The Two Kinds of Random: Why π Compresses and Noise Doesn't — and the Proof Barrier Between Them

Summary

This article explores a deep puzzle in mathematics and information theory: two files of a million digits each — one pure random noise, the other the first million digits of π — are statistically identical (both pass randomness tests), yet one (π) can be compressed to a few lines while the other cannot. The author uses this gap to dissect the concept of 'compressible' into two distinct meanings: compressible in practice (algorithmic) versus compressible in theory (Kolmogorov complexity). The piece traces through randomness, entropy, statistical testing, and the philosophical wall that you can always prove something is compressible but can never prove it isn't — a boundary that touches on Gödel-like incompleteness and the limits of mathematical proof.

Source

Hacker NewsThe Two Kinds of Random: Why π Compresses and Noise Doesn't — and the Proof Barrier Between Themstillthinking.net