George Stibitz

Early developer of digital computing

• Born: April 20, 1904
• Birthplace: York, Pennsylvania
• Died: January 31, 1995
• Place of death: Hanover, New Hampshire

Primary Company/Organization: Bell Laboratories

Introduction

George Stibitz was a part of what may have been the greatest technical innovation organization of the twentieth century, Bell Laboratories. Not as well known as other innovators of the twentieth century, Stibitz conceived the first electric digital computer. The claim has been disputed by some, as similar work was being performed in Germany, Japan, and Cambridge, Massachusetts, but none was as advanced or resulted in an operating machine as early as Stibitz's 1940 Model I Complex Calculator. Stibitz designed the prototype and supervised construction of this machine, demonstrating its remote use in 1940. During World War II, he led the development of increasingly advanced computers. After the war, he continued that work until he transitioned to developing computerized solutions for medical problems and performing medical research.

Early Life

George Robert Stibitz was born in York, Pennsylvania, on April 20, 1904, to George and Mildred Stibitz; he was one of four children. At an early age, Stibitz moved with his family from Pennsylvania to Ohio. His father was a professor of theology; his mother, a former teacher of mathematics. Like other innovators of the twentieth century, such as Larry Roberts and Leonard Kleinrock, Stibitz, from an early age, showed a talent for building and designing electrical devices. In his case, it was electrical motors that captured his interest. His fascination with such devices never diminished, and he eventually entered a technical high school, Moraine Park, in Dayton, Ohio.

After graduating from high school, Stibitz went to Denison University, where he graduated in 1926 with a bachelor's degree. He received his master of science degree from Union College in upstate New York. After working for a year, Stibitz returned to school, specializing in mathematical physics at Cornell. He received his doctorate in 1930 and immediately began his employment at Bell Telephone Laboratories (Bell Labs), where he would spend the next eleven years.

Life's Work

When Stibitz began work at Bell Labs in 1930, his role was as a mathematician. The use of telephones (and consequently the development of telephone networks) at this time was growing exponentially. Even at this early stage, Bell Telephone understood that the topologies (both physical infrastructure and the logical flow) had to be designed and planned to accommodate rapidly expanding requirements. For Stibitz and his colleagues, that meant a great deal of work performing calculations to support design decisions, assisted only by desktop manual calculators.

This situation encouraged Stibitz to consider what could be done to replace the mechanical calculators with something that could perform more work, accurately, in a fraction of the time. In 1937, Stibitz had the idea that an electrical digital computer could handle large numbers of complex calculations in a fraction of the time that it currently took. He proposed his idea to management at Bell Labs but was initially turned down. Later that year, however, the decision was reversed, and Stibitz began to design his calculator. He completed the design in the following year, and in 1939 work on a prototype began, with Stibitz doing the construction in his kitchen at home with odd pieces of equipment he had acquired from Bell Labs.

The prototype worked; the next step was to enlarge and enhance it into what became known as the Model I Complex Calculator, a machine about the size of a closet. By early 1940, the machine was in operation, successfully performing calculations for Stibitz and his colleagues. Problems to be solved (inputs) were handled by an operator sitting at a teletype machine who entered the values comprising the problem. The results would be returned in seconds. The Model I Complex Calculator was a dramatic improvement and represented a major advance that would make it possible for mathematicians to support network design.

There were, however, some real limitations, which would become obvious as time went on and which would contribute to an agenda in coming years for improving computers. For example, although three workstations (basically teletype machines) were connected to the calculator, only one could be used. Creating the capability to handle multiple, simultaneous users would not be possible for several years. Second, at least initially, the calculations were restricted to combinations of multiplication and division, although addition and subtraction would be introduced in later models. Finally, the machine was not programmable; it was mainly a very large and faster version of the desktop calculator. A limited degree of programmability would be introduced as computers developed.

On balance, the machines were faster than their predecessors, always accurate, and could support an ever-increasing demand for work. There was another capability, however, that would be demonstrated in dramatic fashion by Stibitz on September 11, 1940, at Dartmouth College in Hanover, New Hampshire. The Mathematical Society of America was conducting its annual meeting. In attendance were the country's most distinguished mathematicians, including John von Neumann and Norbert Wiener.

Stibitz showed the assembled group a workstation similar to those used at the Bell Labs facility in New York City. The workstation was connected to the calculator by telephone lines. Participants were invited to submit problems that would then be entered into the teletype machines, transmitted via the phone lines to the computer at the New York facility. The correct solution would be returned in seconds. Stibitz had demonstrated not only the capabilities of an electric computer but also the possibilities of remote processing.

The first model remained in use at Bell Labs until the late 1940s, and there would be successor calculating machines of greater complexity in the coming years. In addition to their increasing speed and ability to perform advanced calculations, the newer machines were designed to be more reliable and could be maintained by workers who were not necessarily advanced engineers. Stibitz, however, would not be involved in those projects. In December 1941, the United States entered World War II, and Stibitz, like many advanced engineers and scientists, would be heavily involved in war work.

Military operations required substantial mathematical expertise to calculate ballistics. There were many different types of artillery for various purposes. For crews to be able to hit their targets, firing tables that gave solutions to how to set weapons, depending on range and other factors (such as temperature), were necessary. Consulting these ballistics tables and applying the correct settings to guns and other weapons before they were fired would, one would hope, result in hitting the target on the first try. These ballistics problems were complicated by the increased use of antiaircraft artillery, which fired at objects moving at various speeds and at various altitudes.

To construct these tables, some government agencies hired large pools of junior-level mathematicians (who were called computers at a time when the term did not primarily denote a machine). Automating this process, however, was a priority—a project in which Stibitz became involved during the war years. Such efforts would eventually contribute to the development of early computers.

With the end of the war, Stibitz did not return to Bell Labs. For several years he worked as a consultant, performing research and experimentation with the goal of developing new computers. After that, he made a major transition in his career, becoming a member of the faculty at Dartmouth College, specializing in biotechnology, in 1964. The field focused on using computer science to investigate medical issues, such as the movement of oxygen, the movement of blood, the anatomy of brain cells, and mathematical models of capillary transport. Stibitz officially retired from Dartmouth in 1973 but remained as a researcher there until the early 1980s.

Personal Life

In 1930, the year in which Stibitz began work at Bell Labs, he married Dorothea Lamson. Together they had two daughters.

Stibitz personally held thirty-eight patents, not including those awarded to Bell Labs for technologies to which he had contributed. Stibitz's achievements were widely recognized during his lifetime; he was presented with several awards. These included the Harry M. Goode Memorial Award (1965, with Konrad Zuse) from the American Federation of Information Processing Societies, membership in the National Academy of Engineering (1976), and the Babbage Society Medal (1982). He was inducted into the Inventors Hall of Fame in 1983. Stibitz died on January 31, 1995, in Hanover, New Hampshire, of natural causes; he was ninety years old.

Bibliography

Gertner, Jon. The Idea Factory, Bell Labs and the Great Age of American Innovation. New York: Penguin, 2012. Print. A well-received account not only describing the history of Bell Labs and its accomplishments but also offering a good analysis of the consequences of those inventions on today's technical and economic environments.

Grier, David Alan. “George Stibitz's Values and R. C. Archibald's Slide Rule.” Computer 39.1 (2006): 11–13. Print. A description of Stibitz's legacy as an innovator and developer at Bell Labs.

Irvine, M. M. “Early Digital Computers at Bell Telephone Laboratories.” IEEE Annals of the History Of Computing 23.3 (2001): 22. Print. An account of computer development at Bell Labs from 1937 (the beginning of Stibitz's efforts to develop an electronic computer) to 1958.

Stibitz, George. “Early Computers.” A History of Computing in the Twentieth Century. Ed. N. Metropolis. New York: Academic, 1980. Print. Stibitz's own account of the rise of electronic computing to the dawn of personal computers.