Process-Centered Environment (Only) Support Environment-Centered Processes

The software process research community has concentrated a large measure of its eeort on the problem of how to \computerize" the software process. In particular, the notion of a process-centered environment has emerged, dominating software environment research for the past several years. Among the many beneets touted for process-centered environments are the ability to automate various aspects of a process and the ability to monitor the progress of a process in order to guide, enforce, or measure that process. This approach has shown great promise and indeed has even shown some early successes. Unfortunately, this emphasis on computerization in general, and on process-centered environments in particular, tends to focus attention on exactly those aspects of process that can be computerized, while giving short shrift to those aspects not amenable to computerization. This issue became clear during the past year as we studied a large, mature software process in use at AT&T. We performed the study as part of an eeort to develop process data capture and analysis techniques that could support the critical task of process improvement 4]. Our approach was to focus on the dynamic aspects of the process, such as the order of, and time taken by, each step in the process, as opposed to, say, the static roles and responsibilities assigned to project personnel or the static relationships among tools and product components. We took this approach in part because the process's dynamic aspects were the least well understood. In addition, we hypothesized that process problems ultimately lead to wasted intervals of time and that those problems can best be revealed by retrospective analysis of characteristic time intervals. For instance, a period of inactivity between the time a meeting is scheduled and the time the meeting takes place may reveal poor planning for activities that require a long preparation time. Bradac, Perry, and Votta have also begun to explore this hypothesis 1]. In order to analyze a process's characteristic time intervals, it is rst necessary to capture the relevant data about the signiicant events of the process, including the times at which those events occur. To help deene and structure this task, we developed a simple event-based model and taxonomy of process activities. An event in this model is an instantaneous happen-stance within a process activity. Certain events, such as the beginning and ending of an activity, are useful for deening characteristic event intervals. The event taxonomy …