Maintaining centralised application systems: a cross-country, cross-sector, cross-platform comparison
- PDF / 397,194 Bytes
- 15 Pages / 594 x 797 pts Page_size
- 59 Downloads / 172 Views
1997 Operational Research Society Ltd. All rights reserved 0960-085X/97 $12.00
Maintaining centralised application systems: a cross-country, cross-sector, cross-platform comparison W-G Tan1 and GG Gable2 1
Institute of Systems Science, National University of Singapore, Kent Ridge, Singapore 119597; 2Faculty of Information Technology, Queensland University of Technology, GPO Box 2434 Brisbane, QLD Australia 4001 Past studies of software maintenance issues have largely concentrated on the average North American firm. While they have made a substantial contribution to good information system management practice, it is believed that further segmentation of sample data and cross-country comparisons will help to identify patterns of behaviour more akin to many less average organizations in North America and elsewhere. This paper compares the Singapore maintenance scene with the reported North American experience. Comparisons are also made between: Government organizations, Singapore corporations and multinational corporations (MNCs); mainframe and minicomputer installations; and fourth-generation language (4GL) and non4GL computer installations. Study findings, while in many cases were similar to earlier US studies, do show the importance of Singapore’s young application portfolio, the widespread usage of 4GLs and the severe maintenance personnel problems.
Introduction Software maintenance has often been cited as one of the critical issues confronting organizations with a large investment in computer systems (eg, Bunday, 1988; Neiderman et al, 1991). Researchers have attributed this view to two common concerns of managers who are responsible for the upkeep of the application portfolio. First, the task of maintaining application systems is much more complex than has been portrayed in the literature. This has led to problems of degenerating software quality, poor maintenance productivity, low programmer motivation and other management concerns. Second, the maintenance activity consumes a major portion of the information systems (IS) budget. Both Boehm (1981) and Martin and McClure (1983) estimated that software maintenance incurs as much as 75% of overall software lifecycle costs. Conte et al (1986) claim that maintenance accounts for over 60% of the total software engineering effort. IS researchers have aptly described the high cost of maintaining application systems as the ‘maintenance burden’. The unit of analysis in past studies of software maintenance management issues has tended to be the average North American firm. While this past research has made a substantial contribution to good IS management practice, it is believed that further segmentation of sample data and cross-country comparisons will help to identify patterns of behaviour more akin to many less average organizations in North America and elsewhere. Questions of interest to maintenance managers thus include, “Do all computer using organizations face the same
maintenance problems?” and “What contingencies or contextual factors influence these problems?
Data Loading...
