Dartmouth Symposium on the Future of Computer Music Software: A Panel Discussion

[Editor's note: This article is an edited transcript of a panel discussion, moderated by Eric Lyon, at Dartmouth College in October 2001. See "About This Issue" on page 1 for more information.]

Eric Lyon: I think it's fair to say that without the work of today's speakers, computer music as we know it would not exist. Now that we have them all together, we have an unprecedented opportunity both to focus on their individual achievements and to assess the bigger picture of where computer music software is today and where it might be heading.

At the end of a 1989 comprehensive survey of computer music languages and systems, Gareth Loy wrote the following: "It is remarkable how few of the languages and systems documented above are generally available." Among the reasons postulated for software's demise were the obsolescence of the host computer system, competition from other languages, and discontinuation of support by the programs' creators. This symposium takes Gareth's observations as a point of departure. But we turn the question around and ask not why so much software has disappeared, but why a few exceptions have been so persistent and have developed a loyal and even fanatical following among computer musicians.

In the world of recorded music, which comprises most of what we listen to these days, the term "computer music" is redundant. The prevalence of the use of computers in today's music demands another distinction; at its outset computer music meant experimental music, carried out in laboratories and universities. This experimental work continues here and at many other institutions, but most of today's computer music is created in the field of entertainment, whether film music or the various technology-drenched genres of rap, techno, rock, and pop.

The distinction between experimental music and what might be termed "normative music"—that is, music based on accepted stylistic norms—is mirrored in our software. On the normative side of software are utility programs such as mixers, sequencers, and reverberators. On the experimental side are the programs that we discuss today. This software is open, extensible, and invariably used in ways unanticipated by its creators. While such software does not command a market on the scale of normative utility programs, it is arguably much more influential in the long run, as it facilitates the creation of the music which today exists only in our collective imagination. And the experiments of today will lead inevitably to the norms of tomorrow. This debt is even occasionally acknowledged, such as when the Beatles put Stockhausen on the cover of their album Sgt. Pepper's Lonely Hearts Club Band, or more recently when Radiohead sampled (and credited) Paul Lansky on their album Kid A which went platinum.

My hope is that this will be a little less like a panel discussion and more like a town meeting, given the people in the audience. I am going to start by asking one or two questions, then open it up to questions from the audience.

I would like to start by discussing various aspects of different kinds of software, different specific pieces that we know and love. Perhaps this is more or less about survival paradigms, which is to say that specific pieces of software and ideas have been filtered through—have made it through—the 40-odd years of computer music. In particular, we see the "Music N" paradigm represented by the Music N environments and of course Csound, and we see the graphic layout and connection of signal of Max/MSP, and finally we see the pattern-based and syntactical sort of object-oriented idea of Super Collider.

Figure 1.

Max Mathews and Jon Appleton.

To start our discussion, I would like to ask how you feel about the idea of "software cannibalism" and how that may suggest a direction for the future. In other words, now that we have software models that are very clear and palpable, it also...