MH
Michael Harrington
Systems, engineering, analytics, and applied strategy
Content Page

Data Compression

At this point it is more "tinkering" than a functional system. I can make a pre-sort with an amazing focus on 1's or 0's, where they are sorted in two or more groups, with the front groups tending to be more significantly biased and the back groups trending to be more balanced. Text documents I can usually exceed 95% to 5% ratio with ease for up to half of the file. Especially on some text files.

Pre-Sort Issues

The problem with the success I have, is that modern data compression uses predictive analysis, we know that often after a specific character that there is odds of what comes next. Huffman Encoding, and other mathematical functions, need a slightly higher rate to produce good numbers, but this still ignores the back third to half which is essentially RAND at that point.

I would be willing to team up with someone just to make this a possible system, the numbers are so unique that there has to be some value here,

The Basics

A dual mechanism pre-sort that can hyper focus results as desired.

Optional Bullet List

  • The first part sets up an extreme amount of 0's or 1's in front.
  • The second part then splits them successfully, in a lossless manner.
  • Repetitions of either will hyper focus the trend, further increasing the bias in the front portions while the back continues to grow in total bits that are essentially RAND.
  • There is some ways to alter the results, but the amount of time it would take to examine all is extensive. I just do not have the time to pursue all routes.
Placeholder area for extra content, image blocks, tables, callouts, or section-specific notes.