This proposal is to establish a program leading to the award of the degree of Doctor of Philosophy in Computer Science. It should be noted that a Ph.D. program was recommended by the Graduate School Administrative Review Panel that examined the Department's graduate programs in December 1988. A recent review suggested the development of a Ph.D. program.

The reality is that Howard University must have a Ph.D. in Computer Science if it is to remain competitive at the Doctoral Research Extensive level (formerly known as Research Level I). It is likely that Howard University is the only such research university that does not offer a Ph.D. in Computer Science. A recent symposium held under the auspices of the National Academy of Sciences was entitled “The Impact of Information Technology on the Future of the Research University;” the title is illustrative of the importance our competitors (and our country’s leading experts) place on this essential area. Clearly, having a “digital divide” at the policy-making levels of industry and government is a more complex problem and, in the view of the department, is even more serious than the lack of access to computing by disadvantaged persons (the other “digital divide”), since the latter problem will be reduced over time by the relentless decrease in the cost of computer hardware.

The primary reason that no Computer Science Ph.D. program currently exists at Howard University is that, until recently, the Department of Systems and Computer Science’s resources have been far too inadequate to realistically consider offering a program. Under the leadership of President Swygert, the department has had new space made available to it and has been able to renovate its laboratories, meeting most of the needs of a growing undergraduate program and making it possible for the department to seriously consider this new endeavor.

The department’s research effort has also increased, with nine current members of the Graduate Faculty instead of the two from several years ago. As of February 2002, the department has the highest proportion of its faculty holding appointments on the Graduate Faculty of any department within Howard University. We are still severely understaffed, but the level of faculty research has made offering a Ph.D. feasible if sufficient new departmental resources are added. Specific resource needs of the proposed Ph.D. program will be addressed in Part II of this proposal; resources necessary to make the existing Master’s program competitive are also included in Part II.

The new Ph.D. program in Computer Science is designed to:

· Help meet the need for more African-American computer science educators by increasing the supply of African-American computer science doctor's degree holders.

· Help meet the need for more African-American computer scientists in decision-making positions in the domain of computer science by providing advanced technical training and a background in independent research.

· Operate at a level of meaningful size and of the highest quality that can be accomplished with a reasonable increase in Departmental budget.

· Upgrade the Departmental research environment, thereby increasing the Department's ability to attract and retain outstanding Systems and Computer Science faculty.

· Achieve a ranking among the top 50 small departments of Computer Science, ideally reaching this goal within three years of the graduation date of the first Ph.D. students.

The Department of Systems and Computer Science is confident that these goals can be accomplished by the program described in this proposal.

This program will be integrated with the existing Master's program to provide for continued training of Howard's students beyond the Master's degree, and will be an alternative for students from other institutions who desire to pursue such a program at a predominantly minority institution. Howard would become the nation's first HBCU to offer a Ph.D. degree in Computer Science. The program will train individuals to help fill the critical need in academia, government, and industry for minorities with Ph.D.s in engineering and science.

Because of the explosive growth of general awareness of several computer applications software packages, there is often confusion about the nature of the field of Computer Science. It is important to understand the distinction between various aspects of the computing milieu. The following distinctions are common in the Computer Science literature:

Computer Science is the field whose primary concern is the efficient development of both efficient programs and algorithms. Typical training of academic Computer Scientists is a Ph.D. in Computer Science or a related field.
An important subfield of Computer Science is the area of Software Engineering, which is concerned with the problem of developing software systems that are far too large to be developed by a single person. Appropriate design and testing techniques and methodologies for insuring that programs that control power plants, airplanes in flight, medical life support systems, or the telephone system are free of major faults are also studied as part of software engineering. Typical training of academic Software Engineers is a Ph.D. in Computer Science or related field. Industrial practitioners typically have a B.S. in Computer Science or Software Engineering. Related to Software Engineering is the discipline of Systems Engineering.

Computer Engineering is the field whose main focus is the development of computer hardware. The primary concerns may be as small as single computer chips or integrated circuits, or as large as entire computers. Typical training of academic Computer Engineers is a Ph.D. in either Electrical Engineering or Computer Engineering.
Computer Information Systems is the field whose primary concern is large-scale database design. Typical training of academic Computer Systems faculty is a Ph.D. in Computer Information Systems, Computer Science or related field.
Computer Applications Programming is the field whose primary concern is the development of programs that are specific to a particular applications domain. Typical training of applications programmers in industry or government is a B.S. degree. Typically, there are no computer applications programmers who are members of a Graduate Faculty at a research university, although they are typical members of the faculty of a community college.

Rationale

For Howard University to be on the forefront of research in technology, offering a Ph.D. degree in Computer Science is essential; this is consistent with the mission statement of the University (September 24, 1988). The need for such a program exists because there is a shortage of Ph.D. Computer Scientists in general, and especially of African-Americans with advanced degrees in computer science. The demand for African-American Computer Scientists with advanced degrees far exceeds the supply. The College of Engineering, Architecture and Computer Sciences at Howard University currently enrolls more African-American undergraduate computer science students than any other School of Engineering in the country. In view of the Department's success with the undergraduate and Master’s programs, and considering the need for more African-American Ph.D.-level computer scientists, it would be appropriate to initiate such a program in August 2002, providing that sufficient resources are available.

The program in Systems and Computer Science has grown rapidly at Howard in recent years. Despite the fact that there is strong evidence that the greatest need for trained people in the next twenty years will be at the advanced graduate level, the number of African-American Ph.D. in Computer Science has remained small (averaging fewer than 4 per year for most of the past twenty years). The Department of Systems and Computer Science believes that it has the potential as well as the responsibility for offering the first capstone Ph.D. degree in the Computer Sciences at an HBCU. Howard University involvement in Internet II would certainly require a Ph.D. degree in the most fundamental discipline in this area – the discipline that spawned the Internet.

Need for the Program

One of the definitive surveys of the production and employment of Computer Science Ph.D.s and faculty of Ph.D.-granting Computer Science/Engineering Departments in the U.S. and Canada is the annual Taulbee Report, sponsored by the Computing Research Board. According to the most recently published report, the number of Computer Science Ph.D.s awarded annually dropped below 900. This was the lowest output since 1990. However, of the 881 Ph.D. recipients in Computer Science and Computer Engineering, only 14 (less than 2%) were granted to “African-American, Non-Hispanic”persons. Current enrollment in Ph.D. programs is no better. Of the 7,412 students enrolled as Ph.D. students in computer science, 110 (less than 2%) are listed as “African-American, Non-Hispanic.” This is abysmally low, and has not increased significantly even in recent years.

At the same time that the supply of African-American Ph.D.s is remaining flat at best, there is an ever-increasing need in both academia and industry for doctorate-level computer scientists. Consequently, as bad as the situation is now, the percentage of African-American Computer Science Ph.D.'s will decrease even further if all current trends continue.

The goals of the proposed program are simple: offer a high quality Ph.D. program in selected subject areas; by 2008, double the number of African-American Ph.D.'s in the Computer Science area produced in the United States annually; and produce a critical mass of graduate-level support researchers to have a research training faculty with a strong national and international reputation.

We note that there are several fields that are already well represented at Howard University that may reasonably be expected to have some interaction with the Computer Science Ph.D. program; among them are the departments of Electrical Engineering, Physics and Mathematics all of which already have Ph.D. programs. Potential synergisms also exist in working with the Business School and the School of Communications. (An interdisciplinary graduate program in Telecommunications is also being discussed with these five areas.)

Equally important to consider is the foundation that a Ph.D. program in Computer Science would provide for the new paradigm for research in engineering and the sciences, which places considerable emphasis on complex and computationally intensive computer simulation. Simply put, a strong Ph.D. program in Computer Science is essential to Howard University.

Relevant Department History

The Department of Systems and Computer Science (SCS) is an administrative unit of the College of Engineering, Architecture and Computer Sciences (CEACS) and has been in operation since the Fall of 1981. It is the result of a merger of two formerly independent but interrelated graduate programs of the School of Engineering – the Computer Science Program, which was established in 1971, and the Urban Systems Engineering Program, which was established in 1972. The mission of the Department is to develop and maintain first rank programs of education and research in the fields of Systems Engineering and Computer Science, focusing on minority students, and to apply this knowledge base to resolving critical world problems, with particular emphasis on the African-American community, the nation, and the developing world.

Recognizing the great demand for computer specialists and the relative shortage of minorities in this field, an undergraduate program in Computer Systems Engineering was added in 1982. The merger of two independent programs has heavily influenced the Department's orientation and the course offerings, which currently emphasize computer software systems and systems engineering. The name of the degree program was changed to Systems and Computer Science in 1987. The undergraduate program is accredited by the Computer Accreditation Commission (CAC), a subsidiary of the Accreditation Board for Engineering Technology ABET). Previously, the department’s program was accredited by the Computing Sciences Accreditation Board (CSAB), which has been subsumed into ABET.

With a consistent pattern of growth in the quantity and quality of its students, both undergraduate and graduate; improved laboratory support; and a continued improvement in research productivity of the faculty, the department is poised to go to the next level – offering a Ph.D. program.

Strategic Plan for the Graduate Program

To help focus its academic and research programs, the Department of Systems and Computer Science has developed the following goals:

· To educate students for a fully productive self-fulfilling and creative role in society by offering a rigorous and contemporary program of studies in the Computing Sciences.

To impart to our students the professional competence and confidence to seek leadership positions in industry and society, to assume faculty and research positions at national and international institutions, and to serve as role models in their communities.
To conduct plans and outreach programs that will respond to the needs of the African-American Community, the Nation, and the developing world.
To pursue and conduct research programs that contribute to the state of the art in Computer Science, and serve as a vehicle for education and the development of our Department.

Computer Science is one of the fastest changing disciplines. The field is changing so rapidly that the Computing Research Association (CRA) has stated the following in its web site http://www.cra.org/reports/tenure_review.html (and in Computing Research News):

“The reason conference publication is preferred to journal publication, at least for experimentalists, is the shorter time to print (7 months vs. 1-2 years), the opportunity to describe the work before one's peers at a public presentation, and the more complete level of review (4-5 evaluations per paper compared to 2-3 for an archival journal) [Academic Careers, 94]. Publication in the prestige conferences is inferior to the prestige journals only in having significant page limitations and little time to polish the paper. In those dimensions that count most, conferences are superior.”

This statement is consistent with the previous report of the National Research Council on research in the various computing disciplines in 1994. It is important to address the way that this statement affects the nature of the proposed Ph.D. program because of the composition of the current and projected program faculty.

The current composition of the Department’s faculty is much less weighted toward senior faculty than in most existing Computer Science Ph.D. programs and new faculty hires are projected to be at the Assistant Professor level. At first glance, this may appear to be a severe weakness. However, the rapid rate of change in technology, coupled with the excellent industry relationships of the department, make this lack of seniority more of an advantage than might be the case in other departments.

The most pressing need of the graduate program is additional resources for support of basic instructional and research in the beginning graduate programs. It is assumed that advanced research needs will be met primarily by the efforts of researchers obtaining external support.

The essential obsolescence of computer hardware and software in approximately two or three year periods requires an ongoing commitment from the University for basic support of graduate programs. In addition, any lack of current hardware and, more importantly, up-to-date versions of several key software packages would make it difficult to pursue the potentially lucrative pool of part-time graduate students from local industry and government. Such students are very important to graduate programs, since they provide:

· excellent input into class discussions

· material for case studies and term projects

· increased professional relationships with industry and government

· increased employment opportunities for other students

· potential sources of contracts and grants

Of course such students demand adequate equipment and secure Internet connectivity. Without these resources, such students will both drop out and also provide negative information about the quality of the graduate program.

The department is in an enviable place in this regard, due to laboratory enhancement and renovation, and equipment purchases made to support accreditation of the undergraduate program. For the first time, actual research and office space are available to support the graduate program. All full-time faculty members now have offices instead of cubicles. The facilities are spartan, but will suffice until the department moves into the new Interdisciplinary Science and Engineering Building.

Assuming that a regular commitment of much-needed faculty positions and funds for maintaining computer hardware and software is forthcoming from the University, the Department of Systems and Computer Science expects to do the following by the Fall of 2004:

· increase the number of full-time graduate students in the Master’s program by 30 percent

· increase the number of part-time graduate students in the Master’s program by 30 percent

· enroll a total of 15 full-time graduate students in the Ph.D. program

The department intends to schedule its courses to allow sufficient offerings in the evening and late afternoon to allow part-time students to attend classes, while balancing this with the needs of the full-time students and the faculty’s essential need to do research using student research assistants. The proposed schedule of course offerings has been developed with the following goals in mind:

1. The listed courses increase the number of offerings sufficiently to be able to offer enough credits for the Ph.D. The number offered each semester is consistent with the number offered by the departments of Electrical Engineering and Mechanical Engineering.

2. The suggested modifications allow the program to be flexible enough to accommodate growth of the faculty if research directions change.

3. All courses are consistent with the level needed for a Ph.D. degree with the exception of the Analysis of Algorithms course. As a result, the department will not allow any doctoral graduate students to major in theory.

4. The changes can be implemented easily in such a way as to allow evening students to take MS-level courses, while allowing the program to be primarily a day program for full-time students.

In the academic year 2002-03, the Department has approximately 220 undergraduate and 43 graduate student majors. The faculty consists of nine full-time members, four part-time adjunct professors, a departmental administrative assistant, a departmental secretary, and two full-time network/software engineers to support five laboratories in two buildings.

Faculty

Ronald J. Leach, Ph.D., Chairman and Professor*

Legand Burge, Ph.D., Assistant Professor*

Don M. Coleman, Ph.D., Professor Emeritus (WOC)

William D. Craven, M.S., Assistant Professor Emeritus (WOC)

Moses Garuba, Ph.D., Assistant Professor*

Kathleen Kaplan, D.Sc., Assistant Professor

Peter A. Keiller, D.Sc., Associate Professor*

Harry Keeling, Ph.D., Associate Professor*

Wayne Patterson, Ph.D., Professor (part-time)*

Jeri R. Hanly, M.S., Lecturer

Todd Shurn, Ph.D., Associate Professor*

John A. Trimble, Ph.D., Associate Professor*

* = Member of Graduate Faculty

On leave

Arthur S. Paul, Ph.D., Professor (Interim Associate Provost for Research)

Adjunct

Caroline Wardle, Professor (WOC, part-time)*

Moyette Clark, Ph.D. (part-time)

Arif Khan, Ph.D. (part-time)

Bernard Woolfolk, M.S. (part-time)

Staff

Sharon Lacy, Administrative Assistant

Patricia David, Administrative Secretary

Robert L. Gault, Computer Systems Manager

Derssi Mebratu, Network Engineer

Strengths of the Department

The Graduate Department derives its strength from a variety of sources including the quality of its students, the appropriate placement of its graduates, the experience and dedication of its faculty, and the level of corporate support. The Department has always been able to attract a relatively large body of applicants whose academic backgrounds and other experiences suggest a better-than-average ability to cope with the rigors of the program. In the recent past, two factors have accounted for a considerable upgrade in the quality of the Department's graduate students. One is the strengthening of entrance and degree requirements, which has led the Department to be much more selective and thus, has resulted in admitting only the most capable applicants available. The second factor is the faculty.

The faculty has been very productive in recent years. The total number of publications of the department greatly exceeds 100. Conference papers have received “best paper” awards and several faculty members have been asked to give invited addresses or papers at conferences. Department members have published several books with two more in progress.

Most students are appropriately placed upon graduation and command very competitive salaries. Of equal, and perhaps even greater significance is our contribution to Historically Black Colleges and Universities (HBCUs) where, within the last six years, seven of our graduates have joined their Computer Science faculties despite a wide disparity between industrial and academic salaries.

With respect to faculty, recent external evaluation coincided with the Department's perceptions. In its July 1988 Final Statement of Findings, the Computing Sciences Accreditation Board (CSAB) concluded that

"…the faculty are dedicated and hard-working. They have sufficient breadth of exposure and experience in Computer Science to enable them to teach a broad range of courses and to improve the curriculum and maintain currency. Faculty seem to feel that they have a responsibility to provide the opportunity to learn for each student. Students believe that they are receiving a good education."

While these observations pertain to the undergraduate program, it is also true that the undergraduate faculty are essentially the same faculty who provide the graduate education. In the 1999-2000 academic year, the Department's graduate program was strongly recommended for continuation by the Administrative Review Panel for the Graduate School of Arts and Sciences, and the Department was encouraged to pursue the offering of the Ph.D. degree.

As implied in the discussion on student quality above, the program has been cultivating productive relationships with industry and much of the Department's equipment has been acquired as a direct result of such interaction. This interaction is also generating small but increasing amounts of financial assistance for support of graduate students and/or faculty development activities.

The department has been the beneficiary of several new construction projects:

· A new departmental PC laboratory for basic undergraduate instruction with 32 new Dell computers.

· Upgrading of the existing departmental PC laboratory by replacement of the 14 two-, three-, and four-year old computers.

· A new graduate office facility, offering both visual and physical separation of different research projects.

· A new laboratory to house Sun workstations for use in research and in teaching both graduate and advanced undergraduate courses.