Mehmet Duran is an experienced senior Test Consultant working at one of the largest IT companies Capgemini. Now, through his company he offers experienced-based testing and QA service which improves the software quality within the organizations in the Netherlands. He is a leading expert in software test automation tools. At one of his customers in the rail sector he has implemented test automation within a complex IT landscape. As a senior consultant, he is focused on implementing innovative solutions around the software testing Mehmet also offers training, and consulting services helping clients evolve their software quality and make it more efficient.
Topic: Challenging TESTAR in an industrial setting
TESTAR has an innovative approach at the Graphical User Interface (GUI) level. We did a scientific research in the Dutch rail sector to compare this innovative approach to the used manual approach. We developed a scientific framework to enable this comparison. In this research, two perspectives have been conducted to investigate the sensemaking and technological value of TESTAR. ‘Sensemaking’ is the process by which people give meaning to events. TESTAR is a traversal-based and scriptless approach (without test cases) make it different from other GUI testing tools. Moreover, this tool it automates not only the execution of the tests, but also test design. Since no test scripts are needed, it is expected that TESTAR solves a significant part of the maintenance problem from which other approaches suffer. At present, a limited amount of studies has examined how effective, efficient, learnable and satisfactory the innovative approach is in an industrial setting. The results can provide sufficient information to help companies judge if a specific technology being evaluated will benefit their organization. Our case study compares the test automation approach of TESTAR to the manual test approach within a realistic industrial setting (rail sector) with a real mission critical web-based software application. Both qualitative and quantitative research methods are used to investigate learnability, effectiveness, efficiency, and satisfaction. Beside this factor we also analyzed the sensemaking of the test automation tool. The results show that within this setting, TESTAR was able to detect more faults and had a higher functional test coverage than the used manual test approach.