A review on fault taxonomies for web testing

Abstract

For functional testing of web applications, fault-seeding is a method in which a software tester creates several modified versions of a web application that contain errors. They are useful for discovering faults that can compromise a web application’s behaviour. Using a fault taxonomy to guide the fault-seeding of a system increases the confidence that the seeded faults are realistic. This paper presents a systematic review of fault taxonomies that are introduced for testing web applications. Selection of relevant literature is performed by applying the inclusion and exclusion criteria. Six significant fault taxonomies for web testing are then selected and analysed. We suggest three fault classes or sub-classes that can be incorporated into a fault taxonomy namely the warning notification fault, media fault and orphan page fault. Finally, some major areas where the fault taxonomies are commonly applied are health software applications, cloud computing, Android applications and desktop applications.