Implementing automatic testing is not so automatic

This work describes a process improvement experiment (PIE), that was designed to improve the software testing methodology, of Magic Software Enterprises Ltd. Magic's business is related to the provision of application development tools supported on a wide range of platforms, operating systems, and network, and interfacing with many data base management systems. The baseline project is part of a release of Magic Version 8-Enterprise Edition, that includes expanded capability in handling of Web connectivity. In the PIE, the data base gateways part of the product were re-tested using new automatic procedures, and were compared to the current manual method. This article mainly focuses on the qualitative results of the experiment, rather than the quantitative ones, to serve as an input for organizations who are considering implementation or intend to implement similar methods in their QA process. The project is part of an overall process to raise the level of development technology of Magic Software Enterprises from level one on the SEI CMM scale to level two.

[1]  John D. Musa,et al.  Software reliability measurement , 1984, J. Syst. Softw..

[2]  John D. Musa,et al.  Software reliability - measurement, prediction, application , 1987, McGraw-Hill series in software engineering and technology.

[3]  A. Wood,et al.  Predicting Software Reliability , 1996, Computer.

[4]  Akira K. Onoma,et al.  Practical Steps Toward Quality Development , 1995, IEEE Softw..

[5]  Moshe Pollak,et al.  A Semi-Parametric Approach to Testing for Reliability Growth, with Application to Software Systems , 1986, IEEE Transactions on Reliability.

[6]  Tom Rodden,et al.  Understanding the Software Process as a Social Process , 1992, EWSPT.