Jinx,

On Thu, 30 Oct 2003 23:23:20 +1300, Jinx wrote:

> > Each agent had a known pattern of spelling mistakes, and
> > often the British intelligence found that the errors did not
> > occur, so assumed the Germans had captured the agent.
>
> Ah, sneaky. That type of human failing was an important aid
> to Bletchley Park's deciphering team. Enigma itself was a
> pretty robust encryption system but the human operators'
> carelessness and complacency was its undoing

Indeed - I was there only a couple of weeks ago, and it turns out that the Germans' belief that it couldn't be
cracked because of its 150 million million different settings - ten million times as many as the UK lottery
has combinations (I may have missed a factor of a million there, but you get the drift!) meant that they
didn't bother with some of the procedures, like picking random starting settings for the three rotors.  They
would often use the same ones, so giving a way in.  Similarly, they often started particular messages the same
way, or ended them with "Heil Hitler" so giving a "crib" - a guess at the plaintext, which meant that you'd
know when you'd found a match.

The word-scrambling that we were talking about would have made cracking Enigma almost impossible because it
wouldn't have appeared to be German even when they did get it right, but the problem is that as with most
cyphers the word-spacing wasn't transmitted, with letters sent in groups of five so the recipients would
probably not have been able to read it either!  :-)

tihs is waht i maen aobut not bneig albe to raed it:

tihsi swaht imaen aobut notbn eigal betor aedit

as opposed to:

thisi swhat imean about notbe ingab letor eadit

If they'd encrypted a space as well, word-scrabmling would have worked.

Cheers,

Howard Winter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics