In Conversation With
Museum of Minds
CS–1916–2026
A MATHEMATICAL THEORY OF COMMUNICATION Bell System Technical Journal · 1948 101 100 101 110 011 010 100 110 101 101 001 011 010 011 010 001 100 101 011 001 010 010 110 100 INFORMATION SOURCE M m TRANSMITTER s(t) x CHANNEL p(y|x) NOISE SOURCE n y RECEIVER m̂ H(X) = −∑ p(x) log₂ p(x) SHANNON ENTROPY the measure of information, uncertainty, and surprise C = max I(X ; Y ) = max [ H(X) − H(X|Y) ] CHANNEL CAPACITY THEOREM H p 0 ½ 1 1 H(p) BINARY ENTROPY FUNCTION C S/N C = W log(1 + S/N) SHANNON LIMIT Bell Labs hallways, c. 1950 001 011 010 100 110 101 100 101 011 001 101 010 FIG. 1 — SCHEMATIC DIAGRAM OF A GENERAL COMMUNICATION SYSTEM
Mind on Display
Claude
Shannon
1916 — 2001
Portrait of Claude Shannon
10110010 01101001 10010110 01001101 10100101 01011010 11001011 00110100

Portrait Not Found

He quantified the unknowable — but spent his life playing with things that had no point.

Claude Shannon invented the mathematical theory of communication in 1948, single-handedly founding information theory and making the digital age possible. His paper is among the most consequential ever written.

A Bell Labs polymath, Shannon built chess-playing machines, rode unicycles through hallways, and juggled while solving problems that reshuffled the foundations of mathematics. He saw no contradiction between rigorous abstraction and pure play.

He defined information not as meaning, but as surprise — the measure of what you didn't already know. The concept liberated communication from content, and gave engineers something they could actually count.

10110010
Begin Transmission · Engage the Mind
01001101

Converse with
Claude Shannon

Shannon's 1948 paper arrived not as engineering but as mathematics — a proof that any channel has a maximum rate at which information can be transmitted reliably, regardless of the noise. The bit was born in these pages, and with it, every hard drive, every internet packet, every compressed image and encrypted message.

But Shannon was equally famous for what he didn't do: he rarely sought applications, avoided conferences, and reportedly grew uncomfortable as information theory became fashionable. He was, above all, a man who found problems beautiful when no one else could see them yet.

"I visualize a sea of problems, each with a solution waiting to be found. There are many different ones, and they're not very connected with each other." — Shannon, on his working method

Ask him about entropy and meaning. About whether neural networks are doing information theory or just statistics. About what it felt like to write the paper that defined the modern world and then go build a machine that could solve the Rubik's cube.

Information Theory Entropy Channel Capacity Coding Theorems Boolean Algebra Cryptography Play & Abstraction
It is not the business of science to explain why the universe is understandable. It is enough that it is.
— Claude Shannon · Collected writings

Knowledge Corpus — Primary Sources