The Job Title That a Machine Stole
The word computer is far older than any machine, and it originally meant a person. According to the Oxford English Dictionary, the first known use appears in a 1613 book by the English writer Richard Brathwait, praising "the truest computer of Times, and the best Arithmetician that euer breathed".
That meaning held for over three hundred years. A computer was someone who carried out calculations by hand, often for astronomy, navigation or engineering. Later in that period, women were frequently hired for the work because they could be paid less than men. By 1943, most human computers were women.
Only around the mid-20th century did the word shift to the machine. The Online Etymology Dictionary dates the sense of "calculating machine" to 1897, and the modern sense of a programmable digital electronic computer to 1945.
Ancient Gears That Could Do Maths
People have built calculating aids for thousands of years. The earliest was probably a tally stick. Around the Fertile Crescent, clay spheres and cones sealed in hollow containers stood in for counts of grain or livestock.
Then came the abacus. The Roman version descended from devices used in Babylonia as early as 2400 BCE. In a medieval European counting house, clerks laid a checkered cloth over a table and moved markers around it by fixed rules to total up sums of money.
The most startling object of all is the Antikythera mechanism, believed to be the earliest known mechanical analog computer, built to calculate astronomical positions. It was pulled from a shipwreck off the Greek island of Antikythera in 1901 and dates to roughly 100 BCE. Nothing of comparable complexity appears again until the fourteenth century.
Five Switches, Each Smaller Than the Last
The modern story is really the story of one component getting smaller and faster. Early in the Industrial Revolution, mechanical devices were built to automate long, tedious jobs such as guiding patterns for looms. By the early 20th century, more sophisticated electrical machines were doing specialised analog calculations.
Digital electronic calculating machines arrived during World War II, and the switch at their heart kept changing: electromechanical relays in the 1930s, thermionic valves in the same decade, semiconductor transistors in the 1940s, the MOSFET in the 1950s, and finally monolithic integrated circuits from the 1950s onward.
Each step meant smaller, more efficient circuits, which led to the microprocessor and the microcomputer revolution of the 1970s. Moore's law noted that transistor counts doubled roughly every two years. In the 1930s computers numbered only a few. Today they run key fobs, microwave ovens and supercomputers alike.
What Is Actually Inside the Box
Strip away the casing and a modern computer is surprisingly consistent. There is at least one processing element, usually a central processing unit in the form of a microprocessor, plus some computer memory, typically semiconductor memory chips.
The processing element does the arithmetic and logic. A sequencing and control unit can change the order of operations in response to information it has stored, which is exactly what makes a computer programmable rather than merely fast.
Everything else is peripheral devices. Input devices such as keyboards, mice and joysticks bring information in. Output devices such as monitors and printers send results out. Touchscreens do both. Together they let information come from outside and let results be saved and retrieved later.
