Re: Virtual Concerto

Subject: Re: Virtual Concerto
From: Eliot Handelman (
Date: Thu May 13 2004 - 20:12:56 EDT

Dennis Bathory-Kitsz wrote:

>At 06:00 PM 5/13/04 -0400, Eliot Handelman wrote:
>The list now is not terribly long: Besides
>>Matt and George, there's
>>also Clarence, Brad, and I think Doug. There should be many more?
>Lots more, depending on your taste. Henning Berg and Peter Beyls have
>created their own software. Other software that's interactively
>creative/re-creative was written by Clarence Barlow. Sarah Peebles, yes?
>Maybe Eric Lyon?

>Pre-PC/Mac software would include David Behrmann and,
>gosh, even me. What do you suppose qualifies? :)
A complicated question for me. Ultimately the music.
Program complexity (not necessarily complicatedness)
is probably a big issue. George mentioned that in his interview with you.

The ability to create form, as opposed to using templates, or just looping.

The sense of human agency, that something appears to
be vested with human purpose (better yet, post human, assuming
we're not all going to turn into stupid algorithms).

Richard Zvonar said: "There's something musically satisfying going on
that is
beyond simple Mickey-Mousing."

I'd say that a very salient feature is that the program seems to have
listened to itself. As George said, the program shouldn't behave
like a bonehead.

The program could also frame ideas about what making music is,
or what composing, in any interesting sense, could be. The point
above also suggests that the program could frame ideas about what
listening is.

I personally would not like to see this going in the way of fake AI --
ie, the ALICE chatterbot, by having a huge database and transition
probablities that it can haul stuff out of, without any conceptualization
at all what might actually be taking place. Of course, this is how things
WILL go because the idea is sufficiently pointless and stupid for
it to have mass academic appeal.

-- eliot

This archive was generated by hypermail 2b27 : Sat Dec 22 2007 - 01:46:01 EST