Assessing testing tools in research and education

An evaluation of three software engineering tools based on their use in research and educational environments is presented. The three testing tools are Mothra, a mutation-testing tool, Asset, a dataflow testing tool, and ATAC, a dataflow testing tool. Asset, ATAC, and Mothra were used in research projects that examined relative and general fault-detection effectiveness of testing methods, how good a test set is after functional testing based on program specification, how reliability estimates from existing models vary with the testing method used, and how improved coverage affects reliability. Students used ATAC and Mothra by treating the tools as artifacts and studying them from the point of view of documentation, coding style, and possible enhancements, solving simple problems given during testing lectures, and conducting experiments that supported ongoing research in software testing and reliability. The strengths, weaknesses, and performances of Asset, Mothra, and ATAC are discussed.<<ETX>>