Why "carnivore" type systems can't be (entirely) open source

[David Hansen]

On 30 Jan 01, at 21:44, Owen Lewis wrote:

>>I would
>>also add that the Lorenz machine was broken (by the Swedes 
>>and the  British) without knowledge of what sort of encryption system was
>> used, beyond an educated guess that some sort of wheel was 
>>involved and that the machines operated on teleprinter codes.

> In the example you give, STO failed because it was easy ( hyperbole or
> litotes here?) to break with no prior knowledge of the system.

Beyond the educated gueses mentioned AFAIK the British had no 
knowledge of the system when they broke it. What they of course 
did have was a knowledge of the current state of the art and what 
developments were likely. Later on some machines were captured.

> As argued in another post, Enigma was not so easy

I would disagree. Enigma remained breakable with electro-
mechanical machinery and the use of cribs (though I concede these 
were through standardised headings known to the British).

The Lorenz was far harder to break for many reasons, starting with 
the difficulty of intercepting the data stream which used directional 
radio beams and then breaking up the continuous stream into start 
stop and data bits. The British did not spend a lot of money creating 
the Colossi eor the fun of it.



 David Hansen | davidh@spidacom.co.uk  | PGP email preferred
 Edinburgh    | CI$ number 100024,3247 | key number F566DA0E