The Off-Key Bard
History

This Day in History — The First Public Demonstration of Remote Computing

On this day in 1940

On September 11, 1940, Stibitz ran math problems at Dartmouth on a computer 250 miles away in New York — and Norbert Wiener spent the rest of the day at the terminal.

The Off-Key Bard sits down at a terminal in New Hampshire and enters a mathematical problem. The computer that solves it isn't in the room. In fact, it isn't even in the same state.

"Today we call it remote access. In 1940, it looked like something from the future."

On this day in 1940, mathematician George Stibitz stood before an audience of mathematicians and scientists at the American Mathematical Society and Mathematical Association of America meeting at Dartmouth College in Hanover, New Hampshire, and offered them something remarkable: a chance to use a computer that wasn't in the building.

The machine Stibitz wanted them to use was the Complex Number Calculator — a Bell Labs device built with co-designer Samuel Williams using 450 telephone relays, sitting in lower Manhattan, roughly 250 miles away. The CNC, which had become operational on January 8, 1940, was designed to help telephone engineers perform complex arithmetic that desk calculators handled badly. It could add, subtract, multiply, and divide complex numbers. It was, by the standards of 1940, an extraordinary machine. It was also in New York.

Stibitz had quietly tested the connection on September 9, two days earlier — running problems from a terminal at Dartmouth to the CNC in New York and receiving answers. A plaque at McNutt Hall at Dartmouth later memorialized that first private test. But September 11 was the public demonstration, before the assembled mathematicians.

A teletype terminal at Dartmouth was connected via telegraph lines to the CNC at Bell Labs. Attendees approached the terminal, typed in mathematical problems, and waited. The signals traveled 250 miles to New York, where the relay circuits did their work, and the answer came back and printed on the teletype at Dartmouth within seconds.

The computer and its users no longer had to be in the same place.

The Model K and the Path There

The machine that made this possible had begun on a kitchen table.

In late 1937, Stibitz sat at home and built a simple binary adder from spare Bell System relays, strips of metal cut from a tobacco tin, flashlight bulbs, and batteries. A lit bulb represented binary 1. An unlit bulb represented binary 0. The device could add binary digits. His colleagues, amused, called it the "Model K" — after the kitchen table where it was assembled. Replicas now reside in the Computer History Museum and the Smithsonian Institution.

Bell Labs executives were initially unimpressed. They changed their minds. By late 1938, Bell had funded a full research program with Stibitz at the lead. The resulting Complex Number Calculator — 450 relays, housed in a room-sized installation — proved that relay circuits could perform the kinds of calculations that Bell's engineering staff needed.

What the Demonstration Meant

Among those who spent considerable time at the Dartmouth terminal on September 11 were Norbert Wiener and John Mauchly. Wiener, already developing what would become the field of cybernetics, was so struck by what he saw that he immediately sent a letter to Vannevar Bush enclosing a memo outlining a machine that had nearly all the features of an electronic digital computer — an idea he had begun formulating partly in response to the Stibitz demonstration. Mauchly would go on to co-design ENIAC, the first general-purpose electronic computer, which was completed in 1945.

The demonstration planted seeds in exactly the right minds.

Stibitz had not invented computer networking in the modern sense. The connection relied on existing telegraph and teletype infrastructure. The CNC was a specialized calculating machine, not a general-purpose computer. The demonstration was not connected to the internet, to packet switching, or to anything resembling the modern network stack.

But the underlying principle — that a terminal could send information to a distant machine, have that machine compute something, and return the result — was exactly what time-sharing systems in the 1960s would build upon, exactly what ARPANET would extend, and exactly what the cloud computing infrastructure of the modern world scales up to billions of simultaneous users.

The word for what Stibitz demonstrated on September 11, 1940 is remote computing. The concept was so obvious and so powerful that once someone demonstrated it was possible, it became an organizing principle of the entire computing industry.

"A problem typed, a distant sound,
Through miles of wire the answer found.
The machine was far, yet close somehow —
A glimpse of how we compute now."

History reminds us: the greatest revolutions don't always require inventing entirely new hardware. Sometimes the breakthrough is realizing that two existing technologies — telephones and calculators — can be made to talk to each other across the miles. Stibitz connected a telegraph line to a relay calculator, and without knowing it, sketched the architecture of the modern networked world.

Sources

← Back to the Desk — History