The Innovators: How a Group of Inventors, Hackers, Geniuses, and Geeks Created the Digital Revolutio - Isaacson Walter. Страница 18

Atanasoff’s enduring romantic appeal is that he was a lone tinkerer in a basement, with only his young sidekick Clifford Berry for a companion. But his tale is evidence that we shouldn’t in fact romanticize such loners. Like Babbage, who also toiled in his own little workshop with just an assistant, Atanasoff never got his machine to be fully functional. Had he been at Bell Labs, amid swarms of technicians and engineers and repairmen, or at a big research university, a solution would likely have been found for fixing the card reader as well as the other balky parts of his contraption. Plus, when Atanasoff was called away to the Navy in 1942, there would have been team members left behind to put on the finishing touches, or at least to remember what was being built.

What saved Atanasoff from being a forgotten historical footnote is somewhat ironic, given the resentment he later felt about the event. It was a visit that he had in June 1941 from one of those people who, instead of toiling in isolation, loved visiting places and snatching up ideas and working with teams of people. John Mauchly’s trip to Iowa would later be the subject of costly lawsuits, bitter accusations, and dueling historical narratives. But it is what saved Atanasoff from obscurity and moved the course of computer history forward.

JOHN MAUCHLY

In the early twentieth century, the United States developed, as Britain had earlier, a class of gentleman scientists who congregated at wood-paneled explorers’ clubs and other rarefied institutes, where they enjoyed sharing ideas, listening to lectures, and collaborating on projects. John Mauchly was raised in that realm. His father, a physicist, was a research chief in the Department of Terrestrial Magnetism at the Washington-based Carnegie Institution, the nation’s foremost foundation for promoting the advance and sharing of research. His specialty was recording electrical conditions in the atmosphere and relating them to the weather, a collegial endeavor that involved coordinating researchers from Greenland to Peru.³⁸

Growing up in the Washington suburb of Chevy Chase, John was exposed to the area’s growing scientific community. “Chevy Chase seemed to have practically all the scientists in Washington,” he boasted. “The director of the Weights and Measures Division of the Bureau of Standards lived near us. So did the director of its Radio Division.” The head of the Smithsonian was also a neighbor. John spent many weekends using a desktop adding machine to do calculations for his dad, and he developed a passion for data-driven meteorology. He also loved electrical circuits. With his young friends in his neighborhood, he laid intercom wires that connected their homes and built remote-control devices to launch fireworks for parties. “When I pressed a button, the fireworks would go off 50 feet away.” At age fourteen he was earning money helping people in the neighborhood fix faulty wiring in their homes.³⁹

While an undergraduate at Johns Hopkins University, Mauchly enrolled in a program for exceptional undergraduates to leap directly into a PhD program in physics. He did his thesis on light band spectroscopy because it combined beauty, experiments, and theory. “You had to know some theory to figure out what the band spectra was all about, but you couldn’t do it unless you had the experimental photographs of that spectrum, and who’s going to get it for you?” he said. “Nobody but you. So I got plenty of training in glass blowing, and drawing vacuums, finding the leaks etc.”⁴⁰

Mauchly had an engaging personality and a wonderful ability (and desire) to explain things, so it was natural that he would become a professor. Such posts were hard to come by in the Depression, but he managed to land one at Ursinus College, an hour’s drive northwest from Philadelphia. “I was the only person teaching physics there,” he said.⁴¹

An essential component of Mauchly’s personality was that he liked to share ideas—usually with a broad grin and a sense of flair—which made him a wildly popular teacher. “He loved to talk and seemed to develop many of his ideas in the give-and-take of conversation,” recalled a colleague. “John loved social occasions, liked to eat good food and drink good liquor. He liked women, attractive young people, the intelligent and the unusual.”⁴² It was dangerous to ask him a question, because he could discourse earnestly and passionately about almost anything, from theater to literature to physics.

In front of a class he played the showman. To explain momentum he would whirl around with his arms flung out and then pulled in, and to describe the concept of action and reaction he would stand on a homemade skateboard and lurch back and forth, a trick that one year resulted in his falling and breaking an arm. People used to drive miles to hear his end-of-term pre-Christmas lecture, which the college moved to its biggest auditorium to accommodate all the visitors. In it he explained how spectrography and other tools of physics could be used to determine what was inside a package without unwrapping it. According to his wife, “He measured it. He weighed it. He submerged it in water. He poked it with a long needle.”⁴³

Reflecting his boyhood fascination with meteorology, Mauchly’s research focus in the early 1930s was on whether long-range weather patterns were related to solar flares, sunspots, and the rotation of the sun. The scientists at the Carnegie Institution and the U.S. Weather Bureau gave him twenty years of daily data from two hundred stations, and he set to work calculating correlations. He was able (this being the Depression) to buy used desk calculators cheaply from ailing banks and to hire a group of young people, through the New Deal’s National Youth Administration, to do computations at fifty cents an hour.⁴⁴

Like others whose work required tedious calculations, Mauchly yearned to invent a machine to do them. With his gregarious style, he set about finding out what others were doing and, in the tradition of great innovators, putting together a variety of ideas. In the IBM pavilion at the 1939 New York World’s Fair, he saw an electric calculator that used punch cards, but he realized that relying on cards would be too slow, given the amount of data he had to crunch. He also saw an encryption machine that used vacuum tubes to code messages. Might the tubes be used for other logical circuits? He took his students on a field trip to Swarthmore College to see counting devices that used circuits made with vacuum tubes to measure bursts of cosmic-ray ionization.⁴⁵ He also took a night course in electronics and began to experiment with his own hand-wired vacuum-tube circuits to see what else they might do.

At a conference at Dartmouth College in September 1940, Mauchly saw a demonstration by George Stibitz of the Complex Number Calculator he had built at Bell Labs. What made the demonstration exciting was that Stibitz’s computer was sitting at Bell’s building in lower Manhattan, transmitting data over a Teletype line. It was the first computer to be used remotely. For three hours it solved problems submitted by the audience, taking about a minute for each. Among those at the demonstration was Norbert Wiener, a pioneer of information systems, who tried to stump Stibitz’s machine by asking it to divide a number by zero. The machine didn’t fall for the trap. Also present was John von Neumann, the Hungarian polymath who was soon to play a major role with Mauchly in the development of computers.⁴⁶

When he decided to build a vacuum-tube computer of his own, Mauchly did what good innovators properly do: he drew upon all of the information he had picked up from his travels. Because Ursinus had no research budget, Mauchly paid for tubes out of his own pocket and tried to cadge them from manufacturers. He wrote the Supreme Instruments Corp. asking for components and declaring, “I am intending to construct an electrical calculating machine.”⁴⁷ He discovered during a visit to RCA that neon tubes could also be used as switches; they were slower but cheaper than vacuum tubes, and he bought a supply at eight cents apiece. “Before November 1940,” his wife later said, “Mauchly had successfully tested certain components of his proposed computer and convinced himself that it was possible to build a cheap, reliable digital device using only electronic elements.” This occurred, she insisted, before he had even heard of Atanasoff.⁴⁸