[extropy-chat] the structure of randomness

[gts]

On Sun, 01 Jan 2006 13:36:09 -0500, Russell Wallace  
<russell.wallace at gmail.com> wrote:

> On 12/31/05, gts <gts_2000 at yahoo.com> wrote:
>>
>> The distributions do not deviate in any significant way from normal, but
>> the *fine structures* of multiple histograms seem correlated in time.  
>> For example a bell-curve with a slightly "m-ish" shape (two subtle  
>> peaks) will tend to appear again in the next test. That "m-ish"  
>> propensity then falls off with time.
>
>
> How is that different from considering that batch of tests to be a single
> test, adding all their samples together, and saying the aggregate has an
> "m-ish" shape?

Assuming the effect is real, it pertains only to comparisons of two or  
more histograms.  If radioactive decay happens randomly then one would not  
expect to see similarities in the fine structures of the histograms of two  
successive samples.

It is possible (maybe) that the effect is real but that the fine  
structures of these histograms, though correlated through time,  
nevertheless change rapidly enough that larger samples would not reveal a  
deviation from normal.

You can think of the effect in terms of the graphic visualization mode in  
computer media players, like Windows Media Player or iTunes. You can "see"  
music as it plays. The pattern changes constantly. However, because music  
has structure, similarities exist in the pattern from one moment to the  
next. The similarities fall off as a function of time. If the noise were  
completely random then you would see no such similarities.

If the Shnoll effect is real then it is truly "The Music of the Spheres."

> (I might be overlooking something obvious here; there's
> *mumblety* years of rust on my knowledge of statistics.)

Same here. :)

For what it's worth, I found a website published by Jack Sarfatti on which  
he states his opinion that John Walker refuted the Shnoll effect to his  
satisfaction. Not sure Shnoll agrees. According to the article posted by  
Damien, Shnoll is planning to publish another paper on the subject.

-gts