random

[David C. Ullrich]

On 4 Jun 2001 10:41:55 GMT, tage at pasta.cs.uit.no (Tage Stabell-Kulo)
wrote:

>"Tim Peters" <tim.one at home.com> writes:
>
>>[Darren New]
>>> how random is the binary string "1001010010011100100"? How
>>> random is the binary string "0000000000000000000"? Such questions
>>> don't make any sense.
>
>>BTW, you can find attempts to answer such questions in Knuth, Vol 2.  The
>>questions not only make sense, but finding non-trivial answers is important
>>in real life:  even if you have a theoretically perfect physical RNG, how
>>can you have confidence in a specific real implementation?  
>
>
>Knuth answers a different question.  His question is "Algorthm X
>produced these bits: 010101.  Based on them, can we detect some
>patterns that should refrain us from relying on this algorithm as a
>source of (pseudo)random bits?"  In other words, he is analysing
>sources of randomness, not whether some string is random.  An
>important difference.

He's agreeing with me then. When this gets out people are going
to stop buying his books.

>Randomness is interesting only for the next bit, not the ones you
>already have.  Because you have them, they are not random.  Thus, the
>bits 1001010010011100100 or 0000000000000000000 are not random, although 
>they might have been.
>
>
> [TaSK]
>
>
>-- 
>--
>////       Tage Stabell-Kuloe        |e-mail: Tage at ACM.org (at=@)////
>///Department of Computer Science/IMR|Phone : +47-776-44032         ///
>//9037  University of Tromsoe, Norway|Fax   : +47-776-44580         //



David C. Ullrich
*********************
"Sometimes you can have access violations all the 
time and the program still works." (Michael Caracena, 
comp.lang.pascal.delphi.misc 5/1/01)