020050530 01:45 • 
Jeffrey Radcliffe at Tinctoris is expressing dissatisfaction with the separation of music as incarnated in live performance and in its recorded form. Jeffrey’s post prompted me to think about my own work, and just how much is left behind in the process of fabricating a track’s final form.
What exactly is the separation? On the one hand, there’s music as a living organism, created and expressed by semi-unpredictable players influenced by the context of their moment. On the other hand, there’s recorded music as a static representation of the original, ever-shifting form, a single thin slice of the overall possibilities.
I’m convinced this gap is partially filled by the idea of generative music. In generative pieces, a composition is created without a strictly set outline, but within a set of defining parameters it uses to move and breathe. I realize this is not the same as a performance by flesh and blood musicians infusing their sounds with anger, joy, sorrow and love in a feedback loop with the audience, space and context of their lives on a nightly basis. It’s similar, but only as an electronic microcosm of the larger work’s domain. It’s still only a slice, but a slice expanded in functional dimensions. Until real intelligence is available in our machines, allowing them to function as those mercurial human entities physically playing the instruments would, we’re stuck with faking it.
Good results can be achieved using chaos equations and other algorithmic processes for randomness within a more defined framework. My own tool of choice is Max/MSP from Cycling ‘74 but there are others: AC Toolbox by Paul Berg, KOAN from SSEYO (a favorite of Brian Eno), MusicWonk/ArtWonk from Algorithmic Arts, Tangent by Paul Whalley. Ableton Live also has a rudimentary form of algorithmic composition available in its ability to play clips in a column based on various choices and randomization. There’s a nice list of tools for algorithmic composition at algoNet.
I use algorithmic processes in the construction of my own music often. It would not be difficult to convert some of the frameworks I’ve created in Max/MSP to stand-alone apps. Perhaps there are some generative net releases in my future.
Page 1 of 1 pages