In 1804, a French weaver named Joseph Marie Jacquard built a loom that wove its pattern from a paper tape made of punched cards. Swap the tape and the same loom wove something else. Nothing in the machinery had to change.
That idea sits near the middle of a long story. People had been counting with their fingers, tally sticks and clay tokens for thousands of years. The Lebombo bone, found in the mountains between Eswatini and South Africa, carries 29 deliberate notches cut into a baboon’s fibula and dates to about 35,000 BCE. It may be the oldest known mathematical artifact. The Roman abacus was in use in Babylonia as early as around 2700 BC.
Gears and logarithms
John Napier, the Scottish mathematician, found that multiplying and dividing could be done by adding and subtracting logarithms. While producing the first logarithmic tables he grew tired of long multiplications and designed “Napier’s bones,” an abacus-like aid. The slide rule followed in the 1620s. William Oughtred built a circular version in 1630 and the modern form in 1632, and engineers relied on it until the pocket calculator arrived.
Gears came next. In 1609 Guidobaldo del Monte made a four-gear multiplier for fractions of a degree, and he is the first person known to have documented gears used for calculation. Wilhelm Schickard designed a calculating machine in 1623, but its single-tooth gear could jam, and a fire destroyed at least one unit in 1624. Blaise Pascal started on his own calculator in 1642, still in his teens. It took three years and 50 prototypes. He built twenty machines over the next decade, and nine survive, mostly in European museums.
Gottfried Wilhelm Leibniz came up with the stepped reckoner around 1672. He never got a fully working carry mechanism, but he did describe the binary system. His complaint about calculation is often quoted: “It is unworthy of excellent men to lose hours like slaves in the labour of calculation which could safely be relegated to anyone else if machines were used.”
Around 1820, Charles Xavier Thomas de Colmar produced the Arithmometer, which became the first successful mass-produced mechanical calculator. Mechanical calculators stayed in use into the 1970s.
Cards, relays and tubes
In the late 1880s the American Herman Hollerith worked out how to store data on punched cards that a machine could read. He built the tabulator and the keypunch to process them. His method was used in the 1890 United States census, which was finished two years faster than the one before. His company eventually became the core of IBM.
By 1920, electromechanical tabulators could add, subtract and print running totals, with dozens of wire jumpers in removable control panels directing the work. When the United States introduced Social Security in 1935, IBM punched-card systems handled records for 26 million workers.
Meanwhile, the word “computer” was a job title, mostly given to women who worked these calculators. In the 1920s the British scientist Lewis Fry Richardson proposed using human computers to model the weather. Desktop mechanical calculators from Friden, Marchant and Monroe could add, subtract, multiply and divide from the 1930s. In 1948 the Austrian inventor Curt Herzstark introduced the Curta, a small hand-cranked calculator.
The first all-electronic desktop calculator was the British Bell Punch ANITA, released in 1961. It ran on vacuum tubes, cold-cathode tubes and Dekatrons, with 12 “Nixie” tubes for its display. It sold well because nothing else like it existed, and because it was silent and quick. In June 1963 the American-made Friden EC-130, built entirely from transistors, replaced the tube design.
