THE PROFESSION
An “OffshoreResistant” Degree Program David A. Swayne, Qusay H. Mahmoud, and Wlodek Dobosiewicz University of Guelph
T
he new University of GuelphHumber has created an innovative, combined university degree and college diploma program in computer science and communications technology. This venture between the university and the Humber Institute of Technology and Advanced Learning offers jointly an honors baccalaureate in computing science from Guelph and a diploma in wireless and telecommunications systems technology from Humber. The program, one of six joint programs between the two institutions, specializes in distributed and wireless computing.
PROGRAM DETAILS The program, which comprises eight semesters of full-time study plus two eight-month cooperative education work placements, is divided into two distinct phases of equal duration: • a lower division of two years, in which courses cover the basics of programming, mathematics, physics, and communications systems; and • an upper division that consists of the two cooperative work placements and material designed to deliver a highly specialized and detailed program of study in advanced communications technologies and distributed computing. 104
Computer
believe our program’s integration of these two disciplines will produce graduates whose skills cannot be easily outsourced. The “A New University Model” sidebar describes in detail how we crafted this program. There is nothing to prevent our program from being emulated by others, particularly from other countries. However, we feel that many of the applications we address must remain close to the source. Examples of such applications would be secure communications—applications with hardware and software dependencies. When all else fails, the need for the hardware to
Will proficiency in both computer science and communications give students a global edge?
Figures 1 and 2 show the curriculum’s largely self-explanatory course names. Two calculus courses, an applied algebra course, and a probability and statistics course cover elementary mathematics. A course in ethics forms part of the third semester’s curriculum. The first five programming courses give a fairly complete picture of the underpinnings of programming, algorithms, and elementary data structures. The requisite physics courses for communications consist of the two firstyear electronics courses and the fourthsemester electromagnetics course. Physical application principles dominate the courses in communications systems. The University of Guelph-Humber’s combined-program concept offers a distinct advantage over the standard undergraduate computer science model because it integrates two different aspects of computer applications. Students will graduate with skills in both computing and communications, ensuring that a broad spectrum of positions will be available to them. We
be situated locally also carries the natural need to maintain the software locally. That is, these sorts of students will be most useful in projects that do not travel well.
Faculty and staff The program draws its faculty members evenly from Guelph and Humber, with three Guelph faculty members currently at the new campus. New staff members have been hired to support the program. The University of GuelphHumber has its own staff to handle student recruiting.
Students Most of Guelph-Humber’s students came from Ontario province. While some had no exposure to programming, many were self-taught and some even had industry certificates such as those granted by the A+ and Microsoft Certified Systems Engineer (MCSE) programs. We accepted 36 students for Continued on page 102
The Profession Continued from page 104
Year 1
Year 2 Semester 1 F2002
Procedural Programming 1 Electronics Fundamentals Telecommunication Computations Calculus for Computing I Elective
Semester 3 F2003 User-Centered Programming 3 Principles of Telecommunications Communication Systems Data-Centered Programming 4 Ethical Issues in Info. Technology Workplace Prep and Tech. Writing (half course)
Semester 2 W2003 Intro. to Object-Oriented Programming 2 Semiconductor Electronics Calculus for Computing II Anatomy of a PC Digital Systems
Semester 4 W2004 Data Structures 5 Probability & Statistics Data Networks Electromagnetics Elective
Figure 1. Lower-division program, programming courses numbered 1 through 5.
Work 1 (Spring 2004) and Work 2 (Fall 2004)
Work 3 (Fall 2005) and Work 4 (Winter 2006)
Year 3
Year 4 Semester 5 W2005
Operating Systems Distributed Programming Digital Signal Processing Fibre Optic Communications Telecom Elective
Semester 6 S2005
Semester 7 S2006 Distributed Software Engineering Internet Technologies Senior Paper I (half course) Telecom Elective DCCT Elective DCCT Elective Semester 8 F2006
Database Systems & Concepts System & Network Simulation Wireless Data Networks Cellular Technology DCCT Elective
Distributed Info. System Architectures Telecom Elective Telecom Elective or CS Elective Technology & Society Senior Paper II
Sample Telecom Electives: Broadband Communications Telephone Switch Technology
Sample CS Electives: Cryptography and Network Security Artificial Intelligence
Figure 2. Upper-division program.
the Fall 2002 semester, 30 of whom are now in their second year. We also accepted 61 students for the Fall 2003 semester, 54 of whom have successfully finished their first year. Second-year students will be away from May through December 2004 on an eight-month work placement, then they will return in January 2005 for two more semesters of study. 102
Computer
Staying small Part of the Guelph-Humber equation relies on a funding formula based on a maximum class size of 65 students and a maximum laboratory size of 35 students. This improves significantly on typical university class sizes in Ontario, particularly with the double cohort—a situation created by the elimination of the 13th year of fully
funded public education. It is a mixed blessing, though. From a selfish perspective, high attrition rates would lead to the early demise of the program. With this in mind, we framed retention strategies that focus on the introductory programming language and system choices and our control over the mathematics curriculum.
PROBLEM AREAS Our experience teaching technological programs at several universities led us to believe that at least three main subjects have excessive failure rates or bad outcomes, defined as a D grade, a fail, or a drop: calculus, computing, and chemistry—the so-called three Cs. Computing science programs typically have only calculus and computing. Apparently, the damage usually is done in the first semester. Students who survive often have their cumulative averages reduced and their careers blighted by difficulties that frequently evaporate when they catch on to the subjects. Thus, our curriculum focuses on building expertise and maintaining a sense of accomplishment, rather than discouraging and frustrating students.
The programming problem We recognize that it is better to become proficient at one language and its environment than to risk confusion through diversification. Therefore, our first four programming-language courses focus on the use of HTML, JavaScript, and Java. We stay with Java through the third semester, introducing user interfaces, elementary data structures, and sockets programming in Java. Then, in the fifth course, we adapt the programming for data structures to C++, combining the learning of data structures and algorithms with a new but very similar language—but doing so in a semester in which the curriculum covers no other major programming component. Our most important retention strategy is programming for fun (Qusay H. Mahmoud, Wlodek Dobosiewicz, and David Swayne, “Making Computing
Programming Fun and Flexible,” Computer, Feb. 2004, pp. 108, 106107) and otherwise managing student expectations in the first semester. We achieve this by encouraging and controlling collaboration in programming, facilitated by using programming tools that give good feedback to the student.
The calculus problem We recognize that far too many students drop out of computer science programs for the wrong reasons—the transition from secondary to postsecondary school being a principal one. We decided to exercise control over the mathematics courses to ensure that whatever time is available to the students is well spent. Calculus, particularly in the first semester, involves three hours of lectures, three hours of labs and tutorials, and one hour of testing per week. Extensive training in the use of tools such as algebraic software empowers the students to solve rather dramatic problems that would only devolve into messy calculations if solved using pencil and paper. he time has come for computer science to become interdisciplinary and for computer science departments to offer innovative programs that will produce graduates whose skills cannot be outsourced. Our program started after the dotcom crash. Because of the Ontario Access to Opportunities Program (ATOP), which sought to have all computer science programs in the province double in size over four years, we are competing for a shrinking pool of students in a time of computing industry retrenchment. Our strategies include staying small, community building, and bolstering retention without sacrificing integrity. The University of Guelph-Humber has been a huge success. As of this writing, more than 80 percent of the computing students who finished their second year have found co-op jobs with medium to large companies, such as CIBC, RIM, and Sybase. ■
T
A New University Model From 1999 to 2001, the presidents of the University of Guelph and the Humber Institute of Technology and Advanced Learning negotiated an agreement to offer joint programs sanctioned by the Ontario Ministry of Training, Colleges, and Universities. This venture received $28.63 million in funding as part of the government of Ontario’s SuperBuild initiative to meet the needs of the increasing postsecondary student population. The Department of Computing and Information Science at the University of Guelph and the Faculty of Information Technology and Accounting at Humber thus entered into co-development of the Wireless program. Originally, Wireless provided a package for renovating existing programs at both institutions. Resistance to the program at Guelph and the constraints imposed by squeezing the joint program into eight semesters posed an immediate threat to this form of advancement by small steps, however. Development thus began in the fall of 2000 on a completely new program, and continued until December 2001. This program subsumed earlier efforts and expanded the Wireless notion into a broader study. The development team obtained requisite if guarded approval to proceed with the new program, whose calendar designation became its abbreviated name—Distributed Computing and Communications Technology or DCCT. The team developed a preliminary curriculum and several analyses to show how the proposed degree program would enhance the University of Guelph’s learning objectives and the development of generic skills. In December 2001, we held a workshop with participants from both institutions as well as guest presenters Ralf Denzer and Ernst-Erich Doberkat, professors who had developed similar curricula in Germany. Significant revisions to the Wireless program resulted from this workshop. Construction of a joint Guelph-Humber building began in November 2001, and the building opened in the fall of 2003. The University of Guelph-Humber became the first in Ontario to offer both a university degree and a college diploma in just four years of study. The DCCT was one of the three inaugural programs that, from Guelph’s point of view, provided a significant response to the Ontario government’s double cohort, a situation created by the elimination of the 13th year of fully funded public education.
David Swayne is a professor in the Department of Computing and Information Science, University of Guelph, and program coordinator of the Distributed Computing and Wireless & Telecommunications Systems Technology program, University of GuelphHumber. Contact him at dswayne@cis. uoguelph.ca. Qusay H. Mahmoud is an assistant professor in the Department of Computing and Information Science, University of Guelph, and associate chair of the Distributed Computing and
Wireless & Telecommunications Systems Technology program, GuelphHumber. Contact him at qmahmoud@ cis.uoguelph.ca. Wlodek Dobosiewicz is a professor and chair of the Department of Computing and Information Science, University of Guelph. Contact him at
[email protected]. Editor: Neville Holmes, School of Computing, University of Tasmania; neville.holmes@ utas.edu.au
August 2004
103
