ABSTRACT
During the past two years, I have led a study team that has investigated in-depth how soœware maintenance was being conducted by large maintenance organizations.1 To develop §ndings, conclusions, and recommendations, the investigation has canvassed over two hundred soœware maintenance projects in §ve major locations via questionnaires and interviews. e goal of this e¥ort was to try to more fully understand what tasks soœware maintenance personnel actually perform day-to-day as part of their jobs. Based on these tasks, we next examined how the work evolved as the projects were planned, organized, sta¥ed, and executed. Instead of just focusing on the product, the team looked at all the work involved including management and support. We looked at the processes being used and the people using them from a skills and experience point of view. As our §nal thrust, we investigated how emerging tasks would in¶uence future e¥ort allocations. For example, many of the systems we examined were not built with security in mind. We felt that the work required to retro§t security into the product in the future would become a signi§cant part of the maintenance workload.
