outer@utcsrgv.UUCP (Richard Outerbridge) (04/30/84)
Last year the Polish government issued a stamp commemorating the 50th anniversary of the breaking of Enigma (using pencil and paper) by Rejewski, Rozycki and Zygalski. Noting that UNIX crypt(1) is a rotor-based system, just how secure can it be? -- Richard Outerbridge <outer@utcsrgv.UUCP> 416 978 2742
henry@utzoo.UUCP (Henry Spencer) (05/01/84)
To quote from crypt(1):
...Methods of attack on such machines are known, but not widely;
moreover the amount of work required is likely to be large.
It is thought that rotor systems no longer hold any great mysteries for
organizations like the NSA, but a rotor system would still be a major
challenge to a non-professional. I've seen an outline of a method for
attacking such systems. It's not trivial. At the very least, breaking
crypt(1) would require a knowledgeable cipher enthusiast, quite a bit of
encrypted material to work with, and a lot of effort.
--
Henry Spencer @ U of Toronto Zoology
{allegra,ihnp4,linus,decvax}!utzoo!henrygwyn@brl-vgr.ARPA (Doug Gwyn ) (05/02/84)
The answer is that UNIX crypt(1) is not very secure, as its manual page warns! Any competent cryptanalyst should be able to crack a crypt(1)ed normal text of length ~2000 chars in a matter of minutes, assuming he has his usual set of tools at hand. P.S. No, I am not issuing a challenge. I do not have even a few minutes apiece to spend cracking test messages, so please don't mail me any!
bae@fisher.UUCP (The Master of Sinanju) (05/03/84)
Hmmm....
Anyone want to create a public-domain decrypt(1) for us??
--
Brian A. Ehrmantraut
When the going gets weird, the Weird turns Pro!
{ihnp4, decvax, ucbvax}!allegra!fisher!bae