Xref: utzoo rec.ham-radio:8662 sci.electronics:5159 sci.crypt:1603
Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!ulysses!smb
From: smb@ulysses.homer.nj.att.com (Steven M. Bellovin)
Newsgroups: rec.ham-radio,sci.electronics,sci.crypt
Subject: Re: Short Wave Numbers "Racket"
Message-ID: <11224@ulysses.homer.nj.att.com>
Date: 15 Feb 89 14:42:39 GMT
References: <439@brambo.UUCP> <1349@atari.UUCP>
Organization: AT&T Bell Laboratories, Murray Hill
Lines: 21

In article <1349@atari.UUCP>, mn@atari.UUCP (Mike Nowicki) writes:
> I did a computer analysis of 3 of these types of stations and
> found that strings of 3 or 4 letter or number pop up quite often and might
> signify use of a lookup pad code where for example, "1AC" can be looked up
> in your handy dandy copy of Popular Spy Code monthly, and find out it means
> "get photo of". These codes are impossible to decode because of the one to
> one relationship of strings to phrases.

It sounds like you may be describing a code, as opposed to a cipher.
Codes, especially superenciphered codes, are much more difficult to
crack than ciphers, but it's been done many times in the past.  For
example, the course of both World Wars was significantly affected by
cracked codes (as well as cracked ciphers).  During World War I, the
British solution of the so-called Zimmerman Telegram was a major factor
in the entry of the U.S. into the war; during World War II, the U.S.
was reading the Japanese naval code.  (This is distinct from PURPLE,
the top-level Japanese machine cipher.)  As usual, see Kahn for details.

Mind you, I don't know that those messages really are code, but if they
are cracking them is not impossible, especially under reasonably favorable
circumstances (probably texts, etc.)