My solution to the Halting Problem: Question for Peter Olcott

From: Marc Goodman (marc.goodman_at_comcast.net)
Date: 07/30/04 

Date: Fri, 30 Jul 2004 05:39:21 GMT

Write a Turing Machine simulator H that takes as
input a Turing Machine M and input string w. H
simulates M on w for 20 state transitions. If M
has halted, H correctly stops in state "Y". If,
after 20 state transitions, M has not halted,
then H stops in state "N" which sometimes correctly
means that M does not halt, and sometimes incorrectly
means that M does not halt.

Perhaps Peter would care to explain why my solution
to the Halting Problem is any different from his
solution? Both work (would/will/whatever work)
correctly some of the time and incorrectly some
of the time. On the other hand, my solution is clearly
superior because I can implement it straight-forwardly
without relying on a magic "WillHalt" function.

So, Peter, why do you think your solution is better
than my solution?