Velocity Reviews > Re: Why is rand special?

# Re: Why is rand special?

Roc
Guest
Posts: n/a

 07-03-2003

"Richard Heathfield" <(E-Mail Removed)> wrote in message
news:ba8pho\$15\$(E-Mail Removed)...
> Malcolm wrote:
>
> >
> > "Richard Heathfield" <(E-Mail Removed)> wrote in message
> >>
> >> You just can't tell what is random and what isn't, just by looking at

the
> >> sequences. 1, 1, 1, 1 is no more, and no less, useful than 5, 4, 9, 2.
> >>

> > If the sequence is 1,1,1,1 you need to scan a random sequence of about
> > 10000 digits to get that result. Not impossible.
> > Move to 1,1,1,1,1,1,1,1,1,1 and you need 10^10 digits. Too many for a
> > human to scan, ergo the numbers were not taken from a random sequence.

>
> Not necessarily true. Consider the possibility that the domain is { 0,

1 },
> in which case you would need considerably fewer digits. Or consider

another
> possibility, that the RNG just happened to generate that sequence early.
> You can't tell just by looking.

Yeah, but show us how that RNG will produce 5, 4, 9, 2. But wait, I already
know the answer now - no one claimed that the same RNG was used to produce
1,1,1,1 as 5,4,9,2... yeah.

What I'm learning from this discussion is that whats truly random about
rand() is the number of twisting offshoot threads that are produced when it
is brought up as a topic on clc. =)

At least I know what glark means.