Re: manual cryptography

From: Phil Carmody (thefatphil_demunged_at_yahoo.co.uk)
Date: 10/29/03 

Date: 29 Oct 2003 22:48:46 +0200

jh113355@hotmail.com (John Hadstate) writes:
> Neko <neutral_insomniac@yahoo.com> wrote in message news:<bnmp5m$12rc7p$1@ID-198243.news.uni-berlin.de>...
> > What is the best paper and pencil method other than the one-time-pad of
> > encrypting a message without spending more than a minute per letter
> > encrypting?
>
> Have you considered Vigenere's original autokey cipher?
>
> Given an alphabet for plaintext of 26 letters (A-Z), choose a random
> letter from the same alphabet as the starting point of the key. Then,
> for the rest of the key, append the original plaintext to the starting
> letter. Add the plaintext letter values to the corresponding key
> letter values mod 26 to get the ciphertext letter values.
>
> Thus:
>
> Send reinforcements
>
> becomes:
>
> SENDREINFORCEMENTS <plaintext>
> QSENDREINFORCEMENT <Q + plaintext>
> IWRQUVMVSTFTGQQRGL <ciphertext>
>
> To decrypt, take the value of the first ciphertext letter, add 26,
> subtract the letter value of the key character and convert the result
> mod 26 to a plaintext letter (A-Z). This plaintext letter also
> becomes the next key character. Repeat until all the ciphertext is
> consumed. If this is too slow or error-prone, you can build a
> cardboard "slide rule" that will help you add the letters mod 26.
>
> Obviously, the example can be attacked by testing all 26 letters of
> the key to see which one yields readable plaintext. However, if you
> use two random characters for the key, the attacker may have to try
> 676 combinations. If you use 3 random letters for the key, he may
> have to try 17,576 combinations. So how long do you want to protect
> your plaintext, and from what kind of computing power?

If it's keyed with text, then can't you just throw random text at it,
and investigate anything that looks like real text that emerges.
e.g. try placing 'the' everywhere, and follow the sensible trigraphs?

Phil 

-- 
Unpatched IE vulnerability: dragDrop invocation
Description: Arbitrary local file reading through native Windows 
             dragDrop invocation.
Reference: http://msgs.securepoint.com/cgi-bin/get/bugtraq0302/12.html
Exploit: http://kuperus.xs4all.nl/security/ie/xfiles.htm

Relevant Pages

  • RE: PGP scripting...
    ... no matter your key size (if the encrypting key is known; ... >> for encryption of bulk data. ... it is referring to the actual plaintext (subject to any ... beyond brute-force viability with current hardware, ...
    (SecProg)
  • Re: cryptanalyzing bitwise tramps?
    ... plaintext bit vector, and M is a NxN binary ... the N^2 unknowns, ... You can attack each of the ... For example, encrypting ASCII ...
    (sci.crypt)
  • Re: SPES (my new encryption) one of its kind
    ... the first step into encrypting any thing in my ... |> system is to encrypt it with AES first ... | contents and the length of the plaintext itself ... Middle finger poised over the plonk button... ...
    (sci.crypt)
  • Re: Passwords
    ... Or is it just a case of encrypting the password you ... >> supply and comparing it to the encrypted version in the Registry, ... some kind of one way hash of the plaintext password. ...
    (borland.public.delphi.non-technical)