History of Cryptography

There is a vast literature on cryptography as this has often caught the public imagination. Interestingly, and different from other fields of science, but not so surprisingly after a little thought, the writers here often do not correspond to the discoverers.

A prime example here is Charles Babbage, famous independently for devising the first working general purpose "computing machine", who broke a more complex version of the Vigenere cipher in the 1840s but never published his results. A solution to a simpler version of the cipher was published some 20 years later, but this did not prevent widespread use of the cipher over the next 50+ years or Scientific American for claiming it was "unbreakable" even later still in 1917. Make sure your information sources are reliable.

Almost all of the famous work on cryptoanalysis during the first half of the 20th century was never discussed by the cryptoanalysts themselves (although some like Alan Turing became famous for their work) but was left for historians to piece together. In many cases this has not been hindered by the closure of government files. Much of the popular surveys here have been written by professional journalists rather than historians.

Exceptions to this trend started to happen in the modern era when the commercial aspects of cryptography became important. This was especially true in the case of theoretical results. A famous example here is the fundamental work by Claude Shannon:
Shannon, Claude: "Communication Theory of Secrecy Systems", Bell System Technical Journal, 28 (4), pages 656-715, (1949).

Some classic writers have been fascinated by cryptography (Allen Edgar Poe, Arthur Conan Doyle and Dorothy Sayers for example) and incorporated ciphers into their fictional (often mystery or "who-did-it") work. Perhaps the earliest occurence is the Greek Iliad.

A classic book, an excellent although lengthy read and widely regarded as highly accurate, is that by David Kahn:
Kahn "The codebreakers" (1967).
Kahn is a military historian but the book is still in the "journalistic history" style (which is why it is a best-seller!!). This is of course completely out of date in terms of modern electronic communication cryptography. Some more modern attempts here by journalists have been heavily criticised for their factual errors and overstatements. (for example, the book by Singh).

You can get started in further exploration by the wiki link:
http://en.wikipedia.org/wiki/History_of_cryptography

Other on-line articles include:
http://all.net/books/ip/Chap2-1.html
But in general one should treat many of these web sites with some scepticism. There is a huge amount of copying of other sources here and parts of it not accurately.

Number Theory Background

There are, naturally, numerous textbooks on number theory. These run the full gamut in terms of level, emphasis and readablity. If you only could pick one then it probably should be the classic:
An Introduction to the Theory of Numbers by Hardy and Wright.
G. H. Hardy was perhaps the finest number theorist of the early part of the 20th century. This of course dates the book, but for the purposes of this class we don't require more modern materials; we won't use any number theory that wasn't known at least 150 years ago!

Algorithms and Cryptosystems

The choice here is easy;
Handbook of Applied Cryptography by Menezes, van Oorschot and Vanstone .

It is dictionary-like in its presentation but within this handbook style it is extremely well done. It should be the first point of reference if you plan on doing a project based on a more in-depth study of a particular algorithm.

Even better, it is available online free of charge courtesy of CRC press.

Some sites have online information and even "black box" codes to help you. The best of these are very good, but for the non-.edu/non-.org sites, Caveat Emptor.

At the opposite end of the spectrum in terms of the popular press, an interesting "field manual" for cryptosystems can be found here:
http://www.fas.org/irp/doddir/army/fm34-40-2/
although this is to say the least, bone dry in its exposition.