On quick comprehension and assessment of software

Abstract

By an architecture of a software system we mean the fundamental organization of the system embodied in its components, their relationships to one another and to the system's environment. It also encompasses principles governing the system's design and evolution. Architectures of complex systems are obviously complex as well. The goal of our research is to harness this complexity. In this paper we focus on providing software architects with ability to quickly comprehend the complexity and assess the quality of software. The essential tools we use are: (1) a graph-based repository for collecting information on software artefacts, accompanied by (2) tools to perform software intelligence tasks, like analyzing dependencies among those artefacts, calculating their importance, and quality. On top of those tools we implement visualization methods that render the relative importance using size and the quality using colours. By means of such methods a software architect can at glance comprehend and assess the software, He/she can (1) find the starting points to dig into a complex system; (2) judge the cohesion and coupling of system components; and (3) assess the overall quality. We demonstrate this method using selected open-source projects of various sizes and qualities.