Other Directory Sites: SeekWonder | Directory Owners Forum
The following archive was created by hippie-mail 7.98617-22 on Fri Aug 21 1998 - 17:16:58 ADT
Werner Koch (wk@isil.d.shuttle.de)
Sun, 26 Apr 1998 15:34:37 +0200
[ This is from the engineers point of view and
not that of a mathematician ]
Thomas Womack <thomas.womack@merton.oxford.ac.uk> writes:
> That's just not true; consider the Jacobi-sum primality-proving techniques,
> or ECPP, both of which are guaranteed, if executed correctly, to declare
And here is the point: you can prove that an algorithm does a
correct primality test but it is not possible to build a machine
that is always correct - from time to time it happens that e.g.
cosmic radiation flips a bit somewhere in the RAM or the CPU and
this may lead to a wrong prime. Yes, this is rare but the probability
of this effect is higher than a non-prime slipped through several
Rabin-Miller tests.
Sorry for the misunderstanding.
Werner
The following archive was created by hippie-mail 7.98617-22 on Fri Aug 21 1998 - 17:16:58 ADT