One important mission of Bell Labs was to figure out ways to amplify a phone signal over long distances while filtering out static. The engineers had formulas that dealt with the amplitude and phase of the signal, and the solutions to their equations sometimes involved complex numbers (ones that include an imaginary unit that represents the square root of a negative number). Stibitz was asked by his supervisor if his proposed machine could handle complex numbers. When he said that it could, a team was assigned to help him build it. The Complex Number Calculator, as it was called, was completed in 1939. It had more than four hundred relays, each of which could open and shut twenty times per second. That made it both blindingly fast compared to mechanical calculators and painfully clunky compared to the all-electronic vacuum-tube circuits just being invented. Stibitz’s computer was not programmable, but it showed the potential of a circuit of relays to do binary math, process information, and handle logical procedures.16
HOWARD AIKEN
Also in 1937 a Harvard doctoral student named Howard Aiken was struggling to do tedious calculations for his physics thesis using an adding machine. When he lobbied the university to build a more sophisticated computer to do the work, his department head mentioned that in the attic of Harvard’s science center were some brass wheels from a century-old device that seemed to be similar to what he wanted. When Aiken explored the attic, he found one of six demonstration models of Charles Babbage’s Difference Engine, which Babbage’s son Henry had made and distributed. Aiken became fascinated by Babbage and moved the set of brass wheels into his office. “Sure enough, we had two of Babbage’s wheels,” he recalled. “Those were the wheels that I had later mounted and put in the body of the computer.”17
That fall, just when Stibitz was cooking up his kitchen-table demonstration, Aiken wrote a twenty-two-page memo to his Harvard superiors and executives at IBM making the case that they should fund a modern version of Babbage’s digital machine. “The desire to economize time and mental effort in arithmetical computations, and to eliminate human liability to error is probably as old as the science of arithmetic itself,” his memo began.18
Aiken had grown up in Indiana under rough circumstances. When he was twelve, he used a fireplace poker to defend his mother against his drunk and abusive father, who then abandoned the family with no money. So young Howard dropped out of ninth grade to support the family by working as a telephone installer, then got a night job with the local power company so that he could attend a tech school during the day. He drove himself to be a success, but in the process he developed into a taskmaster with an explosive temper, someone who was described as resembling an approaching thunderstorm.19
Harvard had mixed feelings about building Aiken’s proposed calculating machine or holding out the possibility that he might be granted tenure for a project that seemed to be more practical than academic. (In parts of the Harvard faculty club, calling someone practical rather than academic was considered an insult.) Supporting Aiken was President James Bryant Conant, who, as chairman of the National Defense Research Committee, was comfortable positioning Harvard as part of a triangle involving academia, industry, and the military. His Physics Department, however, was more purist. Its chairman wrote to Conant in December 1939, saying that the machine was “desirable if money can be found, but not necessarily more desirable than anything else,” and a faculty committee said of Aiken, “It should be made quite clear to him that such activity did not increase his chances of promotion to a professorship.” Eventually Conant prevailed and authorized Aiken to build his machine.20
In April 1941, as IBM was constructing the Mark I to Aiken’s specifications at its lab in Endicott, New York, he left Harvard to serve in the U.S. Navy. For two years he was a teacher, with the rank of lieutenant commander, at the Naval Mine Warfare School in Virginia. One colleague described him as “armed to the teeth with room-length formulas and ivy-covered Harvard theories” and running “smack into a collection of Dixie dumbbells [none of whom] knew calculus from corn pone.”21 Much of his time was spent thinking about the Mark I, and he made occasional visits to Endicott wearing his full dress uniform.22
His tour of duty had one major payoff: at the beginning of 1944, as IBM was getting ready to ship the completed Mark I to Harvard, Aiken was able to convince the Navy to take over authority for the machine and assign him to be the officer in charge. That helped him circumnavigate the academic bureaucracy of Harvard, which was still balky about granting him tenure. The Harvard Computation Laboratory became, for the time being, a naval facility, and all of Aiken’s staffers were Navy personnel who wore uniforms to work. He called them his “crew,” they called him “commander,” and the Mark I was referred to as “she,” as if she were a ship.23
The Harvard Mark I borrowed a lot of Babbage’s ideas. It was digital, although not binary; its wheels had ten positions. Along its fifty-foot shaft were seventy-two counters that could store numbers of up to twenty-three digits, and the finished five-ton product was eighty feet long and fifty feet wide. The shaft and other moving parts were turned electrically. But it was slow. Instead of electromagnetic relays, it used mechanical ones that were opened and shut by electric motors. That meant it took about six seconds to do a multiplication problem, compared to one second for Stibitz’s machine. It did, however, have one impressive feature that would become a staple of modern computers: it was fully automatic. Programs and data were entered by paper tape, and it could run for days with no human intervention. That allowed Aiken to refer to it as “Babbage’s dream come true.”24
KONRAD ZUSE
Although they didn’t know it, all of these pioneers were being beaten in 1937 by a German engineer working in his parents’ apartment. Konrad Zuse was finishing the prototype for a calculator that was binary and could read instructions from a punched tape. However, at least in its first version, called the Z1, it was a mechanical, not an electrical or electronic, machine.
Like many pioneers in the digital age, Zuse grew up fascinated by both art and engineering. After graduating from a technical college, he got a job as a stress analyst for an aircraft company in Berlin, solving linear equations that incorporated all sorts of load and strength and elasticity factors. Even using mechanical calculators, it was almost impossible for a person to solve in less than a day more than six simultaneous linear equations with six unknowns. If there were twenty-five variables, it could take a year. So Zuse, like so many others, was driven by the desire to mechanize the tedious process of solving mathematical equations. He converted his parents’ living room, in an apartment near Berlin’s Tempelhof Airport, into a workshop.25
In Zuse’s first version, binary digits were stored by using thin metal plates with slots and pins, which he and his friends made using a jigsaw. At first he used punched paper tape to input data and programs, but he soon switched to discarded 35 mm movie film, which not only was sturdier but happened to be cheaper. His Z1 was completed in 1938, and it was able to clank through a few problems, though not very reliably. All the components had been made by hand, and they tended to jam. He was handicapped by not being at a place like Bell Labs or part of a collaboration like Harvard had with IBM, which would have allowed him to team up with engineers who could have supplemented his talents.