Measuring interest (was: Re: Limits and Sequences (was: Re: [hackers-il] Interesting time_t))
- On Fri, 20 Apr 2001, Nadav Har'El wrote:
> Of course this is not true for a general property! But for our (very broad)Back to definition of "interesting-ness".
> "interesting-ness" property, you must agree that if a whole sequence is
> interesting, then so is its limit? If you knew an interesting sequence,
> wouldn't you be interested in its limit?
How about trying to define it, and seeing whether it satisfies the
property that: the limit of a sequence of elements, satisfying property P,
also satisfies property P?
One possible way to try to define it:
Expose the possibly-interesting object to the sense organs of N persons,
and determine their reactions by means of a polygraph.
If at least M (M <= N) subjects exhibit responses higher than a certain
threshold, then the object is deemed to be interesting.
From the above empirical and physical definition, the next step is to
abstract a mathematical definition (such as one using the value of a
cross-correlation between the subject and sexually explicit material of
the same dimensionality), and henceforth use it as the mathematical
definition of "interesting-ness".
There is no IGLU cabal. It has abysmally low "interesting-ness" score.
WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html