Perplexing proof

[Major Variola (ret)]

At 08:23 AM 9/10/04 -0400, R. A. Hettinga wrote:
><http://www.vnunet.com/print/1157970>
> Perplexing proof
>
>E-commerce is only one mathematical breakthrough away from disaster
>Robert Valpuesta, IT Week 09 Sep 2004
>
>The fact that even experts often do not fully understand how IT systems

>work was underlined by recent reports that the Riemann hypothesis,
>established in 1859, may finally have been proved.

This doesn't follow.

>It seems the hypothesis would explain the apparently random pattern of
>prime numbers that form the basis for much internet cryptography, used
for
>e-commerce and online banking to guard accounts and credit card
details.

Can someone explain how finding regularity in the distribution of primes

would affect any modexp() system?   Suppose that you have a function
F(i) which gives you the i-th prime.  Since the PK systems (eg RSA, DH)
use *randomness* to pick primes, how does being able to generate
the i-th prime help?

>Louis de Branges, a renowned mathematician at Purdue University in the
US,
>has claimed he can prove the hypothesis. But the maths is so
complicated
>that no one has yet been able to say whether his solution is right.
>
>"[The suggested proof] is rather incomprehensible," professor Marcus du

>Sautoy of Oxford University told The Guardian, adding that if correct
it
>could lead to the creation of a "prime spectrometer" that would bring
"the
>whole of e-commerce to its knees overnight".

Methinks the "expert" du Sautoy is an expert in number theory, not
crypto...

>Unfortunately, most managers have no way of telling whether the proof
is
>right or its implications are indeed as stated.

Most managers don't understand crypto.


This could be an
>embarrassment if they are asked to assess risks for corporate
governance
>reports, since they clearly now have a duty to own up and admit that
>business could be threatened by a theoretical prime spectrometer.
>
>Alternatively they might accept that security is a matter of faith,
declare
>that nothing can truly be "known", and add that the way of Zen shows
that
>security is probably an illusion anyway.

I think this latter indicates the cluelessness of the author.