New Yorker Kahn
"Maybe my problem growing up was too much idle time," Kahn told IEEE Spectrum. When not practicing the piano or participating in sports in Queens, he played mind games. He could glance at a puzzle, shut his eyes, and solve it as easily as if he could still see it on the paper.
For the most part, the adolescent Kahn put play before schoolwork. But his father, Lawrence Kahn, was a public high school administrator in New York City and pushed education so hard that both his children got Ph.D.s. Trained as an accountant, he set them tough math goals. "The joke in the family was if you got a 98 on the math test he would ask where the other two points were," said Kahn's sister, Diana Taylor (her doctorate is in commutative algebra).
Kahn enrolled in a five-year engineering program, spending the first two years at nearby Queens College and the core years at City College in Manhattan, where, challenged at last, he at last grew diligent. In the '50s, City College was a "big, factory-like school--technically terrific," said Kahn. Adept in science as well as mathematics, he tried industrial, then chemical engineering before deciding on electrical engineering.
He worked summers at Bell Laboratories, including the headquarters then in New York City. At a time when color TV and digital communications were among the exciting fields, he was drawn to telephone traffic engineering, a mature but math-intensive field. The work involved analyzing the telephone system's infrastructure to decide where to put new lines or switches, as well as using computer simulations to estimate likelihoods of delays and blocked calls. He learned Fortran and assembly code on an IBM 704 mainframe.
It was not research, but he gained a macroscopic view of telephone system operations and an appreciation for large-scale system design and simulation. "We were into modeling and analysis tools that could apply to many things," he said.
Kahn graduated from City College in 1960 and received a fellowship from the National Science Foundation (NSF) to attend Princeton University in New Jersey. There, he devoured books in library carrels, following an agenda of his own "far broader than they had laid out for me." He made time for music, too. A stint as a group social chairman netted him season tickets to New York City's Metropolitan Opera and initiation into a now favorite activity. For his Ph.D. thesis in electrical engineering, he examined two applied math problems--using bandwidth more efficiently and representing signals by sampling--and had his findings published in two different IEEE journals.
His next move was in 1964 to the Massachusetts Institute of Technology (MIT), Cambridge, as an assistant professor. After just two years of teaching information theory, communications, and probability (and buying a 1965 racing green Porsche convertible, which he still drives today), Kahn took leave to get some practical experience.
He chose BBN Corp., a small Cambridge-based organization and a magnet for creative research in computing. Kahn focused on the virgin territory of computer networking. Eventually, he got involved with recruits mostly from Lincoln Laboratory with experience in real-time computer systems. As the communications guy, Kahn said, "I was the stranger."
His tools were "pencil and paper and timesharing computers." He also struck up an association with Larry Roberts, who had been shanghaied from Lincoln Laboratory by Robert Taylor to jump-start computer networking at the Pentagon's Advanced Research Projects Agency (ARPA).
(c) Copyright 1996, The Institute of Electrical and Electronics Engineers, Inc.