Concepts familiar from grade-school algebra have broad ramifications in computer science.
There's a slight shift occurring in the Course 6 educational paradigm. It's not really a revolution, but it could affect the evolution of some of the graduate students.
"The Structure of Engineering Revolutions," a course offered for the first time this fall, provides an integrated approach to engineering, one that considers the interplay between socio-political influences and technological developments. The course is listed jointly in the Program in Science, Technology, and Society (STS) and the Department of Electrical Engineering and Computer Science (Course 6 or EECS).
In true engineering fashion, Dr. David A. Mindell, an assistant professor in STS who is a historian of technology as well as an electrical engineer, designed the course to solve a specific problem.
"I asked Course 6 what kind of educational problems they were facing that an STS course might be able to address. They said they wanted to give the master's students a better idea of how engineering works in the real world, familiarize them with working in teams and improve their communications skills," said Professor Mindell.
The new course, STS.185/6.972, does all those things through readings in sociology and history of technology, class discussions and lectures, and group projects presented to an audience that includes experts who respond to -- and sometimes challenge -- the presenters. It is co-taught by Professor Mindell and Professor of Computer Science Charles Leiserson.
"The incoming students were naive with respect to real-world engineering. This course teaches them about the technical and nontechnical forces that contribute to design choices," said Professor Leiserson. "Developing the course with STS is a high priority of my department."
Engineering students may learn about black boxes early on, but many years can pass before they understand that their work takes place in a sort of conceptual box, the engineering equivalent of Thomas Kuhn's scientific paradigm.
The group projects encourage students to look at innovative electronic machines as artifacts of a particular time and place, rather than seeing them as naturally occurring developments in a linear history of technology.
Students choose a single technological artifact or a particularly innovative engineering firm and make it the subject of a thorough investigation. They try to recreate the social and political world of the historical engineers and metaphorically climb into the conceptual box with them -- to see the process behind the engineering artifact from the designers' perspective.
SHOOTING FOR THE MOON
For instance, one group tackled the Apollo Guidance Computer (AGC) that was developed in the early 1960s by the MIT Instrumentation Laboratory (IL), the precursor of today's Draper Lab. By studying popular publications, news sources and the IL team's own AGC documentation, as well by interviewing key IL design engineers, they learned that AGC's conception was fraught with conflict.
Not only can political and social pressures be the primary impetus for technology development, but the tension among competing groups often influences design decisions. In turn, the artifacts' design can have major ramifications for future technology.
"The American people saw the Apollo program as the country's manifest destiny. The reality was that President Kennedy was desperately clutching at something to save face in a political arena where his country was losing," said Mohan Gurunathan, a graduate student in EECS, during his group's presentation on December 2. His teammates were fellow EECS graduate students Jennifer Kleiman, Matt Lau, Chris Rodarte and Keith Smith.
In their presentation, they described how the Cold War struggle for military power led to the desire for space supremacy and finally to the development of technology capable of achieving that national goal. The United States reacted quickly after Soviet cosmonaut Yuri Gagarin made the first manned orbit of Earth in 1961.
"Against the advice of his scientific advisors," Mr. Gurunathan said, "Kennedy declared that the United States would be the first to walk on the moon. The scientific objections were buried away in classified documents."
The students asserted that America's need for heroes greatly influenced the design of the Apollo Guidance Corp. computer, forcing the engineers to allow the astronauts more manual control over the guidance system than the engineers thought necessary.
"Tensions arose between the astronauts and the engineers. The engineers believed the computer was capable of recalibrating itself, but the project administrators wanted the astronauts to have control. So they were given the illusion of control," said Mr. Smith. The astronauts were responsible for adjusting the computer's location information. But the adjustments themselves were actually computer-assisted because the computer controlled the optics.
"It's debatable whether or not the AGC could have flown the entire mission without the need for external correction," he said.
Mr. Smith also discussed the decision process behind the inclusion of integrated circuitry in the AGC, which was then an "immature technology."
Donald MacKenzie of the University of Edinburgh, a sociologist of science and technology who is visiting at Harvard University, served as commentator for the AGC presentation. He noted that the decision to include integrated circuitry in AGC prompted long-term technological change. The resources put into developing that particular technology for such a high-profile project served as an endorsement for further development of integrated circuits.
Another group presentation followed the trajectory of the Thinking Machines company from its 1983 founding in the "AI paradigm" through a shift of focus to scientific computing in 1989 to bankruptcy in 1994. Other teams studied RSA Data Encryption, the development of spreadsheets, and magnetic core memory.
Professors Leiserson and Mindell were pleased with the students' progress over the term, "especially their ability to synthesize and make persuasive arguments and rigorously support them with evidence," said Professor Mindell.
Although it was experimental this fall, STS.185/6.972 will become a regular part of the Course 6 curriculum next year, offering students a way to broaden their engineering education.
"This represents a statement by MIT that the best engineers combine the highest technical abilities with a broad perspective and the ability to communicate," said Professor Mindell.
A version of this article appeared in MIT Tech Talk on December 10, 1997.