Challenge: Letter Frequency Statistics
Can you take the average frequency of letters in the English language and create a formula or algorithm that compares them with some decrypted text, to see the percentage of which it is correct? That way if your program prints out a 70 or so percent match, you can give it the time of day to see if it's correct.
I'm going to try and use Spearman's rank correlation coefficient, though I'm interested to see other people's approach.
"Computer science is no more about computers than astronomy is about telescopes."
~ Edsger W. Dijkstra