Software Testability

AbstractA software component's testability or application's testability is regarded as a measure of the difficulty incurred when testing a component or application. If testing is considered as not being difficult, then the testability of the software is considered as being high. If testing is considered as being a difficult process to perform, testability is considered as being lower. The reason for seeking testability ratings is to determine early in the software development life cycle how many testing resources will be needed in order to complete the testing process. And typically, the testing process is not completed until some predefined testing stoppage criteria have been satisfied.Note that whether a program receives a high testability score is a function of the definition of testability. Each unique testability definition will require a different type of analysis to measure a program against that definition.Software testabilityhas historically referred to the ease with which inputs can be selected to satisfy a specific structural testing criterion (e.g.,branch coverage). With this definition, if it were extremely difficult to find inputs that satisfied a particular structural coverage criterion for a given program, then the program would be labeled as having “low testability.” Another view of software testability defines testability as a prediction of the probability that existing faults will be detected during testing, given a specific input selection criteriaC. Here, software testability is not regarded as an assessment of the difficulty to select inputs that cover software structure, but as a way to predict whether a program would reveal existing faults during testing according toC.Definitions based on coverage are fundamentally different from definitions based on the likelihood of fault detection. Testability is predicated on the definition employed. There are numerous definitions.