Evaluation criteria of software visualization systems used for program comprehension

Abstract

The program understanding task is usually very time and effort consuming. In a traditional way the code is inspected line by line by the user without any kind of help. But this becomes impossible for larger systems. Some software systems were created in order to generate automatically explanations, metrics, statistics and visualizations to describe the syntax and the semantics of programs. This kind of tools are called Program Comprehension Systems. One of the most important feature used in this kind of tool is the software visualization. We feel that it would be very useful to define criteria for evaluating visualization systems that are used for program comprehension. The main objective of this paper is to present a set of parameters to characterize Program Comprehension-Oriented Software Visualization Systems. We also propose new parameters to improve the current taxonomies in order to cover the visualization of the Problem Domain.