[Courses] [security] Crypto Scientists Crack Prime Problem

Raven Alder raven at oneeyedcrow.net
Thu Aug 15 13:42:38 EST 2002

Heya --

	Thanks for all the detailed information; this is interesting.
Most of the crypto stuff I've done has been implementation based; it's
neat to see the academic/historical side of it.

Quoth Jacinta Richardson (Wed, Aug 14, 2002 at 10:59:06AM +1000):
> This [being sure that the large primes actually are prime] has always
> been a problem.  That doesn't change because we've got a new primality
> test.  It would change lots if we could efficiently factor the damn
> things.  ;)

	I slightly disagree; there are high-end folks looking at
brute-forcing keys, which generally does involve factoring the numbers.
(For example, the Bernstein paper on parallel cracking --
http://cr.yp.to/papers.html, it's the first one under "Integer
factorization").  Assuring that numbers are prime before they're used in
an RSA-like algorithm would change setups like this.
> We should be overjoyed about having a new method to test for primality.  
> It might give us some interesting insights into new factorisation methods.  
> PS: Lots of the text of this email was pulled out of a computer science
> essay I had to write last year.  I can provide code or algorithms for all
> theories and am happy to provide such for generating RSA strong primes if
> anyone wants them.
	I'd like to see it -- could you throw out a link?

"Do you know where the RSA t-shirt is?"
"Well, I need the algorithm, so I'm doing laundry."
  -- me and RavenBlack

More information about the Courses mailing list