Software-systems that expectedly or unexpectedly live very long frequently challenge software developers with problems of laborious and error-prone reverse engineering. Software measurement is a method that can be used to predict maintainability risks as early as during the design and implementation phases. However, its is a technique, that needs a sound empirical foundation and naive application can unintentionally easily yield invalid predictions. This case study sheds light on the question: What are risks of software measurement for the validity of the results and the acceptance of the surrounding quality assurance process? Examples from realistic project situations at DFS are given to answer these questions. One of the most important findings is that early involvement of software developers into software measurement programs is crucial for its success.
