From: Mu In Taiwan (email@example.com)
Date: Mon Oct 12 2009 - 23:02:45 MDT
> > Please present an algorithm, or state-transition function for a turing
> > machine, implementing:
> > a. A real time clock.
> > b. A true random number generator.
> > Good luck with these. You'll need it.
> > Of course, my Mac has these features as an integral part of the machine.
> I could be wrong but I don't believe the the Mac has a true random
> number generator, just pseudo random
You are wrong, and you could have checked your facts even by using Wikipedia
for about 30 seconds, rather than continuing to spew your junk claims out
all over the mailing list.
" In Unix-like <http://en.wikipedia.org/wiki/Unix-like> operating
, */dev/random* is a special
serves as a true random number
as a pseudorandom number
It allows access to environmental noise collected from device
"/dev/random and /dev/urandom are also available on
, Mac OS X<http://developer.apple.com/documentation/Darwin/Reference/Manpages/man4/random.4.html>
> ; and it can tell the time because
> it's connected to the outside world"
I am quite certain the real time clock in my Mac would continue to work even
if it were isolated from the universe by cold vacuum and zero incoming
radiation, assuming the battery were sufficiently charged. If you disagree,
please state reasons and evidence.
> But no matter, try beating a Jupiter Brain armed with nothing but a
> random number and the time of day. Good luck. You'll need it.
I don't know what your idea of a Jupiter Brain is, since you continually
refuse (or are unable) to define the vocabulary and ideas you are using.
I shall assume that it is an algorithm running on an asynchronous large
finite state machine (implementing a crude almost-turing machine, as most
computers do), without a real time clock or true random number generator.
I therefore challenge your "Jupiter Brain" to a game of
stone-paper-scissors, armed with my random number generator. There is a
possibility that I may be able to model its mental state, and thus beat it
consistently over the long term. If it appears capable of modelling my own
mental state better than I am of modelling its, I will switch to using my
random number generator to force a draw over the long term. If I have any
doubts about my ability to model its mental state, I will use the random
number generator from the beginning. In the worst case I can force a draw;
in the best case, I can win consistently. Looks good to me.
Oh, and if I am feeling a bit cheaty, I will play it once, hit reset, and
then beat it consistently by consistently replaying the first game over and
over via the reset button, knowing what it will output.
Next I will challenge it to a game I played as a child. Remaining silent,
without looking at a watch, count to 60 seconds in your head, and call out
when the minute is up. The individual whose guess is closest to the real
time measured on a clock, wins.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:04 MDT