Visualizing Software Quality

摘要

When software projects reach a considerable size (multi-person, multi-man-year) they are difficult to maintain. Which results in high efforts for software maintenance, code duplication, bugs, and other quality related problems. The consequences include limited software life cycles, especially in areas where software evolves rapidly. For example in an industry setting, business applications are usually re-implemented every 3-5 years. This is commonly considered as the only foolproof strategy currently available. It is also considered as a huge waste of resources and a true sign of the software crisis. Quality initiatives could significantly improve the software quality, but they usually suffer from tight budgets, lack of support and other neglectful factors. Also, due to the fact that software can be immensely complex, complete testing is impossible. So it is desired to drive testing efforts to parts of the software product that would benefit the most (risk considerations). The root cause of the software crisis lies in software engineering itself. Software is the most complicated product we make and we can not see it, or touch software like we can see or touch the work products of other engineering disciplines, such as construction or mechanics. During visual inspections you instantly see whether a car misses the bumper, a door or the wind-shield. The quality of a software project is not as obviously apparent through visual inspections. What if we could make the quality of software visible? This could be used to govern testing initiatives and could help to decide if the software is really ready for release. This paper provides in-depth insights and experience on interactively visualizing different aspects of software quality.