Chapter 4 - Networking: Vision and Packet Switching 1959 - 1968
4.1 The Intergalactic Network: 1962-1964
When Dr. J. C. R. Licklider became Director of the IPTO (Information Processing Techniques Office) office of ARPA (Advanced Research Projects Agency) in October 1962, he had a vision of computing vastly grander than time-sharing. Yet knowing it would take all the resources and management he had at his disposal to implement the desired number of time-sharing projects, he decided to leave the issues of interconnecting the time-shared computers for the future. In April 1963, Licklider first described his over-arching vision of an “Intergalactic Network” in a memo to computer scientists.
The term ‘Intergalactic Network’ was a kind of intentionally grandiloquent way to express the idea, because we didn’t really expect to get at that right away. It was all we could possibly do to make timesharing systems work.1
Licklider conceived a future of networked computers when few computers supported more than one user, and most people would have labored to conceive a future of thousands, let alone millions, of computers. It required a true soaring of the imagination to see end-user computing when the paradigm of the day was batch processing with users passing decks of computer cards to trained operators and sometimes having to wait days to receive their results, hardly interactive computing.
Only for the self-effacing Licklider, computer networking came naturally. A gracious man with endless enthusiasm for new ideas and the will to implement them, Licklider’s career mirrored his innate gift for connecting with people. As a member of the Project Charles, an MIT summer study that advanced the need for an air defense system that became the genesis of the SAGE Project, he came into contact with some of the best science mangers of his time. He then taught at MIT and became a group leader at MIT’s Lincoln Labs, witnessing first hand the countless collaborations between these organizations, government agencies and commercial contractors. Throughout this period Licklider’s insatiable quest for new ideas brought him into contact with the finest computer scientists of his day. When in 1957, he left MIT to join Bolt Beranek and Newnan, he would rely on this network of talented individuals to assemble an exceptional staff. In creating an environment of unbridled energy for exploring how computers could be made more ubiquitous and useable, he advanced the state of many basic ideas, especially time-sharing. Mining this milieu, he made the fantastic leap that guaranteed that the future of computing would evolve into a gigantic network: an Intergalactic Network.
Once at IPTO, Licklider again drew upon his network of individuals as well as a number of forward-thinking corporations to bring this grand new world into being. For the most part he was successful, but not always. When he invited the participation of AT&T and its Bell Labs, he generated little more than personal frustration. He recalls:
The telephone company wasn’t interested. Well, Buick used to have the expression: ‘When the better cars are built, Buick will build them.’ I have a feeling the telephone company felt: ‘When there’s a time for networks, we’ll know about it.’2
In July 1964, after only two years, Licklider left IPTO for IBM to assume a role he later described as more promoter than scientist or researcher. In September, Dr, Ivan Sutherland, a young prescient computer scientist and a member of Licklider’s network, took over as the new Director of IPTO. One of his first, and easiest, decisions was to fund the work of his friend Dr. Lawrence G. (Larry) Roberts, a principal investigator at Lincoln Labs. Roberts informally led the group Wes Clark once headed; the same Clark who had earlier so influenced Licklider. Roberts, who was to become one of the giants of computer communications, remembers:
Nobody knew what anybody should be doing with the computer. The TX-2 was really a tremendous machine. When Ivan went to ARPA, we worked together to supply money to the group, and as a result, I wound up being in charge as the contract monitor for ARPA, but on no formal terms at Lincoln.
Roberts and Sutherland had known each other since entering the MIT doctoral program in 1959. They quickly became friends, sharing both an office and a passion for computer graphics. When preparing their computer demonstrations to complete their degree requirements, they used the TX-2 computer at Lincoln Labs on alternate nights. (In their final exams, each demonstrated algorithms that become seminal in computer graphics.3)
In November 1964, Licklider, Sutherland and Roberts were among a contingent of ARPA scientists traveling by train to Hot Springs, Virginia, to attend the Second Congress of the Information System Sciences sponsored by MITRE Corporation.4 The trip afforded the scientists the rare opportunity to interact as a group and they crammed every minute with conversations later characterized by Licklider as intense and spirited. When the trip began Roberts had at best a modest interest in computer communications. He remembers:
I didn’t have any strong direction at that point, or before that point. Lick was arguing for the ‘Intergalactic Network’ concept. Lick and I and a number of other people talked at the conference about what the next thing would be, and I concluded that the thing to work on would be the communications between computers and to computers, because the computer stuff itself was a big team activity at that point, and one that I thought I knew how to handle anyway. The real challenging task was to try to interlink these systems.
Interview with author
Licklider: “The telephone, that was the object of their [AT&T] research, was a physical system that kind of stopped with the microphone on one end and the earphone on the other.”
For Sutherland it was Sketchpad, and for Roberts it was a hidden lines algorithm that was minor to his thesis.
Another organization associated with MIT and a technology prototyping organization doing work for government agencies, especially the Air Force.