In a recent posting on the Cipher Mysteries blog, Nick Pelling provides a list of contemporary cipher challenges he collected from the Web. One of those listed is by one Andrew Fergus, who posted his own supposedly unbreakable cipher which apparently uses a pseudo-random number generator for encryption. As Mr. Fergus proclaims, “I have spent some time developing a cipher, which I genuinely believe is unbreakable”.
What left me in a state of disbelief was the following gem found on Mr. Fergus’s site:
“I am familiar with Kerckhoff’s principle which states that a cryptosystem must be secure, even if everything is known about the cryptosystem, except for the key. This cryptosystem employs secret keys in accordance with Kerckhoff’s principle. However, nowhere does Kerckhoff’s principle state the methodology employed must be provided to cryptanalysts, so please don’t ask for this. “
Gee, if the cryptosystem employs keys in accordance with Kerchoff’s principle, would he care to tell us what the keys are? Alternatively, the methodology can securely be divulged to anyone (including my Aunt Matilda) but please don’t tell it to cryptanalysts.
It gives new meaning to the term “cryptographically-challenged”. Needless to say I will not be wasting a single second of my time on this cipher.