2 Replies - 1618 Views - Last Post: 17 December 2012 - 02:12 PM Rate Topic: -----

#1 pbl  Icon User is offline

  • There is nothing you can't do with a JTable
  • member icon

Reputation: 8334
  • View blog
  • Posts: 31,857
  • Joined: 06-March 08

Decrypting WW2 code

Posted 17 December 2012 - 12:41 PM

http://www.bbc.co.uk/news/uk-20749632

Come on they say that they cannot decript it ... with all the computer power the Head Quarter of British decripting agencie have !!!

I don't buy that
Is This A Good Question/Topic? 0
  • +

Replies To: Decrypting WW2 code

#2 jon.kiparsky  Icon User is online

  • Pancakes!
  • member icon


Reputation: 7756
  • View blog
  • Posts: 13,116
  • Joined: 19-March 11

Re: Decrypting WW2 code

Posted 17 December 2012 - 01:04 PM

It would be very odd if a collection of random acronyms happened to fall into 5-letter codegroups.
If there's any existing message using these abbreviations, I'd believe it, sure, but I don't think there is.
Considering that cryptography had standardized on 5-letter codegroups in the early part of the century, it's not hard to see that his would have been either a trench code (in which case you'd have to find the codebook, which might be difficult, since there were a lot of trench codes created, or impossible, since it's not likely they would have been preserved) or else a standard cipher of some sort. If it were a standard monoalpha or polyalphabetic cipher, it's long enough that computer analysis would have at least offered some plausible guesses as to the content. Sheer brute force would give you the possible combinations for reasonable-length keys. I'm assuming the Brits are not stupid, so they tried that. That means it's either a mechanical cipher, something of the enigma sort, or a code.
As far as I know the British did not deploy anything like the Enigma, and certainly not to a man on foot behind enemy lines - those things were built on portable typewriters, not something you want to lug around, and it's certainly not something you want captured. So that suggests it was a trench code. I don't know for sure that 5-letter groups were the format for British trench codes, but I believe they may have been - I'll have to check in Kahn when I get home.

If this is correct, and it's a code (not a cipher) then it's going to be impossible to break without the code book, since each five letter group will be a "word" in the code, and we just don't have the material to establish the patterns.


So, in short, no this guy's just making stuff up, he hasn't solved anything.
Was This Post Helpful? 0
  • +
  • -

#3 baavgai  Icon User is online

  • Dreaming Coder
  • member icon

Reputation: 5833
  • View blog
  • Posts: 12,687
  • Joined: 16-October 07

Re: Decrypting WW2 code

Posted 17 December 2012 - 02:12 PM

My first thought when I heard about this was some kind of book cipher. While this is considered "security through obscurity," which modern cryptographers snear at, it can still be nearly unbreakable without the key. Perhaps if you had all the text from all published works and few millenia...

If the encoder is clever and doesn't repeat a reference the pattern wouldn't really be derivable.

I don't find it surprising that something thought up in a world before computers would frustrate efforts of the modern computer centric. I rather enjoy that, actually.
Was This Post Helpful? 1
  • +
  • -

Page 1 of 1