[Cryptography] Generating random values in a particular range

[Ray Dillinger]

This is dumb, and if they ever try to extort money from anyone based
on this classic algorithm, shooting them down will be horribly easy.

If the court so desires I can show them a computer science textbook
printed in _1982_ that specifically gives that very algorithm and
specifically warns against the modular construction as being biased.  I
think I call that "prior art."

In following pages it even goes on to show how to calculate the exact
degree of bias so you can tell within engineering constraints when the
modular construction is acceptable.  Then it gives a six different tests
for predictability that many common pseudorandom generators at the time
were failing, four 'whitening' algorithms (one of which even looks a lot
like cryptographic hashing) to extract bit streams passing those
statistical tests for predictability given bit sources that do not, a
couple of pseudorandom generators that were passing all the tests known
at the time and comparisons of their efficiency, and algorithms for
calculating the period of lagged-fibonacci and LCG type generators.

Did RIM try to patent all of those basic well-known algorithms, too?

				Bear



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20160806/acb72513/attachment.sig>