This is a summary of a poster session presented at the 1993 CAUSE Annual Conference held in San Diego, California, December 7-10. It is the intellectual property of the author(s) and may not be published without permission. Disseminating the material otherwise is permitted, provided that the author(s) and the original presentation of the paper at CAUSE93 are acknowledged. A SUCCESS STORY OF IN-HOUSE DEVELOPMENT OF A STATE-OF-THE- ART STUDENT INFORMATION SYSTEM AT THE UNIVERSITY OF MARYLAND/COLLEGE PARK INTRODUCTION Buy or develop? That's the question. Many directors and administrators of University DP shops confront this question under dwindling resources and shrinking budgets. Five years ago, the University of Maryland at College Park decided to develop a in-house Student Information System. It comprised 1,500 programs that were housed in the UNISYS 1191 and Hewlett Packard 3000/64 to a new IBM 3090 mainframe. The Student Information System (SIS) serves seven major user areas including undergraduate and graduate application, registration and grade processing, advising, degree clearance and transcripts, and course management and scheduling. Rather than simply converting all existing programs from one system to another, the campus used this opportunity to re-engineer and improve the way some offices did business with the student and academic community, as well as the way the system interacted with its users. The office of Academic Data Systems (ADS) was charged with the development of the new SIS. A task force consisting of members from ADS and the user community was formed to map the goals of the new system, create a migration methodology and establish resource requirements. DEVELOPMENT REQUIREMENT During development, the campus should not be disrupted in its educational mission and users should continue to function with no loss of service. While we were migrating, current systems and the developing systems needed to be synchronized. The migration was not a straight conversion of the systems 'as is', but provided increased functionality whenever possible. Any additional function that were not possible during the basic migration occurred afterwards. The prioritization of how the various SIS systems were migrated was based on user needs. Additionally, the 15-year old Student Records system residing on the UNISYS was given priority because of the age of the system and a campus mandate. It was estimated that the basic migration would occur over a three year period. Long before the entire system completed, it was necessary for all SIS users to begin reaping the benefits of the new system. As the first step in the migration, the decision was made to populate and maintain a full, up-to-date SIS database on the IBM. OLD SIS PROBLEM There were problems. System access was limited only to first- line users, such as admissions, records, registrations, etc. Advisors and administrators at colleges and departments had no direct access to the information other than via paper. The overall look and feel of all the online programs was very inconsistent. This also caused problems with training. It was often painful for a user to move from one online program to another, and search-type data such as student ID numbers had to be re-entered each time. Control card parameters for nightly batch programs were entered by production control staff at the computer center via paper forms that were submitted by the user. Errors in the entry process often led to erroneous parameters being supplied, resulting in aborted jobs. Ad-hoc requests for information always required that a program be written; hence these requests require a programmer's time. Because security to access individual online programs was hardcoded, it also required programming effort whenever a user wished to limit or expand program access. Certain services to students, such as enrollment verifications and requests for official transcripts, though semi-computerized, were provided in a nightly batch mode and required much manual sorting and checking. SIS DEVELOPMENT GOALS The ADS staff met with the various users to determine what new capabilities and functions they envisioned for their respective systems. The programming staff, many of whom had worked very closely with the users for years, provided much input to this process. As a result of this interaction with the user community, the following goals and objectives were incorporated into the re-engineering process: 1)Information should be made available to colleges and departments either via direct access on the mainframe where appropriate, or via regular data download to smaller, individual platforms. 2)All online programs and reports should have a common look and feel across sub-systems, and navigation between online programs should be simple and should minimize re-entry of key information. 3)The security mechanism that allows entry to various online programs and sub- systems should be dynamically controlled by designated managers in user offices, thus eliminating the need for a programmer's involvement. 4)Users should be able to process a high percentage of ad- hoc requests themselves without the need for programming support. 5)Control card parameters for nightly batch jobs should be entered directly by the user, rather than by a third party, with the benefit of immediate editing. 6)To improve service to its clientele, user offices should be able to provide on-the-spot service whenever possible rather than overnight. NEW SIS HIGHLIGHT As a result of all this effort, the student services offices across campus have enhanced the way they do business, and in turn the quality of service they provide to students and faculty. Below are some concrete examples: 1) A fully integrated online navigation system was developed that allows users to go from screen to screen, program to program, and sub-system to sub-system with minimal effort. The system 'remembers' key information between transition to minimize re-entry of data. Student information is available by ID number or name. All function keys are standardized. 2) Online help is available and text can be entered by the user. The help can be brief, or it can provide full documentation. 3) The system allows a manager to dynamically set security for his or her system. Within seconds, access to any particular program or sub-system can be provided or denied by the office manager. In view of existing budget constraints, this capability has been particularly useful when user offices share personnel at various times during peak periods of activity. 4) 'Super selection' programs are available that allow the user offices to satisfy ad-hoc requests for information in a query-by-example mode. 5) A walk-in information request system was implemented that allows students to pick up, or have mailed to the desired destination, official transcripts and enrollment verifications. Under the old system, this task sometimes required as much as two weeks to complete. BENEFITS OF IN-HOUSE SIS DEVELOPMENT The major reward from all of this work and effort for the programming staff in ADS has been the satisfaction of seeing improvements that are tangible and visible on a daily basis. The new system has empowered users by placing more information at their fingertips, and by allowing them to take control with the service aspect of their operation. At the University of Maryland, every office is charged with the responsibility of improving service to students and faculty. By enhancing the way that various campus units deal its' clientele, ADS feels that it has contributed to that mission. The benefits of the in-house development are enhanced application software, flexible application software maintenance, maximization of user input through frequent user testing, quick return of investment of programmer training, conversion of programmer's new technology into another system development, high programmer and user morale, and substantial cost saving. WHY THIS EFFORT SUCCEEDED Several distinctive factors contributed to making ADS developmental efforts a success. The first was timing. The approach of cost-effective in-house development gained support from campus administrators during the budget constraint of 1991. It was necessary to develop, in house, a system that should accomodate the University's policy changes. If we had purchased a package, we would have had to pay exorbitant fees whenever policy changes were made. When creating SIS, the developmental goals had to satisfy the University's mission to provide flexible computing services to all levels of users. Second, frequent communication between users and programmers was a vital element to our success. No developmental effort can succeed without direct user involvement. Third, ADS was lucky to have able, dedicated programmers who were willing to learn and grow. From the Director to student programmers, one objective was to program efficently. Finally, state-of-the-art computer technology has provided tools that fuifill developmental goals. Jacob Lee, Systems Analyst, Email-jlee@deans.umd.edu Barbara Riggs, Assistant Director Records and Registrations Email-briggs@record1.umd.edu Date Revised: 02/24/94 (bdz)