108 IEEE TRANSACTIONS ON EDUCATION, VOL. 36, NO. 1. FEBRUARY 1993

[3] P. B. Hansen, Brinch Hansen on Pascal Compilers. Englewood Cliffs, NJ: Prentice-Hall, 1985.

[4] B. Hartfield, T. Winograd, and J. Bennett, “Learning HCI design: Mentoring project groups in a course on human-computer interaction,” ACM SIGCSE Bulletin: 23rd SIGCSE Tech. Symp. Comput. Sci. Educ., vol. 24, no. 1, pp. 246-251, Mar. 1992.

(51 I. L. Janis, Wctims of Groupthink: A Psychological Study of Foreign Policy Decisions and Fiascos.

[6] G. M. McCue, “IBMs’ Santa Teresa Laboratory-Architectural design for program development,” IBM Syst. J., vol. 17, no. 1, pp. 4-25, Jan. 1978.

[7] L. M. Northrop, “Software engineering at the undergraduate level,” ACM 20th CSC / 23rd SIGCSE Joint Panel Session P5: Software Eng. Undergraduate Comput. Curricula, Mar. 1992.

[8] H. Pournaghshband, “The students’ problems in courses with team projects,” ACM SIGCSE Bulletin: 21st SIGCSE Tech. Symp. Comput. Sci. Educ., vol. 22, no. 1, pp. 44-47, Feb. 1990.

[9] I. Sommerville, Software Engineering, 3rd ed. MA: Addison-Wesley 1989.

[ lo ] G. Weinberg, The Psychology of Computer Programming. New York: Van Nostrand Reinhold. 1971.

Boston: Houghton-Mifflin, 1982.

Hong Liu received the B S degree with Honors in Computer Science and Mathematics and the M S degree in Computer Science from Hefei Polytechnic University, China, in 1982 and 1984, respectively, and the Ph D degree in Computer Science from Polytechnic University, Brooklyn, New York, in 1990

She joined the faculty of Polytechnic University in 1987 for three years, teaching both the undergrad- uate and graduate courses in various topics such as compilers, programming languages, and switching

and automata. After receiving the P h D degree, she went to University of

courses in compilers, programming languages, and computer networks In the summer of 1992, she was with Columbia University, New York as a Visiting Scholar. Her research interests include programming languages and compilers, commumcation network design, and computer graphics. For each of the areas, she had publications.

Dr. Liu is a member of ACM, ACM SIGCOMM, AND Upsilon Pi Epsilon. She is a recipient of Pearly Brownstein Doctoral Research Award and Westinghouse Research Grant for Woman in EEKS

1 i Massachusetts, Dartmouth, where she is now an Assistant Professor teaching

Computer Science and Engineering Curricula

Abstract- The challenges of maintaining relevancy and cur- rency in computer engineering curricula are examined. The emerging trend to emphasize design experience early on, the need to stress written and oral communication skills, and the need for industry-university collaboration in funding instructional labora- tories are identified as the three main ingredients for success in an undergraduate computer engineering curriculum.

I. INTRODUCTION

NE of the perennial challenges faced by educators is 0 the need to maintain currency and relevancy to their curricula in the face of changing times. The new Infor- mation Revolution, spurred by the development of low-cost and high-speed computers, is touching every facet of our lifestyle. With PC-sized supercomputers, cellular radios, and fax machines becoming as ubiquitous as radios and televisions; with microprocessors controlling such household goods as washing machines, dishwashers and ovens; and fuzzy logic and neural nets becoming common household words, the modern computer and communications technology is playing a pivotal role in the economic well-being of many nations. Has the academic community recognized this new reality? Are the curricula keeping up with the current changes and challenges?

During the past twenty years, this emerging discipline has assumed various names. For convenience, we refer to this here as computer science and engineering (CSE). In addition to the rapid pace of developments in this area, other important forces are at work prompting a need for frequent curricular

b

Manuscript received June 1992. The author is with the Department of Applied Science, Graduate Group in

IEEE Log Number 9206224. Computer Science, University of California, Davis, CA 95616.

V. Rao Vemuri

revisions. The needs of employers and the characteristics of a computer engineering career are continually changing. The prepafation and motivation of students are also experiencing a transformation. These phenomena are not entirely new to our times, nor are they confined to CSE. Engineering educators faced several such challenges in the past and they routinely responded with recommendations for model curricula for four- year undergraduate programs [ 11 - [5] . Indeed, there has never been a shortage of studies on model curricula. (A sample model curriculum is shown in the appendix.) What seems to impede progress is that curricular recommendations have the tendency to exhibit tremendous “implementation inertias.” They resist all but the most incremental changes.

11. THE SYMPTOMS AND SOLUTIONS

The “Phoenix Committee” (at Worcester Polytechnic Insti- tute in Massachusetts), named thus to reflect its mission of completely overhauling the undergraduate engineering curricu- lum, conducted an extensive survey of their EE alumni and identified some important categories of comments received from 300 of the respondents [6]. I conducted my own, albeit informal and anecdotal, survey of some undergraduate and graduate students. The results from these surveys, summarized below, identify the following needs:

1) More practical and hands-on experience, such as a research or design project, to accompany traditional lectures.

2) Courses that emphasize the computer as an engineering design tool, in general, and improved computer literacy, in particular.

0018-9359/93$03.00 0 1993 IEEE

IEEE TRANSACTIONS ON EDUCATION. VOL. 36. NO. I . FEBRUARY 19Y.1 109

Diversity of courses beyond the (computer engineer- ing) major. The scope of this diversity included not only courses from other engineering disciplines, such as thermodynamics, fluid mechanics, materials, and project management, but also topics in general education, espe- cially writing and rhetoric. Cultivation of the art of thinking and communicating clearly. Many “computer-generation” students learn to communicate only with their terminal or workstation. They are ill prepared in written and oral communication skills as well as in interpersonal skills. A philosophical approach that makes the student realize that the whole is more than the sum of the parts. The body of knowledge a student should learn is often parcelled into semester- or quarter-long courses. Many students have difficulty realizing the interrelationships among these subjects or how one uses the knowledge thus acquired to solve a practical problem.

The following remedial steps were often suggested by the respondents of the above survey. First, requiring the students to attend at least two sets of undergraduate seminars in two different terms. Unlike graduate seminars, these are tutorial and expository in nature with the topics drawn from contempo- rary research issues, but presented in an exciting and thought provoking manner. The speakers, drawn from both industry and academia, should be carefully selected for their ability to inspire and excite, with the speaker’s own contribution, to the topic under discussion, being a secondary consideration. The idea is to offer one of these seminars in the freshman year, the formative period in a student’s undergraduate life, in order to assist the student to make an informed decision in selecting a major. Students are required only to listen and make a brief report at the end of the term. The goal here is to sustain and nourish any budding interest in the sciences (say, CSE), thus forestalling the possibility of students switching majors or even dropping out from engineering.

The second remedial step is to require all freshman en- gineering students (native as well as foreign) to take a one term course in writing or rhetoric [7]. The third step is to require students to write a term paper, not in lieu of any written laboratory reports in a design engineering course. This can conveniently be done in conjunction with the second seminar course suggested earlier; the term paper topic could be one inspired by the topics of the seminar. This is typically done in the junior and senior years. Selecting a topic, doing a literature search, scoping the problem, doing the problem without knowing the “correct answer,” documenting the results concisely, and making a brief presentation of the results in front of a class are skills that are as important as writing a correct program and proving it correct.

111. IMPLEMENTATION

How do we compress all this course work into a four-year curriculum? Even the best of students find it necessary to spend 12 hours per course per week. Further demands on a student’s time will only be counterproductive as there is less time left for reflection and contemplation. This brings us to one of the frequent challenges faced by curriculum

designers, namely striking a compromise between what is basic and what is current while balancing the department’s ever shrinking budgets. Preserving the basics shall always be the prime objective of any curriculum. Pursuing this objective in its purest form has its pitfalls; rote memorization exercises and drills will surely turn off the student seeking a challenge.

Currency and relevancy in computing could be influenced by cooperative efforts between industry and universities. The common complaint that academia really does not understand the problems that are important to industry is, I believe, moti- vated by short-sighted considerations. The prevailing “client- server model” that portrays the student as the product of the “university factory” for consumption by the industrial CUS- tomer is wrong. Corporations, with an enlightened self interest, should participate with universities in the development of the necessary instructional infrastructure. These efforts should go beyond the traditional cooperative education, it should also include the development of instructional and research labora- tories, an area where severe budgetary disincentives exist.

APPENDIX

The Computer Society of the IEEE published a model curriculum for CSE in 1977. By 1981, the society recog- nized the changing trends of the times and commissioned a committee (on which this author served as a member) with the charge to make a major revision. This committee’s report, The 1983 IEEE Computer Society Model Program in Computer Science and Engineering appeared in the April 1984 issue of the COMPUTER magazine, along with several other articles on the role of computers in education. The committee eschewed identifying a single complete curriculum spanning four years. “They preferred instead to identify a body of knowledge that should be a part of any curriculum. The criteria for selecting the fundamental concepts-the core of the curriculum-were to provide a student with broad background in engineering principles coupled with in-depth knowledge of hardware, software, application tradeoffs and the basic modeling techniques used to represent the computing process. Another criterion for the core was that it stay within a 33-semester credit hour limit.” With these objectives, the following 13 “subject areas” were identified as the Core of the Model Program. The Lecture/Recitation components of the Core Model Program are:

1) Fundamentals of Computing, 2) Data Structures, 3) System Software and Software Engineering, 4) Computing Languages, 5) Operating Systems 6) Logic Design, 7) Digital System Design 8) Computer Architecture, and 9) Interfacing and Communications.

1) Introduction to Computing Laboratory, 2) Software Engineering Laboratory, 3) Digital Systems Design Laboratory, and 4) Project Laboratory.

The Laboratory components of the Core Model Program are:

1 IO IEEE TRANSACTIONS ON EDUCATION, VOL 36, NO. 1, FEBRUARY 1993

Then the committee went on and provided a detailed descrip- tion of each of these core areas.

The committee also listed 15 advanced subject areas, for in- depth study. The committee felt that an individual institution should string together these subject area modules in ways that are suitable to their individual needs. That no other Model Curriculum came out of the Computer Society during the past decade attests to the apparent satisfaction felt by the Society with its 1983 Model Curriculum.

[3] ACM Education Board: Commit. Comput. Curricula, ACM Recom- mended Curricula for Compufer Science and Information Processing Programs in Colleges and Universities 1968-1981. New York: ACM Press, 1981.

[4] J . T. Cain, G. G. Langdon, Jr., and M. R. Varanasi, “The IEEE computer society model program in computer science and engineering,” Computer, IEEE Computer Society Press, vol. 17, no. 4, pp. 8-17, Apr. 1984.

[5] “The ACM Response: The scope and directions of computer science,” CACM, vol. 34, no. 10, pp. 121-131, Oct. 1991.

[6] D. Christiansen, “Do students really get it?”, IEEE Spectrum, vol. 20, no. 6, p. 17, June 1992.

[7] V. Booth, Communicating in Science: Writing and Speaking. Cam- bridge, MA: Cambridge University Press, 1985.

ACKNOWLEDGMENT

The author would like to thank Sunil Vemuri, a graduate L - student at Stanford University; Mytilee Vemuri, a freshman at the University of California at Davis and Jim Shackelford, Associate Dean of Undergraduate Studies, University of Cali- fornia, Davis for helpful discussions.

REFERENCES

[ l ] Nat. Acad. Eng.: Cosine Commit., Commission on Eng. Educ., Com- puter Science in Electrical Engineering, Washington, DC, 1967.

[2] IEEE Computer Society: Educ. Commit., Model curricula in Com- puter Science and Engineering, Rep. No. EH0119-8, IEEE CS Press, Los Alamitos, CA Jan. 1977.

V. Rao Vemuri received the Ph.D. degree in engi- neering from University of California, Los Angeles, in 1968.

His ten years of industrial experience includes assignments at Bhilai Steel Works, RCA, Environ- mental Dynamics, and TRW. Prior to his current assignment, he taught at Purdue University, West Lafayette, IN and at the State University of New York in Binghamton.

Prof. Vemuri served the Computer Society of IEEE in many capacities including a membership on

the committee that produced the 1983 Model Program in Computer Science and Engineering. He also served as the Editor-in-Chief of CS Press.

Teaching Design Thro.ugh Computation Garret N. Vanderplaats

b I. INTRODUCTION

OMPUTERS have been commonplace in engineering for C nearly three decades. Today, personal computers are so powerful and so cheap that it is reasonable to expect each engineering student to have his or her own computer. Indeed, the PC on my desk provides me with much more computa- tional power than I had available on a NASA mainframe when I graduated in 1971. Without moving from my desk, I have complete word processing and desktop publishing, as well as immense scientific computing, available for a cost well below that of a cheap used car.

In an effort to offer perspective to my comments here, some background is in order. I was trained as a Civil Engineer, worked for eight years for NASA as an Aerospace Engineer, taught Mechanical Engineering (I am a registered ME) for eight years, and now am affiliated with a small high tech engineering company specializing in design optimization soft- ware. While both of my brothers became Electrical Engineers, my own background in this specialty ended when Laplace

Transforms got the best of me. Thus, I offer the perspective of the generalist with specialization in numerical optimization methods, which themselves span virtually all engineering disciplines. Having been a student for 21 years, an engineer for 8, an educator for 8 and an engineedmanager for 5, I naturally consider myself to be an expert on the topic at hand. However, by virtue of my background and disposition, I expect that my view may not be consistent with that of most academicians.

In order to address the teaching of design through com- putation, I will first offer my perspective of the present state of engineering education. I will then suggest a general approach to engineering education, which is independent of any particular discipline. Finally, some words will be offered about how I think change may be made.

11. ENGINEERING EDUCATION TODAY

First, consider the strengths of our engineering education. Having traveled to various countries, both in Europe and Asia, I’m convinced that our educational system .is the best in the world. It’s certainlv true that several maior countries

Manuscript received June 1992. outperform us in some areas. Mathematics, is the most notable example, and there’s little debate that we need to improve. The author is with VMA Engineering, 5960 Mandarin Avenue, Suite F,

Goleta, CA 93117. IEEE Log Number 9205783. However, there is an important trade off. It seems to me that

0018-9359/93$03.00 0 1993 IEEE