The Planner programming language, developed during the late 1960s, introduced a procedural embedding of knowledge that provided an alternative to existing logical approaches in artificial intelligence. This project served as a central pillar in the career of Carl Eddie Hewitt, a computer scientist and mathematician whose research spanning over three decades influenced the development of concurrent computation and logic programming.
Academic Foundation at MIT
Carl Eddie Hewitt earned his PhD in mathematics at the Massachusetts Institute of Technology in 1971. His doctoral supervisors included Seymour Papert, Marvin Minsky, and Mike Paterson. Following his graduation, he joined the MIT faculty within the Department of Electrical Engineering and Computer Science, where he taught until the 1999–2000 academic year. Upon his retirement, he was granted emeritus status in the department. Throughout his tenure, he mentored numerous doctoral students, including Gul Agha, Henry Baker, William Clinger, Irene Greif, and Akinori Yonezawa.
Twenty questions, eight minutes on the clock, and a percentile measured against everyone who has taken it. No sign-up.
Take the IQ test →Development of Planner
During his time in the MIT Artificial Intelligence Laboratory, Hewitt worked on Planner, a language designed for automated planning and theorem proving in robots. Unlike previous logical methods favored by researchers such as John McCarthy, Planner utilized pattern-directed invocation from assertions and goals. A subset of the language, Micro-Planner, was implemented by Gerry Sussman, Drew McDermott, Eugene Charniak, and Terry Winograd, subsequently powering programs such as SHRDLU and various natural language story understanding experiments. Hewitt later collaborated on Muddle, an extended version of Lisp intended to refine the implementation of Planner concepts.
The Actor Model and Concurrent Computation
In 1972, Hewitt shifted his focus after developing the actor model of concurrent computation alongside his graduate students. Formally introduced in a 1973 paper co-authored with Peter Bishop and Richard Steiger, the model proposed a new framework for parallel processes. This work influenced the creation of the Scheme programming language, where researchers Sussman and Steele noted that their operator for actor creation shared identity with the lambda calculus operator. The actor model subsequently informed the development of various languages, including E, SALSA, and ActorScript, as well as the theoretical π-calculus.
Professional Outreach and Later Research
Hewitt maintained an international academic presence, serving as the IBM Chair Visiting Professor at Keio University in Japan from September 1989 to August 1990. He also held a visiting professorship at Stanford University. In his final decade of research, he focused on inconsistency robustness, aiming to create rigorous foundations for information systems that handle pervasively inconsistent data. His academic output spanned diverse areas such as open information systems, paraconsistent logic, cloud computing, and organizational multi-agent systems.
Fast facts
- Born: 1944
- Died: 2022, Aptos
- Citizenship: United States
- Educated at: Massachusetts Institute of Technology
- Primary Field: Computer Science
- Notable Work: Planner
- Academic Supervisor: Marvin Minsky
- Visiting Professor: Keio University
Questions readers ask
What was the purpose of the Planner language?
Planner was designed for automated planning and proving theorems, specifically for use in artificial intelligence and robotics.
How did Hewitt contribute to concurrent programming?
He developed the actor model, a formal system for concurrent computation that influenced multiple programming languages and the development of the π-calculus.
Achievements
- Notable work: Planner
- Held posts at Massachusetts Institute of Technology
- Fields: computer science
