Critical issues in evaluating freely improvising interactive music systems

As freely improvised music continues to be performed, it also continues to be implemented in interactive computer systems. For the scientific study of such systems to be possible, it is important to ensure the fitness for purpose of available evaluation methods. This paper will review several approaches to evaluating interactive computer music systems. It will also examine the uncritically-accepted assumption that quantitative evaluation invariably yields significant data, irrespective of context. Ultimately, it will be argued that, for some interactive computer systems, such as those designed for freely improvised music, qualitative evaluation by experts is the most appropriate evaluation method.