Larry Roberts was the most important person to the birthing of the Arpanet, and then, arguably, one of the most important people in the creation of the x.25 standard. Yet he was much more than an accomplished scientist/computer engineer, for he was a founder of Telenet and as the CEO took the company public, all while devoting considerable time and energy within CCITT, working to create the X.25 standard. Two years later, in 1979, he managed the sale of Telenet to GTE Corporation for approximately $90 M. He remained with GTE for another year, before leaving to found four more companies.
While earning his PhD in Electrical Engineering from MIT in 1963, Roberts worked at MIT Lincoln Labs as did his two office mates Ivan Sutherland and Leonard Kleinrock. Roberts, who had been working for Wes Clark writing system software for the TX-2, informally took over Clark’s project after Clark left in a dispute with management. Roberts returned to Lincoln Labs after graduation. Then in 1966, Bob Taylor recruited him to be Project Manager of the Arpanet Project. When Taylor resigned to join Xerox PARC, Roberts took over his responsibilities as Director of the IPTO Office. He would then fund numerous other projects, including ALOHAnet and Douglas Engelbart’s Augmentation Research Center at SRI international.
It was clear from the start that the reconstruction of the Arpanet history would not be credible without an interview with Roberts. So I waited until I better understood that history before calling him. To my delight, he joked that since I had talked to nearly everyone else, he had better make sure my facts were correct. Nearly three decades later I can still see the two of us sitting in a room full of tables holding computers and other electronic equipment in various stages of being assembled and tested. He sat not at a desk, but before a computer with me next to him and we talked. And while it was clear his attention was mostly on the computer, I feel it made his answers more spontaneous. When I later read the transcription and realized how much we did not have time to discuss, I called for a rare second interview, and he readily agreed to meet again. So enjoy two interviews with a man secure in his accomplishments and with nothing to prove or assert.
Keywords: Arpanet, x.25, Lincoln Labs, IPTO