Category talk:Cryptography

Shouldn't this category be called Cryptology? Lapo Luchini 20:47, 12 Dec 2004 (UTC)

This question pops up from time to time (along with the naming of cryptography, list of cryptography topics etc). The argument is that there are (at least) two ways of describing the subject. The first is to describe the entire field as "cryptology", divided into "cryptanalysis" and "cryptography". The second is to describe the entire field as "cryptography". I think the second style is preferable simply because it's much more commonly used. — Matt Crypto 00:38, 13 Dec 2004 (UTC)

More Subcategories?

Shouldn't 'Classical ciphers' 'Stream ciphers' and any other categories for ciphers be subcategories of Cryptography? Hugo999 00:59, 14 November 2007 (UTC)

Block ciphers, too? There are lots of great categories out there that are not linked to Cryptography. --WiseWoman (talk) 10:02, 19 October 2008 (UTC)

Alice and Bob - the theory

Any chance of a section on the theory of Alice and Bob, where things can go wrong, and how? Maybe a guide to tests one can perform to establish what is going wrong and where? --JeremyJones56789 (talk) 11:39, 9 January 2008 (UTC)

We have an article on Alice and Bob. I'm not sure I understand what you mean by "theory"?--agr (talk) 11:53, 9 January 2008 (UTC)

Quicker manual one time pad generation

Perhaps you could suggest using a bingo machine for manual generation of a one time pad? Depending on how secure it needs to be (theory, or a guide would be handy) you can rattle balls out at about 11 seconds per character - much faster than dice or scrabble, untouched by human hand, absolutely unambiguous. Can use original numbers with conversion chart or make your own balls out of, eg: FIMO plastic. See mine on my web site, with example set of balls, sample OTP, and stats: [1] --JeremyJones56789 (talk) 11:39, 9 January 2008 (UTC)

A bingo machine as normally used wouldn't be suitable. As I understand how they work, there is a fixed number of marked balls that are extracted one at a time. So if you had, say three of each letter in the ball pool, there could be no more than three consecutive same letters in a string, for example. That is not fully random. You could put each ball back after you recorded it and let them mix, but that would slow things down a lot. Besides, how many people own Bingo machines?--agr (talk) 11:53, 9 January 2008 (UTC)

The best way to generate one time pads manually, and fairly quickly, is to roll a handful of 10-sided dice. Say you throw 5 dice, that gives you a 5 number group. With a bit of practice, you can easily do a 5 number group every 5 or 10 seconds. Ten-sided dice are readily available at gaming stores. I've tried a number of different manual ways to generate one time pads, and using 10-sided dice is absolutely the quickest, most convenient, and most secure way. Don't use regular dice, as messages encrypted with pads generated by conventional dice are susceptible to attacks where you 'guess' at a chunk of plaintext, and 'slide' it through the ciphertext looking for places where you get a possible result (ie., ciphertext-plaintext = 1 through 6) and to weed out areas where the plaintext isn't possible (ciphertext-plaintext = 0,7,8, or 9, which can't occur on pads generated by conventional 6-sided dice). Dittybopper (talk) 15:19, 27 April 2011 (UTC)

Hashing

I was looking through the links on Cryptography and was wondering why there are little to no resources regarding Hashes. If someone could tell me why this is so or rectify this that would be great. — Preceding unsigned comment added by 202.72.161.115 (talk) 01:58, 14 June 2013 (UTC)