Speak PeopleSoft

Here are some phrases that often come up in PeopleSoft discussions:

client/server architecture	mainframe architecture
distributed workload	workflow
end-user	Year 2000 (Y2K)

Client/server architecture uses the computing power of individual desktop computers to respond to data requests. Software installed on a desktop machine (client) uses the network to request and process data from a central database (kept on a "server" computer). The client software can be configured to the needs of the individual user at that workstation; the server data is uniform to all users. A desktop program such as Netscape or Eudora is built on the client/server model.

The university's old systems use mainframe architecture, characterized by "dumb terminals" that with communicate with a central (mainframe) computer, which processes the request and returns the result to the requester. On a mainframe system, all of the data processing work is done by the mainframe computer.

Distributed workload (environment), in the technical sense, is one result of client/server architecture -- the computing work is distributed to desktop clients rather than being centralized on the mainframe. In a general sense, distributed workload creates a decentralized business structure that allows workers nearest the actual site of work to report, manage and evaluate its efficiency and effectiveness. Many of the university's current business processes are centralized: work information projected by a central office (for example, a unit payroll report) is sent (usually on paper) to individual units for confirmation, then returned to the central office for processing. Each exchange of information in a centralized environment may corrupt data. In a distributed environment, supervisors would report each employee's work hours directly to the database, reducing the number of transmission points and the possibilities for error that result. On the student side, the current processes for creating and distributing class registration lists and reporting grades are centralized. Although PeopleSoft permits such processes to be managed centrally, it also allows for them to be distributed when the university is ready to do so -- system development during the enhancement phase (2000-2001) may include the option for professors to request their class lists or report their grades directly to the database, either through PeopleSoft or through a Web application that would require no PeopleSoft client software at all.

An end-user runs a client program to request information. Ideally, the end-user and the person who actually needs the information are one and the same; the larger the number of intermediate transfers (i.e., of persons who have to pass the information along to the one who needs it), the more likely it will be erroneous -- or just too late -- on arrival. PeopleSoft and the ancillary software being implemented with the new system (for instance, Web applications) will give many people end-user status. In the enrichment phase (2000-2001), for example, Web applications will make it possible for students, faculty and staff to update their directory information online and in one process themselves -- no more going from office to office, hoping you've remembered to change your address with all those who might have reason to need it.

Workflow, in general terms, is the series of steps necessary to complete a task. Workflow analysis examines the steps to see which are redundant, duplicative, overly complex or entirely unnecessary. In the PeopleSoft system, workflow automates some business processes through the use of system updates and e-mail: forms are automatically transferred through successive steps until the required process is complete. Setting up a workflow path is also a good opportunity to analyze and enhance the efficiency and effectiveness of business processes. Once the system is implemented in the foundation phase, the university will have the opportunity to use system workflow to improve administrative efficiency and reduce bureaucracy.

Year 2000 (or Y2K) is the nickname of a large-scale programming fault in the way that year dates have been conventionally rendered in software. When computer memory was at a premium and every byte (the "1" or "0" to which all information is reduced in a computer program) was expensive, rendering a year date as two digits was half as costly as the four-digit version. Without those additional two digits, however, a program will not identify "00" as the year 2000 but will rather assume it to be 1900 -- and the program may crash (stop in an abrupt, unpleasant, or permanent way) as a result. These faults can be repaired only by expensive and time-consuming reprogramming. PeopleSoft is written to address this problem, saving the reprogramming costs that would otherwise be necessary.