in reply to Re: Markov Chain Program
in thread Markov Chain Program
I'm currently implementing one right now (looking at as few others as possible) that is just words - boring and makes many mistakes.
But I'm planning on rewriting it shortly after being done and will be doing it with digraphs (two character pairs, including punctuation and spaces, although not likely line breaks) after learning how nice those can be for old-school cryptography - the digraphs should theortically be better than trigraphs and mono (single letters).
just a thought.
(my current problems lie less in the above theory or programming - all very easy - and in the way I strip it and from where... trying several boards, as well as doing it in an amusing way in newsgroups as well)