INFORMATION GATHERING IN SOFTWARE PROCESS ASSESSMENT

摘要

Software process improvement starts with assessing the actual development process in a software development organization. While bad input can only produce bad output, it is important that software process assessment is based on stable and reliable information reflecting the actual 'real' process in an organization. Currently used software process improvement methods rely on process information which is gathered during interviews and brainstorm sessions, and from quality manual and process standard reviews. Interviews and brainstorming represent a so-called 'asking strategy' and quality manuals and process standard reviews represent a 'deriving' strategy for information gathering. Accompanying information gathering techniques are respectively interview and brainstorming techniques, and document review techniques. We argue that the selection of these strategies and these techniques is not based on a well-founded and systematic approach. As a consequence, the information that is gathered in a software process assessment is often inappropriate and offers a very restrictive basis for software process improvement. We address the determination of the level of uncertainty in software development systems and their processes as a basis for the selection of an information gathering strategy and techniques. We show that a more suitable information gathering strategy, and accompanying techniques, can be selected on the basis of the high level of uncertainty of a software development system. In highly uncertain software development systems, an information gathering strategy called 'synthesis' and 'experimentation' is needed, and that information gathering techniques derived from software process mining research, can offer a stable and reliable basis for software process assessment and improvement.