Prev: Primitive tri/pentanomials
Next: solutions manual
From: Mok-Kong Shen on 7 Aug 2010 13:19
Ivan Voras wrote:

> I'm not nearly an expert but that is, as far as I understand it, at the
> heart of it. A chaotic system implies both non-periodicity in the strict
> sense - you don't *ever* get exact repeat of a long enough sequence -
> and an "attractor", mostly meaning that, as a pattern-seeking minds, we
> clearly see some type of patterns in the data set. It's exactly what
> happens with multi-body physics: we "see" planetary orbits and can
> approximate planets' positions in the future but we can never calculate
> 100% exact (down do individual atoms) positions.
>
> I think that this difference between what is approximated and what is
> "exact" could be a good source of true randomness but probably not easy,
> as the "attractor" part of the whole thing means there can be false
> periodicity if measured with insufficient precision (for the planetary
> example: if we allow loss of precision we can arrive at completely
> predictable orbits, useless for randomness, but on the other hand,
> unfortunately, we cannot measure planetary positions to within
> atom-sized levels of precision and use that for a source of randomness).

I surmise that chaos theory could well have some substantial problems,
at least when applied to crypto without corresponding care. However,
IMHO it seems desirable that some crypto experts could find some little
time to look at certain publications in established journals on
application of chaos theory to crypto. I mean that, should a bunch of
papers in such reknown journals as Phil. Trans. R. Soc. turn out to be
in fact all essentially wrong, clearly pointing that out to the public
(the readers) would certainly be a good and valuable service to the
sciences.

M. K. Shen

First  |  Prev  | 
Pages: 1 2 3 4 5
Prev: Primitive tri/pentanomials
Next: solutions manual