Software Maintenance Research in the Mire?

The theme of the 1999 Workshop on Empirical Studies of Software Maintenance is phrased as: ‘How to wade through the mire of evidence?’. In this paper, an overview is given of some of the evidence, as it was presented at the 1998 International Conference on Software Maintenance [8]. Being the main international conference on software maintenance, we believe we can assume that the proceedings of that conference contain a fair representation of current research in the area of software maintenance. Though the focus of the conference is not solely on empirical research, we would still expect most papers to be of an empirical nature, simply because the object of study – software maintenance – exists in the real world. In the next section, we discuss a subset of the papers presented at ICSM’98. Most of the papers investigate some process, method, or tool that can be applied in a maintenance environment. Of each paper assessed, we will discuss the research technique applied, the setting of the study, and the source of the data used. Since research in software maintenance is ultimately concerned with improving the efficiency, effectiveness, and hopefully also the enjoyment of software maintenance, we would expect a fair number of papers to report on experiments or experiences using processes, methods, and tools in real maintenance environments. Hence, our a priori hypothesis is that such reports constitute a fair amount of the total number of papers in the ICSM proceedings.