Computing history is full of accidents, side projects and things nobody meant to start. These tiles visit a codebreaker who nearly ran in the Olympics, a chain of tea shops that built a computer, a magazine cover photographed with an empty box, and a coffee pot in Cambridge that got its own camera. You will also find a famous moth, an algorithm written out as a poem, and three words that let the web happen.
Every tile in a Computers and the Internet game carries one of these. Open one to read it now, or leave them for the board to hand you.
Pioneers
nine stories
1The notes that swallowed the paper
On the tile: Ada Lovelace1843
In 1843 Ada Lovelace published a translation of an Italian engineer's French article about Charles Babbage's Analytical Engine. To it she attached notes of her own. The finished paper ran sixty-six pages, and forty-one of them were hers: the commentary had grown larger than the work it explained. It appeared in Taylor's Scientific Memoirs that August with no author's name attached, signed only with three initials at the very end, A.A.L.
Charles Babbage designed Difference Engine No. 2 between 1847 and 1849, worked the drawings out in fine detail, and then made no attempt to build it. The only engine finished in his lifetime was a fragment: about one seventh of his earlier Difference Engine No. 1, roughly two thousand parts, assembled in 1832. More than a century later the Science Museum in London took up the drawings. The calculating section was completed in 1991 for the two hundredth anniversary of his birth, and the printing mechanism was added in 2002.
Alan Turing took up distance running after the war and joined the Walton Athletic Club, where he immediately became the best runner they had. In August 1947 he entered the national marathon championship at Loughborough and finished fifth in two hours forty-six minutes. The winning time at the 1948 Olympic marathon in London was only about ten minutes faster, and he was in contention for the British team. Fibrositis in January 1948 kept him out, and a leg injury ended his serious racing.
The story usually told is that Herman Hollerith took the punched card from a railroad conductor. Tickets of the 1880s really did carry what was called a punch photograph: printed down the margin were choices such as Slim, Medium, Stout, and Light or Dark hair, and every word that did not describe the passenger was punched out. Hollerith knew them. But his idea for tabulating data on cards came out of the 1880 census, in conversation with Dr. Billings, who ran vital statistics there, and his real invention was the machinery that read the cards.
Konrad Zuse completed the Z3 in Berlin in 1941, and the Computer History Museum describes him as working in complete isolation from developments elsewhere. On 12 May he presented it to an audience of scientists, among them two professors from the German aviation laboratory. Assembled from about two thousand relays, it worked in binary, handled floating-point numbers, and stored its program on an external tape, so a new job needed no rewiring. An Allied bombing raid destroyed the Z3 in 1943, and a replica was built in the 1960s under Zuse's supervision.
Claude Shannon came to MIT as a graduate student after finishing at Michigan in 1936, and took a part-time job running Vannevar Bush's differential analyzer. His master's thesis, A Symbolic Analysis of Relay and Switching Circuits, used the two-symbol algebra George Boole had built for reasoning to describe circuits made of electrical relays. Modern switching theory starts there. Harvard's Howard Gardner later called it "possibly the most important, and also the most famous, master's thesis of the century."
On 9 December 1968 Douglas Engelbart sat at a custom console in San Francisco's Civic Auditorium, driving a computer that stood thirty miles away at Stanford Research Institute. For ninety minutes he and his team showed a packed hall what their NLS system could do: a pointing device someone had jokingly named a mouse, hypertext links, collaborative editing, and colleagues who joined by video from the lab. The audience rose in a standing ovation. It is remembered as the Mother of All Demos.
Six women were ENIAC's programmers: Kathleen McNulty Mauchly Antonelli, Jean Jennings Bartik, Frances Snyder Holberton, Marlyn Wescoff Meltzer, Frances Bilas Spence and Ruth Lichterman Teitelbaum. Two of them built the missile trajectory run that impressed reporters when the machine was unveiled at Penn's Moore School in 1946. Afterward there was a celebratory dinner at Houston Hall, and the women were not invited. Archival photographs show women and men together at the machine; the pictures the newspapers chose to print showed only the men.
In March 1989 Tim Berners-Lee, a software engineer at CERN, handed his manager a proposal for an information system that would link documents across the laboratory's computers. Mike Sendall read it and wrote three words on it: "Vague, but exciting." That was enough. Sendall did not fund a project or assign a team; he simply let Berners-Lee work on the idea on the side. The first browser and editor followed in 1990, written on a NeXT computer at his desk.
Colossus went to work at Bletchley Park in 1944, reading German teleprinter traffic, and then vanished. The machines were broken up, the secret held for decades, and in 1960 the original drawings of the mechanical assemblies were deliberately burned. In 1993 Tony Sale set out to build one anyway. His sources were eight wartime photographs taken in 1945 plus fragments of circuit diagrams that engineers had kept, in his words, "quite illegally, as engineers nearly always do." A basic Colossus was working again by early 1996.
J. Lyons and Company ran tea shops across Britain, and from 1951 it ran a computer too. LEO, the Lyons Electronic Office, was built with help from the Cambridge team behind EDSAC. Its first business job, in November 1951, was called Bakery Valuations: it priced the ingredients in every loaf and cake made at the Lyons factory at Cadby Hall in west London. Soon it was running the payroll. The machine was unreliable, but the testing regime was so strict that no one was ever paid late.
ENIAC was demonstrated publicly in February 1946 with about eighteen thousand vacuum tubes, thirty tons of hardware and a room of its own, and its builders, John Mauchly and J. Presper Eckert, held the patent on the electronic digital computer. In October 1973 Judge Earl Larson ruled in Honeywell versus Sperry Rand that they had not invented it themselves, but had derived the subject matter from John Vincent Atanasoff, who with Clifford Berry had prototyped a machine at Iowa State in 1939. The patent was invalid.
Seymour Cray's 1976 machine was shaped like the letter C, and the shape was the engineering. More than sixty miles of wire ran inside it, with no segment longer than three feet, so signals had less ground to cover. Around the base ran a padded circular seat that concealed the power supplies, which made the world's fastest computer a piece of lounge furniture as well. It held the speed record until 1982, sold for as much as ten million dollars, and drew enough power to run about ten homes.
The Apple-1 was a bare single board with four kilobytes of memory. Buyers supplied their own case, keyboard, power supply, display and storage. Steve Wozniak built it and Steve Jobs sold it, starting with an order for fifty from the Byte Shop. The retail price was $666.66, which Wozniak chose because he liked repeating numbers and, by his own account, had no idea of that number's darker reputation. About two hundred boards were built before the Apple II replaced it.
The January 1975 issue of Popular Electronics put a new computer kit on its cover and set off the personal computer business. There was a catch. When editor Les Solomon commissioned the article the machine was not finished and did not yet have a name, so the magazine photographed an empty box and chose the name Altair. Orders buried its maker, MITS, within weeks, and two young programmers named Paul Allen and Bill Gates supplied it with a BASIC interpreter.
The chip that opened the microprocessor era was commissioned by Busicom, a Japanese calculator maker, and Busicom held exclusive rights to the design. Intel had to negotiate, offering Busicom a break on the price of chips in exchange for permission to sell the family for anything other than calculators. Federico Faggin led the design, and before it went into production he added a flourish of his own, laying his initials along the edge of the layout. Every 4004 ever made carried a microscopic F.F.
IBM's Deep Blue could evaluate two hundred million chess positions a second. In Philadelphia in 1996 it won the opening game against Garry Kasparov, the first time a computer had beaten a reigning world champion under regular time controls, though Kasparov recovered and took that match four to two. The rematch came in May 1997 in New York, and Deep Blue won the deciding sixth game and the match, three and a half to two and a half. The machine was then retired to the Smithsonian.
The Commodore 64 arrived in 1982 at $595, with sixty-four kilobytes of memory and graphics that embarrassed pricier machines. Thousands of software titles followed, and Guinness World Records dates production from August 1982 all the way to April 1994. It is still the best-selling desktop computer ever made, and nobody can say by how much: Commodore's founder Jack Tramiel claimed as many as 30 million sold, the company itself said 17 million, and the estimate Guinness now calls credible is about 12.5 million.
At half past ten at night on 29 October 1969, a computer at UCLA sent the first message over the ARPANET to a computer at Stanford Research Institute. The message was supposed to be the word login. Charley Kline typed the letters while talking to Bill Duvall by telephone: got the l, got the o, and then the system crashed. So the network's first word was lo, as Leonard Kleinrock likes to point out, as in lo and behold. They managed the full login about an hour later.
Robert Metcalfe's first name for it was the Alto Aloha Network, after the Xerox computer it linked and the Hawaiian radio network it borrowed ideas from. The team changed it to Ethernet to make clear that any computer could join, picking up a remark by their colleague Chuck Thacker that coaxial cable is nothing but captive ether. Metcalfe's 1973 memo, typed on an IBM Selectric, was titled Ether Acquisition, and he later described it as heavy with handwritten annotations, one of them the word ether in capital letters.
The documents that define how the internet works are called Requests for Comments, and the name was never a pose. RFC 1, written by Steve Crocker at UCLA and dated 7 April 1969, is titled Host Software. Its introduction says: "I present here some of the tentative agreements reached and some of the open questions encountered. Very little of what is here is firm and reactions are expected." Thousands of RFCs have followed, and the internet still runs on documents that invite argument.
Long before most people had heard of electronic mail, Queen Elizabeth II sent some. On 26 March 1976 she pressed a few buttons at the Royal Signals and Radar Establishment in Malvern, England, and sent a message over the ARPANET, the research network that grew into the internet. Peter Kirstein, who had brought the ARPANET to Britain three years earlier, set up her account himself and chose her username: HME2, for Her Majesty Elizabeth II. The World Wide Web that would carry everyone else's mail was still thirteen years from being proposed.
In 1978 the ARPANET had a printed directory of its users, all 2,600 of them. Gary Thuerk sold computers for Digital Equipment Corporation, and to advertise demonstrations of DEC's latest machines on the West Coast he emailed an invitation to about four hundred people he found in that book. It went out on 3 May 1978. The replies were not friendly: users argued the network was not for bulk advertising, and Thuerk's boss made him promise never to do it again.
On 15 March 1985 symbolics.com became the first second-level .com domain on the internet, registered by Symbolics, a computer maker. Nothing about the moment suggested a gold rush. The Macintosh was a year old, the online service that became AOL had not launched yet, and the internet was still a government-funded experiment linking research sites. A decade would pass before home connections were fast and cheap enough for ordinary people to want an address of their own.
In 1988 a twenty-three-year-old graduate student named Robert T. Morris released a worm onto the internet. It destroyed nothing, but it copied itself so eagerly that about 6,000 of the network's 60,000 host machines were crippled for days. His father was a computer security expert at the National Security Agency. Morris became the first person convicted under the Computer Fraud and Abuse Act, and in 2008 he said he had been trying to measure the size of the internet, not damage it.
CERN owned the World Wide Web. On 30 April 1993 two of its directors, Walter Hoogland and Helmut Weber, signed an internal document addressed to whom it may concern, stating that "CERN relinquishes all intellectual property rights to this code, both source and binary form, and permission is granted for anyone to use, duplicate, modify and redistribute it." A particle physics laboratory had decided not to own the software Tim Berners-Lee built there. The next year's version went out under an open source license instead, with CERN keeping the copyright but letting anyone use and modify the web freely.
Before the web won, Gopher was ahead. Built at the University of Minnesota and named both for the university mascot and for go-for, meaning fetch, it organized information in folders rather than clickable links, and in the early 1990s it was growing faster than the web. Universities adopted it, and so did the United States Congress. Then in 1993 Minnesota began charging for Gopher server licenses, in the same spring the web became officially public and free. Gopher's developers were planning hyperlinks. They were too late.
For the Festival of Britain the computer company Ferranti needed something to show the public. John Bennett, an Australian on its staff, suggested a machine to play Nim, the game of taking matches from piles in which you try to remove the last one. Raymond Stuart-Williams turned the design into hardware, and Nimrod stood in Ferranti's display in South Kensington for a festival that ran from May to September 1951. It traveled to a Berlin show later that year, and afterward Ferranti took it apart.
In the autumn of 1951 a BBC outside broadcast team recorded the Ferranti Mark 1 at Manchester playing God Save the King, an abandoned attempt at Baa Baa Black Sheep, and a truncated In the Mood. Afterward an engineer named Frank Cooper asked for a copy and was refused on copyright grounds, so the crew cut him a separate one-off acetate disc instead. That private favor is the reason the earliest known recording of computer music survives at all.
On 9 September 1947 operators of the Harvard Mark II found a moth caught between relay contacts, taped it into the logbook and wrote "First actual case of bug being found." The page is often presented as the origin of the word. Read the wording again: they called it the first actual bug, which only makes sense as a joke about a word they already used. Engineers had been calling faults bugs since the 1800s, and Thomas Edison used the term.
IBM shipped the 305 RAMAC in 1956 with the world's first hard disk drive inside: fifty magnetically coated metal platters, spinning together, holding about five million characters. The capacity was not the point. Tape and punched cards had to be read in order, while a disk could jump straight to the record you wanted, which is what made real-time business computing possible. The first RAMAC went to Zellerbach Paper in San Francisco.
Doug Engelbart's pointing device was a carved wooden block holding two wheels set at right angles, each rolling in one direction only, with a button on top. The patent, filed in 1967 and granted in 1970, calls it an X-Y position indicator for a display system; the word mouse appears nowhere in it, and nobody remembers who said it first. The cord originally came out the front. Engelbart said they turned it around because the tail got tangled when you moved your arm.
The password did not arrive to keep out intruders. It arrived because people wanted privacy from their colleagues. MIT's Compatible Time-Sharing System, first demonstrated in 1961 under Fernando Corbato, let many people use one computer at the same time, which meant many people's files sat on the same machine. CTSS gave each user an individual account with a personal password, and is widely credited as the first computer system to do so. Every login prompt since descends from that housekeeping problem.
Russell Kirsch worked at the National Bureau of Standards on SEAC, one of the earliest American computers. In 1957 he built a rotating drum scanner and fed it a snapshot of his infant son, Walden. The result, five centimeters square and in black and white, was the first image scanned into a computer, which makes it the ancestor of every digital photograph. In 2003 Life magazine listed the picture among the hundred photographs that changed the world.
In 1968 the novelist Len Deighton leased an IBM Magnetic Tape Selectric Typewriter, an early word processor. Its main unit weighed about two hundred pounds, so a crane had to lift it into his house on Merrick Square in London. His assistant Ellenor Handley mastered the machine. The book written on it, Bomber, appeared in 1970, and the scholar Matthew Kirschenbaum identifies it as the first novel written on a word processor.
In 1999 Shigetaka Kurita and his colleagues at the Japanese telecom NTT DOCOMO designed 176 small pictures for mobile phones and pagers. Each had to fit a grid twelve pixels by twelve, the limit of the i-mode system they were built for, so a face or a weather symbol got 144 dots and not one more. Kurita drew on manga and on existing printers' dingbats. Many of the symbols on phones today can be traced back to that set.