Psychological complexity of computer programs

Abstract

In order to reduce the complexity of programs many ideas and techniques have been expounded. However, no quantitative evidence has been given that the quality of the programs has indeed been improved. We believe that experimental studies should be performed to measure those factors which make programs difficult to understand and maintain. The first step in such research is to establish a suitable experimental methodology. This paper describes a series of experiments which have been conducted with the aim of establishing such a methodology.