[Milsurplus] German decrypts

[Mike Andrews W5EGO]

On Sun, Oct 05, 2008 at 08:47:46AM -0400, J Forster wrote:
> Could the Germans have been reading low-level or commercial codes? I
> would seriously doubt the really secure stuff would have been widely
> used as doing that would have greatly increased the risk it would be
> compromised.

Commercial codes? Certainly. In the US, a copy of each commercial code
had to be deposited with some designated federal agency, and I suspect
that most other countries that permitted the use of commercial codes 
had similar requirements. Most of them were available over-the-counter
to anyone who wanted to use them, and the remainder (company private 
codes) had the depository requirement. 

Low-level? I have a training edition of a division-level field code
(<http://mikea.ath.cx.codebook/>). It's a two-part code, which I've
never seen superencipherment tables for, though Para. 7 mentions
that the numeric groups are for use only with superencipherment.
It wouldn't be terribly hard to get pretty well into the bare code, 
given enough traffic, good traffic analysis, and maybe a little luck.
Superencipherment would make it rather more dificult. 

Lower-level stuff, such as company- or battalion-level tactical       
codes, or low-level naval codes, probably would come easier, but      
would be used to carry traffic that wasn't needed to be secure beyond 
a few days or a week.                                                 

The question of how widely used a code can be while still remaining 
secure is far too complex for discussion in one letter, or even in a 
small book.

-- 
Mike Andrews, W5EGO
mikea at mikea.ath.cx
Tired old sysadmin