Last time, we watched a young mathematician imagine the simplest possible computing machine. Tape. Head. Rules. Beautiful. Elegant. Completely imaginary. Nobody had built one. Nobody needed to.

The machines they built were enormous, fragile, expensive, and revolutionary. They filled entire rooms. They consumed enough electricity to power small neighborhoods. They broke down constantly. And they changed everything.

Starting in 1936 — the same year Turing wrote his paper — Zuse began building computing machines in his parents' apartment. His first attempt, the Z1, was purely mechanical, built from hand-cut metal plates. It jammed constantly. But the concept was sound.

The German government showed little interest. The Z3 was destroyed in an Allied bombing raid on Berlin on December 21, 1943. Zuse rebuilt and continued working, eventually producing the Z4 — but his contributions were largely unknown outside Germany for decades.

After the war, Churchill ordered most machines dismantled. Tommy Flowers was sworn to secrecy. He received a modest award of £1,000 — not even enough to cover what he'd spent from his own savings. When he applied for a bank loan, he was denied — he couldn't tell the bank what he had accomplished.

In 1943, the U.S. Army needed artillery firing tables — charts telling gunners how to aim. Human teams, mostly women mathematicians officially called "computers," were doing this work by hand. They were months behind. J. Presper Eckert (24, engineering prodigy) and John Mauchly (36, visionary physicist) proposed building an electronic calculator. The Army, desperate, funded the project.

In the 1940s, the word "computer" referred to a person — usually a woman — who performed mathematical calculations by hand. The U.S. Army employed hundreds of women as human computers, calculating firing tables for artillery. When ENIAC was built, the Army needed people to program it. The engineering leadership considered hardware the "real" work. Programming was seen as clerical. So the job was given to six women from the Army's computing corps.

Marlyn Meltzer, Ruth Lichterman, and Frances Spence — all mathematicians who had been working as human computers.

These six women received no manual for ENIAC — because none existed. They were given only the machine's logical diagrams and told to figure it out. They studied ENIAC's wiring blueprints, understood its architecture from the ground up, and invented programming techniques as they went.

They broke the trajectory calculation into discrete steps. They figured out which of ENIAC's 40 panels needed to be configured for each step. They determined the order of operations, managed data flow between accumulators, and debugged a machine with 17,468 vacuum tubes and no error messages. They were, by any definition, the world's first professional programmers of a general-purpose electronic computer.

Think About It: The ENIAC women were left out of history for decades because their work was classified as "clerical" rather than "engineering." Why do you think certain kinds of work get labeled as less important? How does that shape whose stories get told — and whose get forgotten?

John von Neumann was, by the mid-1940s, possibly the most brilliant mathematician alive. Born in Budapest in 1903, he made foundational contributions to quantum mechanics, game theory, set theory, and fluid dynamics — all before age 40. He was famous for his photographic memory, his ability to do complex arithmetic in his head, and his habit of telling jokes at scientific meetings.

The architecture is often called the "von Neumann architecture" because his name was on the published report, but history is messier than that name suggests. The collaborative reality — conversations among engineers and mathematicians, each contributing different pieces — is more accurate than any single-inventor story.

This is it. The design that runs the world. Five components, connected by pathways called buses. Every computer you have ever touched — your laptop, your phone, game consoles, servers — follows this basic layout. Von Neumann described it in 1945. It is still the blueprint.

3. The Control Unit — the conductor of the orchestra. It reads instructions from memory, decodes them, and tells the ALU and other components what to do. It keeps track of which instruction to execute next using a program counter — a pointer that moves through the program step by step.

Think About It: Your phone follows the von Neumann architecture. Open your phone's settings and look at "About Phone" or "Storage." Can you identify the Memory (GB of storage and RAM), Input (touchscreen, microphone, camera), Output (screen, speakers, vibration motor), and CPU (e.g., A17 Bionic or Snapdragon 8)? Apps are the programs stored in memory. Five boxes. Some arrows. Right there in your pocket.

Vacuum tubes worked. But imagine running your house on candles when someone just invented the light bulb. Tubes were big, blazing hot, and burned out every few days. In 1947, three physicists at Bell Labs invented a replacement that was smaller, cooler, faster, and almost never broke. They called it the transistor. It is arguably the most important invention of the 20th century.

A single vacuum tube was thumb-sized, generated significant heat, consumed substantial power, and burned out unpredictably. ENIAC lost a tube roughly every two days — finding which one had failed was a painstaking job. To build bigger, faster computers, engineers needed a different kind of switch. Same function — ON/OFF, 1/0 — but smaller, cooler, more reliable, and cheaper.

William Shockley, who had been researching a different approach, was reportedly furious that Bardeen and Brattain made the breakthrough without him. He checked into a hotel for several weeks and, driven partly by professional jealousy, developed the theory of the junction transistor — which proved superior and became the basis for mass production.

Shockley, Bardeen, and Brattain received the Nobel Prize in Physics in 1956. By the late 1950s, transistors had begun replacing vacuum tubes in computers. The era of room-sized machines began to give way.

In 1958, Jack Kilby — a new hire at Texas Instruments — had an idea during a quiet summer when most colleagues were on vacation. Alone in the lab, with no interruptions, he asked: What if you fabricated the entire circuit — transistors, resistors, capacitors, and their connections — from a single piece of semiconductor material?

Both men are credited as co-inventors. Kilby received the Nobel Prize in Physics in 2000 (Noyce had died in 1990). Kilby once said of his co-inventor: "I'm sure history will give us equal credit."

In 1965, Intel co-founder Gordon Moore observed that the number of transistors on a chip was doubling approximately every two years — a trend that became known as Moore's Law. This prediction held remarkably steady for over five decades. Today, a single phone chip contains billions of transistors — each far smaller than a virus.

By the late 1950s, computing had undergone a revolution in hardware. Turing's imaginary machine had become real. The von Neumann architecture gave computers a universal structure. The stored-program concept meant they could be reprogrammed without rewiring. The transistor made them reliable. The integrated circuit would make them scalable. But the people who used these machines faced a brutal reality: talking to a computer meant speaking its language — binary.

Note: This uses accumulator architecture — the CPU has a special register called the accumulator that holds intermediate results. LOAD copies a value from memory into it. ADD adds another memory value to whatever is already there. STORE writes it back to memory. Each instruction is two bytes: one opcode, one address. Other architectures exist, but this was the most common in early computers.

That bridge was about to be built — by a Navy officer who was told it was impossible, a team at IBM who invented a new language for science, and a generation of pioneers who realized the real bottleneck in computing was not the machine. It was the space between the machine and the human mind.

Think About It: Every layer of computing technology solved one problem and revealed the next. Vacuum tubes proved electronics could compute but were unreliable. Transistors fixed reliability but created wiring complexity. ICs fixed wiring but computing was still locked in binary. Why do you think progress often works this way — solving one problem only to uncover a deeper one?