Maximizing functional cohesion of comprehension environments by integrating user and task knowledge

Abstract

Program comprehension tools should facilitate the comprehension strategies used by programmers to achieve specific tasks. Many reverse engineering tools have been developed to derive abstract representations from existing source code and to apply a variety of analysis techniques. Yet, most of these software programs fail to provide users with the necessary guidance in choosing the appropriate methods, tools, abstraction levels and analysis techniques, and they frequently expose the user to unrelated information. The author presents a task and user-centered comprehension environment that maximizes the functional cohesion among the tools and comprehension techniques by focusing on a particular user task and its appropriate comprehension strategy. At the same time, we try to minimize the data coupling for the selected task by providing only the necessary task specific information, therefore reducing the data overload. This environment integrates user specific information with reverse engineered information to select the most appropriate comprehension strategy for a particular task.