Perspectives on software quality assurance (session QS4 addition): my perspective

The moderator has offered seven propositions relative to the meaning and conduct of effective software quality assurance. Although I find it difficult to give unqualified acceptance or rejection to any of these propositions, in response to the rules of the session I accept propositions 1, 2, 3 and 5 and reject 4, 6 and 7.After a number of years experience in the software quality arena, both in government and industry, I have come to the conclusion that the most important aspect of software quality assurance is to make a positive contribution to the objective of the software project. I define this objective as the development of a quality software product delivered on time and within budget. I further define a quality software product as one which meets customer requirements and provides usable service. In short my perspective of software quality assurance is one of contribution in contrast to the “righteousness of the cause.”To this end I suggest an “engineering analysis approach” to software quality. Such an approach can take the form of a typical engineering problem solving model with steps of problem formulation, problem analysis, search, decision, specification and implementation. Using such an approach the advocates for quality software can use empirical evidence to establish the contributions software quality assurance efforts make toward the software project objective.