation.)
terms of all other
candidates, and see if If not, one
there's a unanimous could take a
decision. majority vote.
I wonder
which GOOD
method
is best?
--------
[NEXT - META]