The Decentralization of Academic Computing: Defining Future Roles Copyright 1990 CAUSE From _CAUSE/EFFECT_ Volume 13, Number 3, Fall 1990. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, the CAUSE copyright and its dateappear, and notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To disseminate otherwise, or to republish, requires written permission. For further information, contact CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301, 303-449-4430, e-mail info@CAUSE.colorado.edu THE DECENTRALIZATION OF ACADEMIC COMPUTING: DEFINING FUTURE ROLES by William J. Kettinger ************************************************************************ William J. Kettinger is Assistant Dean for Information and Technology Resources at the College of Business Administration of the University of South Carolina. He is responsible for managing the College's James C. Self Computer Center with an IBM mainframe, the Springs Business Library, the Center for Academic Computing, and the Sony Video and Graphics Laboratory. ************************************************************************ ABSTRACT: This article describes the trend toward decentralization of academic computing in higher education. It explores the causes for this trend and discusses how best to manage and support computing activities in this environment. Specifically, future roles of the central computer center, departments or discipline areas, and individuals are outlined. Until recently, centralized computer organizations were in the best position to meet academic computing needs. However, the personal computer and demands for discipline-specific computer support have irrevocably rocked the nature of academic computing. On many campuses, centralized computing's locus of authority has eroded as academic computing resources are increasingly dispersed to departments and individuals. While this trend might be most pronounced at larger research-oriented universities, its traits are manifesting themselves at even the smallest schools. Higher education institutions must begin to ask what is the best way to manage and support their academic computing activities. Clearly, old paradigms will not work. Decentralization of computing demands greater campus-wide planning to ensure effective utilization of resources, but also requires less central administration. Where does this leave the central campus computer center? What role do departments or discipline areas have? What are the responsibilities of the individuals? DECENTRALIZATION OF ACADEMIC COMPUTING In the 1960s, computers were large, expensive, complicated equipment that required the constant attention of a highly trained technical staff. Colleges and universities segregated these machines, making them available to faculty only on a limited basis. In the 1970s, interactive time-sharing fueled the expanded use of computing technology for academic purposes, moving computing out of the computer center and distributing it around campus in the form of remote terminal labs. During the late 1970s to mid 1980s, "distributed computing" encouraged the use of minicomputers at remote locations under the supervision of the central computer operation (see Figure 1). Many larger academic departments purchased their own minicomputers and established miniature operational versions of the central mainframe. Distributed computing usually provided improved technical performance for those processing tasks performed routinely at the remote site. However, it often only served to migrate existing central mainframe computer users to the distributed machine and did not greatly expand the number of academic computer users. [FIGURE NOT AVAILABLE IN ASCII TEXT VERSION] During this same period, a much more profound revolution was taking place. With the introduction of the personal computer came a significant change to the college campus and the long-term shift of computer resources to the individual.[1] Faculty and students began to integrate computer word processing, spreadsheets, and databases into their classes. Where mainframe-based interactive computing allowed academic computing to decentralize to remote labs, personal computers liberated it from the labs and moved it to the classroom, offices, dorms, and homes[2] With advances in networking, people now expect convenient computer connections that provide them access to processing and data resources locally, nationally, and internationally. Multiple access points are being required in libraries, research laboratories, classrooms, faculty offices and homes, dormitories, administrative offices, other institutions, and the research community world-wide. Users expect to have a powerful workstation at their disposal that can provide any level of functionality needed with little or no concern for how this is accomplished technically. However, these expectations should not be viewed as just another version of distributed computing. With decentralization, users believe that they are at the locus of control, that they have at their disposal a myriad of computing options, and that they can be judicious consumers. Other technological advances promise to continue to change the face of higher education. These new technologies include high-speed local and satellite communications, video disk and other large capacity information-storage devices, graphics, two-way cable systems, artificial intelligence applications, and high performance computing.[3] In the future, as prices drop and processing power continues to rise, students and faculty will obtain their own portable personal computer which will be used as a combination typewriter, notebook, calculator, and textbook.[4] One unique feature of these technologies is that they will continue the trend to decentralize computing power. Migration of Academic User Support Centralized academic user support blossomed during the 1970s. It was during this period that the introduction of sophisticated data analysis software (SAS, SPSS, MINITAB, etc.) stimulated computer use by large numbers of inexperienced faculty and graduate students. These new non-technical users had little programming skill and were not interested in learning the details of computer operations.[5] To support these individuals, computer centers had to set up user support organizations (see Figure 2). These centralized academic user support groups grew with consultants who were technically knowledgeable about specific software, but generally not knowledgeable or interested in the subject matter or research of individual faculty and students. [FIGURE NOT AVAILABLE IN ASCII TEXT VERSION] During the early 1980s, discontent began to surface concerning central academic support personnel's general lack of discipline knowledge. Faculty often complained that the user support people really didn't understand what they wanted to do. In an effort to rectify the situation, some central computing organizations established academic information centers modelled after those being used for administrative computing.[6] However, due to a general lack of funding, often unqualified personnel, and the usual lack of proximity between the information center and discipline area, the centralized information center approach did not experience the same level of success in academic computing that was experienced for administrative user support. Rather, many discipline areas decided to set up their own autonomous departmental equivalent to an information center. Decentralized departmental support centers are becoming an increasingly common feature of the academic landscape. These organizations range in size and scope from one faculty member who is designated the departmental computer "guru" to large support organizations such as those at many of the major professional schools. In most universities, the development of decentralized support centers is less planned than evolutionary, and even on the same campus the dimensions of their operation can vary widely among colleges and departments. Typically, these decentralized support organizations do not have a direct reporting relationship to the central computer operation, but rather report to the dean or department head within which the academic computing support is being performed. These organizations can take many forms, but typically provide some variety of the following essential services for faculty, staff, and students: * Assist departments and faculty in determining appropriate hardware and software for their specific discipline. * Install and maintain hardware and software within the department. * Train faculty, staff, and students by providing short courses, user manuals, and demonstrations, as well as by sometimes teaching for- credit courses. * Provide a technical liaison function with vendors concerning products used for research and instruction. * Provide networking assistance on local area networks and in interfacing to campus-wide or external computers and online data services. * Develop and support administrative applications within the department and represent the department concerning campus-wide computing matters. * Offer discipline-specific consulting support such as computer graphing, econometric modeling, interfacing to scientific equipment, and CAD/CAM applications. * Manage the disciplinary research databases such as the Center for Research in Stock Prices (CRSP) and CompuStat for business, and the Inter-university Consortium for Political and Social Research (ICPSR) data sets for the social sciences. The need for this type of discipline-specific computer support will continue. This is particularly true because of the explosion in the number of personal computer applications that meet special needs. While many general purpose PC software packages are still widely used (Lotus 1-2-3, dBase III, WordPerfect), increasingly faculty and students are using specialty software that meets the unique needs of a particular discipline area (financial software and data in business, design tools in engineering, visualization software in the sciences). As computer applications become more specialized, the value of centralized consulting support is even more diminished. This is because the sheer volume of programs each consultant must support precludes extensive subject matter knowledge. As computing matures and is adopted into practice (the "real world"), we can expect the number and complexity of these specialty applications to continue to increase. For this reason many argue that approaches that ignore departmental involvement and specific disciplinary applications will flounder at best and may be subverted into an anti-computing backlash.[7] Because of the specialized nature of academic computing today, it is nearly impossible to fulfill faculty needs adequately by reliance on a single central academic support staff, except in very small college environments. Problems with Decentralized Academic Computing The move to decentralized academic computing is not without its drawbacks. Clearly, the old organizational problems associated with centralized versus decentralized organizations hold true for academic computing. Muffo states that "lack of planning and control at the college (decentralized) level has resulted in confusion, frustration, misunderstandings, and inefficiencies in a number of cases."[8] While many of these problems existed in the past, the advent of decentralization has added a new level of complexity. Specifically, new problems have arisen related to mission and planning differences, management issues, resource allocation, personnel issues, data management, and compatibility. Mission and planning differences Decentralization can result in parochial views with little concern being given to the campus-wide computer infrastructure. While the myth of the "community of scholars" may still exist, in truth, colleges and departments are often competing for scarce university resources and typically have little interest in supporting computing projects that do not directly benefit themselves. College or departmental computer centers are typically organized around the individual research and instructional goals of that sub-unit. Due to the specialization of the various disciplines, the computing tools to accomplish these goals can vary widely. If left to operate in an isolated mode, these decentralized computer centers may quickly disregard the needs of the campus as a whole. Because the summation of their decentralized goals may not equal the institution-wide computing mission, campus-wide infrastructure could soon decay. For example, expenditures on a high-speed fiber optic back- bone network to support a centralized supercomputer may meet the planning requirements of the university, but not those of the college of journalism, which only wants to create a computer-simulated newsroom using personal computers. Clearly, a planning mechanism must be in place that focuses on attempts to address all the institution's computing needs and deals with the entire campus in an equitable manner. Management issues Management and the locus of control for decentralized computing resources continue to be bones of contention. Since most departmental computing operations report to a college dean or academic department head, decentralized staff may feel no sense of connection to the central computing organization. Often, ownership of and obligation for computing resources are viewed as locally based. At many universities, campus-wide mechanisms to effectively coordinate computer and telecommunication resources in a decentralized environment do not exist. This can result in many problems, such as: * duplication -- because of lack of sharing, expensive and often underutilized computer resources have to be replicated on campus; * inventory control -- it may be impossible to identify what computing resources exist on campus and how they are being used; * maintenance -- expensive external maintenance contracts may be used as opposed to centralized on-site maintenance; * ownership of intra-building networks -- isolated decisions concerning networking approaches and use of intra-building wiring may lead to expedient short-term but disastrous long-term decisions; * space -- the need for local computer laboratories and operation centers has placed new pressures on space for faculty offices and classrooms; and * procurement -- the repetitiveness and volume efficiencies of centralized purchasing are lost with decentralized purchases. Resource allocation When responsibilities are dispersed, the battle for money and people begins. With greater scope and complexity of decentralized computer operations comes the need for more support personnel and associated operating budgets for new equipment and maintenance support. In times of tight money, the source for these often comes from either existing computing funds or student computing fees. Reallocation of these funds has sometimes made for professional jealousies and internal squabbling both among departments and between departments and central computing. Personnel issues In the personnel area there is the problem of establishing equitable pay scales and career paths for both departmental and central computing personnel. Decentralized computing staff may feel isolated from their computer colleagues and their upward mobility may be stifled in smaller departmental or college support organizations. Heavy service demands placed on these staff, as well as smaller operating budgets, may result in insufficient opportunities for career development activities. Finally, because they frequently report to a faculty member, decentralized support staff may not have a supervisor who identifies with and understands the demands of their profession. By the same token, centralized staff may view decentralized personnel as being better paid, receiving more exposure, and generally being given an opportunity to be more productive than they are in their large, central computing organization. Data management An important topic that faces departments that have undergone decentralization is the issue of data ownership, control, and update. In the case of academic computing, data tend to be the individual programs and data sets of faculty and students and those large proprietary databases used in research. When all processing took place in a batch mode on the computer center's mainframe, there was little question that it was the computer center's responsibility to act as the steward of these data. However, with decentralization, these data sets are now being run on departmental and individual machines, users are now responsible for their own back-up, security, and updates. In addition, as these data sets are moved to departments and individuals, they may not be generally available to users elsewhere on campus. Compatibility An overall technical and data architecture is important for communication both on and off campus. However, with decentralization this is becoming more difficult to accomplish. In the past, the central computer organization set these technical standards and they generally were held campus-wide. That situation may not exist any longer. Since decentralization has given individuals greater freedom to purchase what they want, the diversity of technical products on campus has multiplied as has the likelihood for incompatibility. Specific discipline areas may have technical preferences that may not easily fit into previous campus- wide technical standards. An example of this has been the widespread adoption of UNIX as a workstation operating system among the scientific and engineering community. Increasingly, the technical needs of the institution are being defined from the bottom up and the importance of departments and individual faculty in standards development is becoming crucial. Yet many campuses that have experienced decentralization have not adjusted to this change. As a consequence they have many incompatible computers, networks, and software packages. Advantages of Decentralized Academic Computing Lower cost and increased capacity have been the driving forces pushing the technological reasons for decentralization of computing. However, technological change is not the only justification for this trend. Decentralized computing offers specific advantages. Responsiveness to academic mission The heart of any institution of higher education is instruction and research. Traditionally, these responsibilities have been the purview of the college faculty. However, due to the efficiencies offered by centralization, the computing resource has been separated from the faculty. While some faculty have made use of computing for research, and to a much more limited extent in teaching, centralized academic computing support has played only a relatively minor role in meeting the academic mission. Because centralized computing has tended to concentrate on solutions that focus on economies of scale and minimize specialization, it has tended to meet only the needs of the lowest common denominator of potential computer users. In contrast, decentralized approaches can be targeted to the specific objectives of disciplines or individuals. Decentralized support staffs, which are also knowledgeable in the discipline, are now assisting in the development of research and instructional applications that were previously unavailable through centralized academic support. For example, departments are developing a wide range of research applications from econometric and fluid dynamics modeling to oceanographic and geo-information systems. Because the computing resource is closer to the people who need it, it is used more, can be customized more easily, and can be disseminated more quickly. Increased user productivity A combination of personal computing, greater computing options through diverse computer networks, specialized departmental applications, and local user support has increased user productivity. Individuals now have the choice as to the most appropriate level for their computing (PCs, departmental computing, campus-wide computing, and external network computing such as NSFNET). The technology is now powerful and widespread enough to allow individual faculty to conduct research and offer instruction that was not previously feasible. For example, high-end workstations are permitting numerically intensive research applications on desktops that were only possible a few years ago on distance supercomputers, and several business schools are now using simulation software and actual business data to analyze "real- life" cases during class time. Greater computer integration into the curriculum While widespread integration of computing into the curriculum may not be universal among all schools and disciplines,[9] the recent landslide of PC software, PC classrooms, and student and faculty discount purchase programs is beginning to have a tremendous effect on the college curriculum. Clearly, business schools have to be viewed as a leader in this regard. A recent survey of AACSB business schools suggests that computer usage is required in 72 percent of undergraduate core business courses and 66.5 percent of graduate core courses. In addition, 90 percent of the business schools surveyed have classrooms that are equipped to display interactive computer output, either by terminal or microcomputer.[10] Students still are required to complete computer assignments away from class, but they are also beginning to be responsible for completing projects during class time in specially designed computer classrooms. With improvements in computer-assisted instruction and distance learning, we are only beginning to experience the impact that decentralized computing will bring to instruction. As Staman points out, the best way "to integrate technology into the teaching/learning process must ultimately evolve into the question of how best to create an environment in which interested faculty can, if they choose to, create changes in individual courses, one course at a time[11] Knowledge of discipline It has already been argued that the proliferation of discipline specific software and unique computer equipment precludes centralized academic support for all but the most general software packages. In the past, lack of knowledge of the discipline by computer center support personnel has frustrated busy research faculty looking for new ways to apply the computer to their problems. Computer support organizations that are locally based within academic departments and staffed with discipline-knowledgeable personnel offer faculty the targeted technical assistance they need to use the computer to increase their research productivity. Demonstration of support for faculty Many academic areas are experiencing tremendous competition in the recruitment and retention of quality faculty. This is particularly true in the professional schools and sciences where industry attracts many of the qualified Ph.Ds. With recent budget cuts, many departments are finding it difficult to even maintain past levels of faculty positions. Increasingly, colleges and departments are turning to adding lower-cost support personnel to increase the research and instructional productivity of their remaining faculty. Adding departmental computer support personnel is one way that departments can demonstrate a high level of faculty support as a means to both attract and retain quality faculty. Catalyst for innovation on and off campus While there are certainly exceptions to the rule, centralized computer centers tend not to be the most innovative organizations. Almost by their very nature they have had to encourage the status quo for the remainder of campus, while focusing on their own growth.[12] Decentralized computing offers a refreshing change that emphasizes diversity of application. The reward structure for departmental computer centers seems to focus on encouraging new ideas and creative approaches that bring recognition from corresponding professional associations and sister schools. This acts as a powerful incentive to innovate and is similar to the process whereby faculty must publish new research. Once a decentralized application is recognized as a good one, it can act as a catalyst for the diffusion of ideas both inside and outside the institution. Ultimately, this process should bring recognition to the college or university as a whole. NEW ROLES IN HIGHER EDUCATION COMPUTING Decentralization of academic computing is dictating a rethinking of how computing is organized on campus and redefining the roles of those involved. Figure 3 outlines technical trends in academic computing. [FIGURE NOT AVAILABLE IN ASCII TEXT VERSION] New Role of the Central Computer Center The central computer center will increasingly have to function in a competitive environment, with individual departments being self- sufficient for the bulk of their academic computing needs and only looking to procure special services from the central computer center. Typically, the computer center will be responsible for operating mainframe computers for campus administration and providing mainframe processing for those academic departments that do not possess their own departmental mainframes or minicomputers. The need for this latter function will diminish over time as the processing capabilities of high- end workstations make them superior machines to run quantitative analysis. Ultimately, all academic computing will be relegated to departments or individuals with the exception of technology that proves to be only cost-effective on a centralized basis, such as supercomputing and networking to external data networks and services (NSFNET, Internet).[13] Since the computer center will function in a competitive environment, it will become more market-driven. This should result in more innovative approaches and, ultimately, it should provide only those academic services demanded by the community. The central computer organization will still play a prominent, but changed, role in academic computing (see Figure 4). [FIGURE NOT AVAILABLE IN ASCII TEXT VERSION] The most important future role of the central computing organization probably will be in facilitating coordination by organizing policy-making activities and providing technical advice. As Alley, Shaub, and Willits point out, the success of decentralization will depend on information technology management's support of the notion of institution-wide coordination.[14] As the computer center loses much of its operational responsibilities, it will be increasingly asked to provide institutional guidance on an information architecture, networking, and hardware and software selection. This planning function should focus on how best to meet both the needs of individual users and the campus community as a whole. Paul Davenport suggests that "key academics need to be involved in IS priority committees and in IS planning generally, with as much access as they need to data on costs, projects, etc."[15] Several leading universities are now in the process of experimenting with various schemes to assist in this type of cooperative planning[16] New Role of Departmental Centers Departments or discipline areas will continue to be the focus of academic computing for some time into the future. This new role was originally the result of the distribution of minicomputers to selective departments, but has grown in intensity with the advent of the PC and the movement of user support to the departments. Specialization of both hardware and software will continue to require discipline knowledge to support increasingly complex instructional and research computer applications. As a result of this trend, we will see more and more new computer support personnel within departments. In most cases, these groups will report directly to the department head or college dean, while in other cases they will act as local representatives of the central computing organization. In either case, as Figure 5 displays, these departmental computer support centers will be responsible for specialized consulting concerning discipline-specific hardware and software, operating distributed administrative systems, running computer-based research libraries, developing presentation graphics, maintaining local area networks, and providing direct instructional assistance in the form of specialized training and curriculum development. In essence these organizations will become the principal contact for computer users with problems and the liaison between the department and the central administration concerning computer policy. New Role of the Individual Academic computer users typically perform information processing tasks in support of their three primary functions: instruction, research, and personal administrative duties. Many of these information processing tasks such as text processing, grading and reporting, graphics and presentation development, statistical analysis, simulations, and experimental applications, were once performed on the computer center's mainframe. Over the next ten years, these functions will increasingly be performed on personal workstations as will the bulk of all academic computing. The explosive growth of PCs and networks has given individuals tremendous control over their own computing futures. However, with this new freedom comes additional responsibility. Autonomous computer users must now take responsibility for developing their own applications as well as for maintenance, backup, and security of their data. In essence, local computer users must now perform on an individual basis many of the tasks the central computer center performed for them on the mainframe in years past. Clearly, the individual user has been the big winner in recent computer resource allocation decisions as a result of decentralization. One only has to look to the plethora of PCs in student labs and faculty offices to see where the money is going! Where at one time almost all important decisions concerning campus computing directions were established within the confines of the computer center, today, key computer users are being brought into the decision-making process. Finally, through computer user groups and departmental representatives on campus-wide computer policy committees, a new cooperative approach to academic computing is beginning to surface on college campuses. Much has been written projecting how computers and telecommunications will make profound changes to higher education. However, until recently, much less concern has been given to how colleges and universities should be positioning their internal policies to take advantage of these changes. We can expect that technological advances will continue to push the decentralization of computing. Clearly, computer organizations and policies that were effective in a 1970s computer environment are not appropriate for a 1990s world. Recently, many higher education institutions have recognized the need to rethink their approach to academic computing and have begun to evaluate their information technology organizations to determine the best ways to manage in an increasingly decentralized setting. The role of the central computing organization has changed! Those computer center managers who were comfortable solely as technicians, acting in isolation, may be very uneasy in the present computer environment. Increasingly, the new posture of the central computer organization requires moving from a reactive-operational role to a proactive-leadership stance. As individual users become more independent, the central computing organization must become information resource planners, facilitators, and technical consultants. This new role relies at least as much on good communication skills as it does on providing computer services. As the computer center's operational responsibilities for academic computing decrease, colleges and universities must face tough issues related to allocation of staff and budgets. These decisions will be influenced by the varying management approaches taken toward decentralization and the unique organizational culture at work on each campus. While no definitive model for dealing with decentralization has been devised, and it is unlikely that one will surface, it is certain that the new decentralized computing scene will demand greater policy-making involvement by campus computer users. ======================================================================== Footnotes 1 Robert G. Gillespie, "Evaluating Campus Computing Services: Taming the Technology," New Directions for Insti-tutional Research, Winter 1987, pp. 83-92. 2 Stanley Warren, "Aca-demic Computing: The Challenges Ahead," New Directions for Institutional Research, Summer 1987, pp. 29-41. 3 Francis D. Fisher, "Higher Education Circa 2005: More Higher Learning, But Less College," Change, January-February 1987, pp. 40-45. 4 "Educational Technology: Where do we stand?, A Report of the Task Force on the Use of Educational Technology, American Society for Engineering Education," Information Technology Quarterly, Fall 1987, pp. 4-21. 5 Robert R. Blackmun, Jeff N. Hunter, and Anne S. Parker, "Organizational Strategies for End-user Computing Support," CAUSE/EFFECT, Fall 1988, pp. 33-43. 6 John Bates and Gerry Leclerc, "A Microcomputer Information Centre: McGill's Experience," CAUSE/EFFECT, November 1985, pp. 24-27. This article depicts the thinking at the time concerning the use of information centers to support administrative units. 7 G. David Garson, Academic Microcomputing: A Resource Guide (Newberry Park, Calif.: Sage, 1987). 8 John Muffo identifies many of these problems in a study of decentralized computing in an individual college at a large public university. He breaks these problems down into: resource issues -- equipment, space, and personnel; academic issues -- instruction, research, and public service; and computing policy issues -- overall coordination, use of the institutional mainframe, and sharing of facilities. In his case study, Muffo recommends greater college-wide management and more campus-wide coordination, and expertise and leadership rather than control. See John A. Muffo, "Decentralized Computing at the Operational Level: Some Lessons Learned," CAUSE/EFFECT, Winter 1989, pp. 42-46. 9 Frank Newman, "Technology on Campus: An Uneven Marriage," CAUSE/EFFECT, Spring 1990, p.10. 10 Jason L. Frand and Julia A. Britt, Sixth Annual UCLA Survey of Business School Computer Usage (Los Angeles: John E. Anderson Graduate School of Management, UCLA, 1989), pp. 30-32. 11 E. Michael Staman, "An Action Plan for Infusing Technology into the Teaching/Learning Process," CAUSE/EFFECT, Summer 1990, p. 40. This paper offers a plan to encourage computer course integration based on a team approach involving a centralized academic computing support group and individual faculty. 12 Robert E. Zimmerman, "Survival Skills for the Computer Center in the University of the Future," in Proceedings of the 1988 CAUSE National Conference (Boulder, Colo.: CAUSE, 1989), pp. 101-108. 13 Alan Creutz, "The Role and Management of Telecommunications in Higher Education," New Directions for Institutional Research, Summer 1987, pp. 43-57. Also see John W. McCredie, "Planning for Diversity," Information Technology Quar-terly, Summer 1988, pp. 20-22. 14 Lee Alley, Michael Shaub, and Stephen Willits, "Institution-Wide Coordination of Decentralized Computing," CAUSE/EFFECT, March 1987, p. 7. 15 Paul Davenport, "Gen-erating Academic Support for Information Systems," CAUSE/EFFECT, Spring 1990, p. 33. 16 For example, since 1986, the University of Michigan has been using planning teams made up of departmental users and the computer center as a primary vehicle for coordinating services for faculty and staff. The basic philosophy behind this approach is that the great expansion in computing expertise on campus is taking place outside central computing in the colleges, schools, and operating units. See Cynthia S. Cross, Marianne J. Elser, and Jill B. Tuer, "The Departmental Planning Team: A Bridge to the Future," CAUSE/EFFECT, Winter 1989, pp. 19-24. ========================================================================