don (04/24/83)
Subject: miscellaneous comments on cryptography I don't think the knap-sack methods have been cracked by factoring. You may be thinking of the RSA public key method. RSA seems more secure at the moment, but a lot of cryptographers don't trust any of the public key schemes. Remember, just because you can break the code by factoring numbers doesn't mean you cannot break it another way. Also, many people fear that a probabilistic method for factoring may be discovered that will work, in practice, much faster than a true algorithm. There is a "decrypt" program floating around. We have it here, but I am not interested in distributing it. Crypt is a "rotor machine", and they are strictly old hat. Use DES! Maybe the NSA can break it, but there is no evidence. No one else has been able to make a dent in it (and lots of smart people are trying!). I would be happy to say more about how DES or rotor machines work, but I will wait and see if there is interest. This is not net.cryptography after all.
bhayes@sri-unix.UUCP (07/08/83)
#R:allegra:-125400:sri-unix:2900001:000:196 sri-unix!bhayes Apr 24 13:13:00 1983 Re: Just because you can break a code by factoring... There's a paper by Rabin (sorry, no ref...) in which he shows that cracking a minor variant of RSA is exactly as hard as factoring the keys.