A Pilot Software Maintainability Study of a Scientific Program Library: Part II
Brad Carter · 1995
This reports the results of the second part of a pilot study to identify software product metrics that could help identify problem (high maintainability) programs being inserted into a scientific software library. The study investigates the relationship between a subjective assessment of maintainability by the technical staff and various product measures for seventeen products ranging from 2000 SLOC to 30000 SLOC. For this environment, module size, information flow, global variable usage, nesting depth, knots, Card''s system complexity, and certain entropy measures appeared to help identify the problem programs. The subjective assessment of the programs by the technical staff were compared to an ESPRIT SCOPE maintainability assessment and to a partial AFOTEC maintainability assessment. Neither compared favorably. The AFOTEC assessment was tuned to the staff assessment by using only fourteen selected questions.