http://www.geocities.com/SiliconValley/2151/huffman.html
Now with the understanding of Huffman Compression, you have the understanding that you need the Frequency Table before you can uncompress or decode which either you seem fit & depending on the programming which is doing the the 'encrypting' & 'decrypting' the Frequency Table could be sent differently to the location via different means than the encrypted document. If you don't get me I like you could keep the frequency table on a USB device, and email the document to another email account for safety reasons of course.
Now the programming I am creating, which I am wondering if I would even be able to create, which I know dependent upon my programming skills and dedication.
Cheers for hanging in their,
Right was encoded into the tree, their will always be a 1111111 of a certain size, this size will depend on the size of the document, how I'm thinking I would be able to figure out the maximum size of the largest 111111 thing by grabbing the largest 11111111(which mean not be the largest could be other number mixed in) and the look for another second longest 111111 but one which ends with a 0 instead of a 1. you getting me so far, and from this we have the far right part of the table and can start to decode the tree into its child segments, now if we make the same decrypting program to start to randomize the letter of which it may be, you will get a pattern of where letters are turning up and other sorts. Now add a dictionary file, which you would do for brute force dictionary attack, and start randomizing until more words are making sense...
of course you would also have to have it print out all copies of randomizing so that you could help it along the way, so you would have a key which stops the randomizing, and then some commands so that you could hardwire some letters to certain (numbers).
Sorry if its not making sense I'm much better at thinking ideas than actually writing them down.
