[1887] in cryptography@c2.net mail archive
summary Re: testing your RNG
daemon@ATHENA.MIT.EDU (Zooko Journeyman)
Tue Nov 25 01:38:12 1997
Date: Tue, 25 Nov 1997 00:22:35 +0100 (MET)
From: Zooko Journeyman <zooko@xs4all.nl>
To: cryptography@c2.net
Thanks to the P'sCL crowd for the randomness testing resources!
I enjoyed the comments and followed the references. In sum:
David Wagner's page
"http://www.cs.berkeley.edu/~daw/netscape-randomness.html" is
quite useful. Eric Blossom pointed out statistical tests in
FIPS 140-1. Nick Szabo mentioned that it was impossible.
(;-)) Antonomasia pointed out rfc1750. Dmitri M. Vulis
recommended that I ask the cypherpunks. (;-)) Everyone agreed
that Knuth is a good start (though of course Knuth doesn't
address cryptographic issues).
A couple of people pointed out that the ad hoc tests I posted
were crummy-- I know! I didn't write them-- I would never
write anything so sloppy. But a wise person once said that
the fastest way to get information from the net was not to ask
but to post a wrong answer. :-) In other words, I thought it
would be somewhat rude of me to post a bare question without
writing anything to get the ball started. You can consider
that code snippet as a datapoint in your perspective on the
state of the art in our industry.
Thanks, all.
Zooko
Software engineer for hire. Have computer, will travel.
http://www.xs4all.nl/~zooko/resume.html
