stuart johnson writes:
a co-worker here has come up with a 'cypher' scheme that he would like to use to send code to our clients. the scheme is this : he would take the file of code and pad all lines to the length of the longest line, he would then preform column swaps, and then row swaps, to 'mix up' the file. the person receiving the file would then preform the opposite functions to recover the file. it seems so simple that it can't be good. i've convenced him to use pgp, but i would like some input if possible on why his cypher scheme is not a good one.
This is a variant on a scheme called a transposition cipher. It was okay, but not very good, technology in the Civil War, when it was last seriously used. It can be broken with a technique called multiple anagramming. Perry