[107795] in Cypherpunks
CDR: Re: Redundancy and Randomness [Information Theory] (fwd)
daemon@ATHENA.MIT.EDU (Jim Choate)
Sat Jan 23 22:08:14 1999
From: Jim Choate <ravage@EINSTEIN.ssz.com>
To: cypherpunks@EINSTEIN.ssz.com
Date: Sat, 23 Jan 1999 20:56:10 -0600 (CST)
Reply-To: Jim Choate <ravage@EINSTEIN.ssz.com>
----- Forwarded message from Tim May -----
Date: Sat, 23 Jan 1999 18:44:48 -0800
From: Tim May <tcmay@got.net>
Subject: CDR: Re: Redundancy and Randomness [Information Theory]
Interestingly, one can never "prove" that a string (or message) is random.
One can establish grounds for believing this to be so, that the message was
generated from a random-like source (e.g., Johnson noise, radioactive
decay, whatever).
And there's Von Neumann's comment about those who believe in mathematically
generated sequences being random are living in a state of sin.
----- End of forwarded message from Tim May -----
Absolutely. In fact no finite length stream can be random in any definition
of random. The odds of any of its elements being a particular character in a
character set may be however. For a sequence to be considered random it must
be infinite and non-repeating on any scale (in RNG talk it has no modulo),
this alleviates the potential of a larger window being available in which to
demonstrate the sequence repetition.
It's a subtle distinction.
____________________________________________________________________
What raises the standard of living may well diminish the
quality of life.
The Club of Rome
The Armadillo Group ,::////;::-. James Choate
Austin, Tx /:'///// ``::>/|/ ravage@ssz.com
www.ssz.com .', |||| `/( e\ 512-451-7087
-====~~mm-'`-```-mm --'-
--------------------------------------------------------------------
