EECS eyes curriculum overhaul
By Jeremy Hylton
The Department of Electrical Engineering and Computer Science is considering the addition of a five-year Masters of Engineering program that would effectively replace the bachelor's degree as the primary professional program offered by the department.
"Barring unforeseen developments, we hope to obtain the necessary Institute approvals to describe the new curriculum in the next MIT catalog, for which copy is required by January 1992," Professor Paul L. Penfield Jr. ScD '60, head of the department, wrote in a memo to the faculty.
The EECS faculty met yesterday to discuss a proposal for the overhaul of undergraduate and master's level degrees. The department's Ad Hoc Curriculum Committee, appointed earlier this year, developed the proposal discussed yesterday. It extended the work of the First Profession Degree Committee formed in 1989 to study a five-year program. The FPDC also conducted department discussions during the 1990-91 school year and held a summer study in 1990.
Most of the EECS faculty in attendance supported changing the current degree program, but not necessarily the specifics of the proposal, according to Professor of Electrical and Engineering and Computer Science Campbell L. Searle SM '51, chairman of the Ad Hoc Curriculum Committee.
"What I heard were two or three very negative comments and the rest were positive," Searle said after yesterday's meeting. "The most serious objection is that we were not ready."
Several professors objected strongly to the less rigorous thesis required for the master's degree. Professor of Electrical Engineering and Computer Science Jeffrey H. Lang '75, associate director of the Lab for Electromagnetic and Electronic Systems, described the proposal as "the scuttling of the master's thesis."
The criticism was expected. Ford Professor of Engineering William M. Siebert '46, another author of the proposal, explained that the meeting was designed to focus on problems with the proposal. "We focused discussion on certain kinds of things that would elicit negative reactions," he said.
Nevertheless, professors had praise and ideas for the proposal. Michael L. Dertouzos PhD '64, professor of electrical engineering and computer science and director of the Laboratory for Computer Science, said he "overwhelmingly supports the idea of moving to a five-year program."
Dertouzos suggested the department should do more to encourage students to student economics, management and manufacturing. Professor of Electrical Engineering Peter Elias '44 suggested that the program allow for specialization in management.
Program would require 20 classes
The program outlined yesterday would consist of an 18-course curriculum beyond the 17 General Institute Requirements. The focal points of the program would be the common core, largely similar to the current EECS core, and three "depth area strings."
The strings would be three-course sequences within several sections of the department, including artificial intelligence, electrodynamics and energy systems. Each depth area would have a header class that is a prerequisite for all other classes in the area.
The common core would also include Differential Equations (18.03) and a probability class.
The M.Eng. degree would be the principal degree offered under the new program, but a bachelor's degree would still be offered after four years. At the end of junior year, all students in good academic standing would be accepted into the M.Eng. program. A thesis would be required of all M.Eng. students, but not of bachelor's candidates.
"By extending the program to five-years, you get not only greater breadth and depth, but greater flexibility for students and faculty," Siebert said. The proposal would increase the number of courses required by the department from 13 to 20, including two science distribution subjects.
The proposal's third author, Professor of Electrical Engineering and Computer Science John V. Guttag, echoed Siebert's sentiments. The new math classes in the common core and the depth strings would guarantee that students in advanced classes would have a common background.
"I think this sort of thing will make a big difference in the long run," Guttag said. "If the content of the subjects doesn't change, we will have missed a great opportunity."
Grad student population
would increase by 100
The number of graduate students in the department would increase by about 100, according to Searle. He estimated 15 or more teaching assistants would be need to handle the extra 100 students. He also claimed the reduced scope of M.Eng. theses would decrease the faculty's supervisory duties.
Financial support for the five-year program would be limited to about 75 students. While students would be accepted into the program at the end of their junior year, financial support would not be guaranteed.
"Support is merit-based, which I think we need to maintain quality," Searle said. "The system is going to be support limited."
Financial support would come from teaching and research assistantships paid for by the tuition income generated from the 100 new students, an expansion of the VI-A internship program and other industrial liaisons. "We would be offering support to 75 of the 194 or so students," Searle said, citing the class of 1991 as an example. About 194 students would have been eligible for the M.Eng. program.
The quality of students in the program was a point of contention at yesterday's meeting, however. "We will be dealing with a diluted student body," said Professor of Electrical Engineering Stephen D. Senturia PhD '66. Many of the best students are admitted to the doctoral program or go elsewhere, he said.
Clifton G. Fonstad Jr. PhD '70, professor of electrical engineering and computer science, concurred. "The lower students will all show up on the doorsteps," he said.
The proposal aimed to have the minimum impact on the doctoral programs already in place. Students in the M.Eng. program would be in a terminal master's program, much like the the VI-A program.
Graduate students who receive undergraduate degrees from other institutions could apply for admission into either the Master's program or the standard doctoral program. "Students coming from outside MIT with electrical engineering and computer science degrees would be able to complete the M.Eng. program in the same time as out students," Siebert said.
Assistant Professor of Electrical Engineering and Computer Science Lynn A. Stein questioned the preparation of computer science undergraduates from outside MIT. They may lack a strong science and engineering background, but still be strong computer science students, she said.
The strongest concern voiced about the master's program was the reduced scope of the thesis. Like Lang, Professor of Electrical Engineering and Computer Science Mike Athans said, "I am very bothered that we're not going to see as much depth in the master's thesis."
Change could be mandatory
for Class of 1996
Penfield asked each member of the EECS faculty to write him a letter explaining whether they support a new five-year program and what problems they have with the current proposal.
"What I hope to get out of the days that follow is a department consensus," Penfield said. He also said that the details of the new program are not final yet, and that copy submitted for the 1992 catalog could be revised later in the semester. Penfield plans to petition the entire Institute faculty for permission to offer the new M.Eng. degree. The motion would be voted on in February.
"We have to get something in the catalog to get the clock running," Searle said. "It does not have to be that detailed."
If the program is approved, a small number of current students could enter the program. According to Searle, most of the current juniors and sophomores could complete the M.Eng. program in one extra year. Financial support would be available for 5 juniors, 35 sophomores, and 60 freshmen during their fifth year.
"In terms of the support issue, we may want to make it optional for this year's freshman and mandatory for next year's entering class," Penfield said.
Even if the program is in place for the class of 1996, MIT will not be the first university to have a five-year master's program, but it may be one of the most influential programs. "I think it's an important step. In 20 years, we will go the route of the medical profession and have extended professional training and specialization," he said.
Two years ago Guttag discussed a similar proposal with a group of heads of computer science departments from across the country. "They thought this was a good idea and they thought it was a good idea for MIT to be the first," he said. Penfield has received similar reactions from electrical engineers. "They have been very interested in what we are doing," he said.