The canonical activities of reverse engineering

This paper describes three categories of canonical activities that are char acteristic of reverse engineering for program understanding. The activities are data gathering, knowledge management, and information exploration. All tasks carried out by a software engineer during a program understanding ex ercise can be mapped to a composition of one or more of these canonical act ivities. The design space formed by the canonical activities can be used to classify the capabilities provided by individual support mechanisms using a common vocabulary. A descriptive model that categorizes important support mechanism features based on a hierarchy of attributes is used to structure the canonical activities.