I was very excited to have the opportunity to review this paper, given my own personal interests in music and software development. Vickers and Alty have contributed a fascinating paper that is among the best I have encountered in computer science. They combine excellent writing with comprehensive coverage of the principals of sound perception, and finish with some of the best experimental design I have seen in the field of computer science.

It is a shame that a written publication cannot include sound, as reading this paper made me want to hear the musical rendering of the software that it describes. There is a Web site mentioned at the end of the paper, but I confess to being very disappointed by the very limited samples it provided. I am left with a great deal of interest in the end results the techniques described in this paper produce, and want to ask many additional questions of the authors: What results would be attained with larger sample sizes, and different experience levels? How would the result differ by performing this experiment in the US, India, and China, to name just a few locations, given the differences these countries represent in their languages and music, and their increasing significant to the software profession? How would a large C++ program sound? Would it really aid the understanding of its behavior, just as we used to be able to hear the disk access patterns in our early PCs? This paper incites many interesting questions, and only hints at their answers.