This article has not yet been rated on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Cryptography: Computer science This article is within the scope of WikiProject Cryptography, a collaborative effort to improve the coverage of Cryptography on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.CryptographyWikipedia:WikiProject CryptographyTemplate:WikiProject CryptographyCryptography ??? This article has not yet received a rating on the importance scale. This article is supported by WikiProject Computer science.

I reverted a bad edit with the incorrect cipher text. Here's some basic python code showing the calculation

>>> f=lambda (x,y):chr((ord(x)-ord('a')+ord(y)-ord('A')) % 26 + ord('A'))
>>> f(('f', 'E'))
'J'
>>> map(f,zip(list('fleeatonce'),list('ERRORSCANO')))
['J', 'C', 'V', 'S', 'R', 'L', 'Q', 'N', 'P', 'S']

Note I only had to revert the bad first four characters of the cipher-string; the rest was ok originally. Jimbobl (talk) 14:44, 7 September 2011 (UTC)Reply

I'm not seeing any other reference to BDA different from it outside of this article. Does anyone know where that came from? Rusca8 (talk) 16:36, 20 October 2019 (UTC)Reply

Running key ciphers (sometimes also referred to as "book key worms" in Germany) are inherent to the language in which the respective book is written: the typical, hardly changeable letter frequency of the individual letters of the language (and the word spacing, if these are also encrypted, then referred to as "informal ciphers"). This statistical weakness can be circumvented if the book text used is first polyalphabetically pre-encrypted to the desired length, preferably with an encryption device such as M-209 or Enigma. The resulting running key has the significant advantage that a) it is not periodic, and b) it is evenly distributed in terms of letter frequencies. A frequency analysis no longer makes sense. The pre-encrypted book text running key (which is not available to unauthorised decryptors) is used as the key for the subsequent polyalphabetic substitution of the plaintext. Of course, not with the tabula recta and the header ABCDEFG..... (Vigenère), but more cleverly with an encryption table, which is a Latin square, but extremely scrambled. The rows and columns of this square cannot be shifted into one another. There existe perhaps 10^320 different latin squares of 26*26 letters. Because it is possible to permute rows and columns of shifted alphabet squares, there are (26!)^2 = 1,6*10^53 different latin squares with shifted rows and/or columns.

There is no doubt that the "book cipher" is a homophonic substitution, but it has all the prerequisites to classify the procedure as a homophonic "book code" rather than a "book cipher", in which the code word is also a "code vector" consisting of page number, line number and number of the word within a line. The running key cipher, on the other hand, is a real polyalphabetic substitution cipher. However, it has the disadvantage of having the not equaly distributed letter frequencies as in every natural language. The one-time-pad eliminates this disadvantage not only through equal distribution, but also through a real random sequence of characters in it, which has no inherent periodicity and which cannot be described by a mathematical algorithm that is shorter than the corresponding one-time-pad. The frequency of double or multiple characters (...BB..., or ...VVV...) must also fulfil the criteria for the occurrence of such character sequences in random texts. Permissiveactionlink (talk) 15:23, 21 July 2025 (UTC)Reply