3.2.1. Identify computing advocates/advisors in each college and administrative unit and coordinate their activities regarding reporting and dissemination of information.

At present, dissemination of information on computing, training opportunities, and applications use is uncoordinated, with many users receiving little or none of the information, which they should receive. This problem is particularly severe in many of the smaller units. Only through establishing a more formal process of dissemination of information can the problem be overcome. Nearly every college and administrative unit has a local computer advocate/advisor. Such an individual may have very specific job responsibilities related to computer support or may simply be an experienced user. These individuals need to be identified and used as conduits for channeling computer and network information and support into the unit. The Task Force favors requiring every unit head to assign someone the task of acting as unit advocate/advisor. This way there is some local recognition of who is responsible as well as official recognition of the time and effort the individual spends in supporting this task. 

3.2.2. Expand training and one-on-one user support for administrative staff.

Administrative staff consistently mentions the need for access to more training and support for computing. Staff said that they were not properly trained in the use of administrative applications programs and that they did not have a clear understanding of the scope of capabilities of the computing and networking technology with which they were working. Many expressed the desire to continually increase their knowledge and skill in the use of computers. Most saw increased knowledge of computing as the key to becoming more effective and efficient in their current jobs and as something which would help them get better jobs. In some cases, it was felt that when a problem occurs in using an application, access to experts would greatly increase productivity. 

This recommendation addresses the need to raise the level of computer use training and support for administrative staff as one way of increasing the productivity and efficiency of University administration in general. Because turnover and movement within the ranks of administrative staff is high, training is a constant and large task, which must be managed if it is to be effective. 

The Task Force recommends that: 

• Staff training be a prerequisite to adding any new database and/or application to the administrative information system. 
  
• The Computing Support Services Center be responsible for scheduling and coordinating all administrative applications systems training and general computer skills courses. The Center should involve professional educators, knowledgeable peers, and applications developers in the training program. 
  
• A team of experts be available for on-call and on-site problem solving and , support for administrative applications and personal computer use. 

3.2.3. Improve and expand training facilities. 

As a corollary to recommendation 3.2.2, effective training will require improved and expanded training facilities. Because administrative staff are direct and intensive users of computer equipment, training needs to be hands- on and in small groups to be most effective. Currently no training facilities are dedicated to training and developing computing skills of staff. By placing training as a major responsibility of the Computing Support Services Center, the Task Force expects to see more intensive use of available training facilities and consideration of administrative staff needs when new facilities for training are developed. New training facilities should address the development needs of staff as well as faculty and students. 

3.2.4. Develop a contract-based system for acquiring NERDC mainframe computing and file resources for administrative production and development systems.

The current method of charging for computing services at the NERDC is perceived by users and developers as being a major stumbling block to full use of current administrative information systems, as well as to innovation and change. The combination of a transaction-based charging algorithm and increased demands for information management has left University data managers and administrators unable to predict annual mainframe computing charges. At the same time, NERDC management is struggling to keep charges in line with its approved budgets while simultaneously trying to satisfy a very diverse user community. The result is periodic manipulation of the charging algorithm which provides budget relief to one user/provider while doing little to help another user/provider. Add to this the need for the University to begin a major administrative system overhaul (see recommendation 3.2.6), and it becomes clear that the current system must change. 

The Task Force recommends that the University move to a contract-based system for paying for NERDC mainframe services and support. Such a system would have the dual benefit of guaranteeing NERDC income for the year while at the same time allowing data managers and developers at the University to budget and plan. It is recognized that NERDC, as an SUS auxiliary, has responsibilities which lie outside the University of Florida. It is felt that these outside users, most of which are part of the SUS, DOE, or another state agency, would also benefit from a contract-based charging system. 

The NERDC Policy Board must approve all major changes to the NERDC operating environment, and the proposed contract-based system is certainly a major change. Responsibility for developing the proposed contract-based charging system shall lie with the NERDC director, the University Comptroller, the University Auditor, and the Provost as Chair of the NERDC Policy Board. The policy should be fully defined and implemented in the 1992-93 fiscal year. 

3.2.5. Assign responsibilities, set deadlines, and fund systems development to move all major administrative information into a centralized database management system.

Users expressed major concern about the general lack of access to information on administrative and academic processes and programs. Academic and administrative staff cannot examine alternate views of existing databases or easily download data/information to a personal workstation for further examination and analysis. Duplication of data elements across a large number of separate files exists, increasing the difficulty of managing the data as well as increasing storage requirements, and hence storage costs. All of these actions are increasing the chances of misinterpretation and improper use of the information in the University's data files. 

The Task Force recommends that the University begin immediately to move all major administrative information into a centralized database management system (CDBMS). This system should allow sharing of common data elements across applications, improve access to data, and improve the applications development environment. The establishment of a CDBMS would require the designation of one individual as Database Manager who would be responsible for the coordination of data element naming and placement, maintenance of required standards of access to and security of the database, and maintenance of the directory of information available in the database. The Database Manager should not exercise policy control over the development of applications which access the CDBMS, nor should the Database Manager have policy control over who has access to the information in the system. The providers of information into the CDBMS should maintain responsibility for security and access to the data elements provided. The structure of the CDBMS should be planned to accommodate both data provider and user needs. 

Because movement to a centralized database management system represents a radical change in direction for administrative computing at the University, it needs careful planning and measured implementation. While the whole process cannot be done in a month or even in a year, the commitment to move in this direction must be established now. The current method of paying for administrative computing is incapable of covering the cost of making this transition (see recommendation 3.2.4). During the transition, additional technical personnel will be needed to supplement current staff who are consumed with the day-to-day activities of their respective administrative units. Along with additional personnel, funds will be needed to purchase state-of-the-art database and applications development tools. 

3.2.6. Establish a uniform security system for administrative and academic information access that is manageable and maintainable.

The current system for controlling access to administrative and academic on-line information works but has been described as a crisis waiting to happen. Because access to applications is currently controlled via passwords assigned by database developers, users are often faced with having to remember which of their multiple passwords will allow access to which information. It is a time- consuming task to obtain approval and a password for a new application or item on the "Deans' Menu." Only a few individuals can establish or replace passwords. The state accounting system, SAMAS, requires a completely separate authorization code for access. The mechanisms for "deactivating" access privileges for employees who transfer to a different job within the University or leave the University entirely are not automatic, and often result in accounts being active long after the employee has moved on. 

The Task Force recommends that the administration of access control and security management be consolidated and that information and application security be a necessary and planned component for all future development. Once an agreed upon solution is reached and specifications developed, a uniform security system can be implemented within a year. A number of systems have been proposed but since no one system is unanimously supported, nothing has been done. It is recommended that the University make the administration of access control and security the responsibility of the Computing Support Services Center, and that a group of technical staff be appointed to define and develop a plan for an access control system for review and approval by appropriate information systems and University administrators. The cost of implementing the system should be covered by the University, and all applications developers and administrative users will be required to use this system. It should be noted that implementation of a uniform security system will require that current applications be rewritten to access the system. One characteristic of this access control system should be that each user has one access code to remember which provides access to all necessary information. 

The developers of the CDBMS will have to define data element security along with the developers of the individual applications which use the databases. These developers should not be involved in the day-to-day maintenance of the security access system, since maintenance activities such as password control, and providing access mechanisms, are service-oriented and not database management oriented. 

3.2.7. Establish a multi-platform query system as the standard for access to administrative information.

It is imperative that the CDBMS selected include a query system that is capable of running on most of the personal computers and local area network systems that exist on this campus. A number of mainframe database management systems provide "query system front ends" that allow for access from within a number of software packages running on personal computers or network servers. With appropriate access authorization, a non-programmer may use these query systems to access individual data elements and records in the CDBMS and download the information to a personal computer for further analysis and graphical display. Because of the flexibility in this approach, existing data can be used more effectively and with less dependence on the database provider, a programmer, or mainframe computing resources. 

3.2.8. Establish a common user view for administrative applications.

It is important that the University establish a common interface for applications for individuals who do not wish direct access to the database, but require a more standard view of the data for daily administrative tasks. This common interface must provide the same view regardless of the type of computer system being used to access the system. It is strongly recommended that the interface be graphical in nature. A graphical interface requires a higher level of intelligence than is currently found in the computer terminals and workstations used for administrative work. 

Before a common user interface can be implemented, the University will need to "retire" hundreds of non-intelligent terminals and replace all administrative workstations with personal computers or "processing terminals." The capital costs are high for a project of this magnitude, but, depending on how the CDBMS and the common user interface are implemented, the processing power of the personal computer could be used in such a way as to share some of the computing load with the mainframe computer, hence delaying the need to upgrade the mainframe. 

3.2.9. Establish administrative strategies to handle major new applications development.

The centralized database management system project and associated upgrading of administrative workstations are very important to the future of the University. A number of other major projects were not examined in detail by the Task Force but were identified as having the potential to beneficially change the University. These include a telephone registration system, an optical disk-based document management system, replacement of older rotary telephone handsets and key systems, a campus-wide video system, and the University ID card system. Each of these projects must be undertaken if the University is to improve its administrative and academic services. In each case, failure of the University to move forward on these projects will result in lost productivity or competitiveness. Other, more innovative uses for information technology resources, such as a network of information kiosks to provide information to students and visitors, have not even been discussed at the University of Florida because of the perceived and actual limitations of resources for projects of this size and complexity. 

There currently exists no easy way for these projects to be evaluated, prioritized, and eventually funded. All of these projects involve basic activities or tasks which are currently being performed without the benefits of current technology. They each involve major investments in money and personnel as well as major changes to many related activities and tasks. 

While the support and coordination structure recommended in this report can handle the operational details of each project, administrative strategies for mandating and funding this level of project must be developed. Discussions of funding major information technology projects and setting priorities for these projects within the context of all of the other types of major projects within the University are outside the scope of this report.