I had the honor of working for Corby on CTSS and Multics development in the 1960s. This note is about my personal experience with his contribution to CTSS and Multics. See Corby's own home page and the 1991 CACM interview for more biographical information. ( An interview with Fernando José Corbató, Commun. ACM 34 No. 9, September 1991.) Corby was also interviewed by Arthur Norberg for the Charles Babbage Institute in 1990; the 80-page transcript is available on the CBI website.
Corby was interviewed by John Fitch for MIT Science Reporter On WGBH-TV Boston on May 9, 1963. He demonstrated CTSS running on MIT's 7090. The program aired on aired on WGBH on May 16, 1963.
Before Corby: No Timesharing.
After Corby: Timesharing.
Corby saw that it could be done, and made it happen. His vision was
- Interactive computing
- Timesharing instead of batch
- Remote access to the computer
- Systems written in high-level languages
- Users protected from each others' errors
- Sharing and privacy
- The computer utility
We're used to these ideas now, but they were revolutionary in the sixties.
Leader of the Team
Corby convinced others to do things the way he envisioned. In the process, he inspired intense devotion to the project and to himself. He was a true gentleman; he never raised his voice to anyone, and he conveyed the utmost regard for each team member. We strove intensely for his approval. Some of his management philosophy is chronicled in "A Managerial View of the Multics Development."
One thing I liked about working for Corby was that I knew he could do my job, if he wanted to. Everybody felt this way. He might not have all the details of a particular situation, but once he found them out, he could program as well as the best.
There was a staff meeting for the whole project every week. Tuesdays at 11:00. (As I remember this time was chosen carefully: you want to have the meeting as early as possible in the week so that people could get new directions and go in the right direction as much as possible. But you can't have the meeting on Monday, not enough time to get all the facts. 11:00 was chosen because many folks didn't get in early in the morning; also starting at 11 gave people a strong reason to finish promptly at noon, so they could have lunch.) The staff meetings had an invariable format: Corby talked for 30 minutes, summarizing each team's accomplishments and describing the next step. Lots of times we all knew just what he was going to say, and yet it was still important to hear about progress from him, and to get his interpretation of the significance of various milestones. The last 30 minutes of the staff meeting were devoted to a technical talk by some team member, educating the rest of the team about a design or a new tool.
Corby served MIT in a variety of management activities: Deputy Director of the Computation Center, Associate Department Head for Computer Science, Director of Information Services, MIT Network Czar, and for a second time, Associate Department Head for Computer Science.
Corby was a recipient of the 1966 W. W. McDowell Award from the IEEE, the 1980 Harry Goode Memorial Award from AFIPS, and the 1982 IEEE Computer Pioneer Award.
In 1990, Corby was honored with the ACM's A. M. Turing Award "for his pioneering work in organizing the concepts and leading the development of the general-purpose, large-scale, time-sharing and resource-sharing computer systems, CTSS and Multics."
Corby's Retirement Party
A party to honor Corby on the occasion of his retirement from MIT was held on November 22, 1996, at the Wellesley College Club. About 120 people attended, including his family, members of the Multics and CTSS teams, and colleagues from his career as MIT professor and department head.
The first person I saw when I arrived was my old boss Dick Mills, Assistant Director of Project MAC in the 1960s, and later MIT's Director of Information Processing Services. Many others arrived, including:
- Dave Clark (Multics/MAC)
- Marge Daggett (CTSS/MIT)
- Bob Daley (CTSS/MIT, Multics/MAC)
- Stan Dunten (Multics/MAC)
- Robert Fano (Director/MAC)
- Bob Frankston (Multics/MAC)
- Jay Forrester (MIT, director of Whirlwind project)
- Ed Fredkin (Director/MAC)
- Dave Gifford (Multics/MAC, MIT)
- John Gintell (Multics/GE)
- Bob Graham (Multics/MAC)
- Frans Kaashoek (MIT)
- Alan Kotok (MIT)
- Bob Morris (Multics/BTL)
- Joel Moses (MIT)
- Peter Neumann (Multics/BTL)
- Bill Poduska (Multics/GE)
- Bob Rappaport (Multics/MAC)
- Dave Reed (Multics/MAC)
- Sue Rosenbaum (Multics/MAC)
- Dan Roos (MIT)
- Doug Ross (MIT)
- Jerry Saltzer (Multics/MAC, MIT)
- Akira Sekino (MIT)
- Mike Solomita (MIT)
- Earl Van Horn (MIT)
- Steve and Nancy Webber (Multics/MAC)
- Doug Wells (Multics/MAC)
Remarks at dinner were MC'd by Professor John Guttag, who called on many members of the audience. Dick Mills asked, "Who's known Corby the longest?" and remarked that he'd known Corby since 1953, when they both worked on Whirlwind together. Prof. John D. C. Little topped that: he was Corby's graduate school roommate at MIT in 1951. Then Prof. Peter Elias spoke up: he was a Navy communications tech with Corby at Treasure Island in 1947. Peter Neumann contributed some of his inimitable puns (e.g., "Corbatov" instead of "Mazel tov" as a toast). Prof. Guttag read messages from other well-wishers who could not attend, such as Lee Scheffler. Corby's daughter Nancy and his wife Emily spoke. There were gifts, and toasts, and applause. Frans Kaashoek showed a video clip from the WGBH Telecast of "John Fitch, MIT Science Reporter" interviewing Corby about time-sharing.
Corby's not the kind of person you could "roast." Everyone who spoke returned to the themes of how much we'd learned from him, from his uncompromising regard for correctness, his integrity and directness, and his penetration into complex issues. His leadership of the CTSS and Multics teams enabled us to build new kinds of systems that had never been done before. And in his career at MIT as professor and administrator, Corby shared his intelligence, honesty, and concern for truth with thousands.
Professor Fano put it best, when he said, "Corby, you'll have to retire again next year, so we can have another party."
On June 19, 2004, a Multics Reunion and Dinner Honoring Corby was held at MIT. The Corbató Graduate Fellowship was announced. Multicians are encouraged to contribute, as a way of thanking Corby for his influence on our lives and profession.
Here is Corby's letter on the occasion of the shutdown of the last Multics.
12/28/96, updated 06/22/04. Thanks to Jerry Saltzer, Corby, and Joel Moses for corrections & additions.