Software metrics: roadmap

摘要: Software metrics as a subject area is over 30 years old, but it has barely penetrated into mainstream software engineering. A key reason for this that most activities have not addressed their important requirement: to provide information support quantitative managerial decision-making during the lifecycle. Good implies risk assessment and reduction. Yet traditional approaches, often driven by regression-based models cost estimation defects prediction, little managers wishing use measurement analyse minimise risk. The future lies in using relatively simple existing build management decision-support tools combine different aspects of development testing enable make many kinds predictions, assessments trade-offs life-cycle. Our recommended approach handle factors largely missing from usual namely: causality, uncertainty, combining (often subjective) evidence. Thus way forward research causal modelling (we propose Bayesian nets), empirical engineering, multi-criteria decision aids.