An algorithm for random data compression

Hello Hacker News, I'm sharing some work I've been doing as a side project. It's an algorithm that attempts to do the impossible -- compress random data. My results seem to indicate that I've succeeded at this task -- but I may be wrong.

Feel free to check out the post and tell me what you think of my methods. It directly attacks the problem in a new way and allows data to be mapped without running into the pigeon hole problem. Again - I may be wrong about this - but I've done my best to verify everything using my small personal cluster and everything seems to add up.