A Programming Environment Evaluation Methodology for Object-Oriented Systems

The object-oriented design strategy as both a problem decomposition and system development paradigm has made impressive inroads into the various areas of the computing sciences. Substantial development productivity improvements have been demonstrated in areas ranging from artificial intelligence to user interface design. However, there has been very little progress in the formal characterization of these productivity improvements and in the identification of the underlying cognitive mechanisms. The development and validation of models and metrics of this sort require large amounts of systematically-gathered structural and productivity data. There has, however, been a notable lack of systematically-gathered information on these development environments. A large part of this problem is attributable to the lack of a systematic programming environment evaluation methodology that is appropriate to the evaluation of object-oriented systems.