Slashdot Mirror
An Experiment in A New Kind of Music
waynegoode writes "Stephen Wolfram's Wolfram Research has produced an new application:
WolframTones-- 'An Experiment in A New Kind of Music'. It combines the principles in Stephen's book, 'A New Kind of Science' and Mathematica to 'instantly create unique music' in many different styles. They describe it as pretty neat as well as being scientifically interesting, and useful. After listening to some compositions and creating a few random ones myself, I must agree that it is. And anyone who has listen to the radio the last few years could certainly use some unique music."
I wouldn't trust anything Wolfram says about his creations. He has a tendency to toot his own horn. Constantly. If you've read A New Kind of Science you know exactly what I'm getting at.
Read the section of ANKoS about the applications of cellular automata. It reads like, "I am the smartest man alive and cellular automata will change the way humans live forever."
I have to second this opinions. I couldn't find any sort of reference or acknowledgement to previous work on the subject.
Of course, I have a slight bias on the topic as my supervisor did something similar back in 1986.
(P. Prusinkiewicz, Score Generation with L-Systems, International Computer
Music Conference 86 Proceedings, 1986, pp. 455-457.)
** Sig-a-licious **
The best thing about music generation software is the potential to automate. Being able to generation thousands of random tracks while you sleep must be a real plus. Like having an artist that never sleeps (or overdose). Eventually some of them would have to sound as good as Megaman tracks.
If Wolfram's confection is indeed an experiment, then it ought to have a falsifiable hypothesis.
Show us the evidence that Wolfram's concoction involves any kind of "experiment."
Crackpots have been churning out music using mathematics for well over 50 years -- but none of this can be described as any kind of "science" or any sort of "experiment." Science involves falsifiable hypotheses...generating music with math involves touchy-feely squishy fuzzy "I like it" or "I don't like it" unfalsifiable subjective personal reactions.
Go ahead. Objectively prove via double blind falsifiable repeatable scientific experiment that a piece of music is good.
You can't. No one can. As Laurent Fichet showed in his 1996 book "Scientific Theories of Music the 19th and 20th Century," every allegedly scientific theory of music over the past 200 years falls apart on examination. It's all vacuous twaddle, nothing more than acoustic gematria. "Mathematical theories of music, based on acoustics, consistently contradict the practice of musicians." -- Paul Hindemith, 1947
Music is an art, not a science. Efforts to scientize the arts are as futile as efforts to mathematically predict the stock market (as the Nobel-induced collapse of Long Term Capital Management proved in 1997). Wolfram is here practicing pseudo-science, and has fallen into a numerological form of superstition no different from biorhythms, ufology, or astrology.
Someone has discovered a unlimited source of muzak! I can sense hordes of senseless HomePage Hobbiests(TM) reaching for their editors...
Any fool can talk, but it takes a wise man to listen.
Quite frankly, RTFA. I got sick of his own horn tooting on that site. You'd think that he was the only person that had anything to do with chaos theory and emergent behaviour; although he doesn't use those terms it seems.
m l
"It's a rather direct consequence of a core phenomenon of Stephen Wolfram's science: that programs with very simple underlying rules can generate great complexity of behavior."
http://tones.wolfram.com/about/faqs/howitworks.ht
His science? Please.
ANKOS would be pretty good with just a few changes:
Reduce page count from 1200 to 400 by removing redundant and self aggrandizing material.
Retract claims that Wolfram is singlehandedly going to change the course of human history.
Choose a title more suitable to the seriousness of the book. Perhaps "An Introduction to Cellular Automata" or "Fun With Graph Paper"
Judging by the "How WolframTones Works" page...
I saw a paper on exactly this a few years ago (perhaps written by these people?). I was particuarly disappointed in the uncreative approach to attaching it to music. Completely one-dimensional, based on a single pattern rule, using the results as a simple piano roll. In this particular example, it seems the programmer has inserted a few generic style and rhythm rules as well. Cute.
If the computation could generate anything more than a bunch of undirected pitches, I might be impressed. Perhaps have variables that can trigger harmonic shifts, considerations of form, independent patterns, definitions of rules for the next 10 seconds for an evolving pattern... SOMETHING more innovative than using it as a piano roll.
It's also disappointing that the score just takes a snippet of the whole pattern and truncates the rest. Some border rule treatment could have added to it.
Hopefully, this will be only the beginning of a much more interesting project. If this is the final result, my fascination has ended.
-- I prefer the term "karma escort."
If you know anyone tooting other people's horns, let me know.
Sounds like something coming from a person who has never created music. It's actually a craft, and needs a certain competence. And even the ancient Greeks knew there was a relation between mathematics and music.
To really see what they could be like, listen to Ramstein's version of Das Modell (or just read it's lyrics) and Siouxie and the Banshees' version of Hall of Mirrors. (Which's original version is the most sopoforic-in-the-literal-sense song ever.) On the other hand, if for nothing else, listen to The Showroom Dummies for the scary factor. But it's true, beneath the super repetitive factor, there is real meaning in their lyrics. Ja tvoi sluga, ja tvoi Rabotnik
A hundred years ago, Atonality was the 'new thing' and it was to give birth to Serialism. Naturally, the uneducated fools of the time proclaimed such things to not be "music" - famously, there was that so-called "riot" at the premiere of Stravinsky's Rite of Spring , which is now considered to be one of the finest pieces of music written in the 20th century.
Fast-forward to modern times though, and you wouldn't enjoy your television, cinema and gaming experiences nearly as much if it were not for the work of those pioneers who were writing things that were not "music" - so perhaps that sort of narrow-minded view is best left buried.
Just because you do not like a thing, does not mean that thing is wrong, does not mean that point of view is incorrect; it's like a farmer telling a city dweller that a domesticated mouse/rat is not a pet, and it should be destroyed because to him, it is vermin.
How is that different from rap?
Music generated from algorithms could ultimately be analogous. It might not be "art", but it could still be beautiful... with the beauty arising from the same simple, natural, relationships which underly a lot of how the world works.
Modern Classical composers(and neo-classical) do this to an extent. Music that is only composed if it conforms to certain rules of any given style. Yngwie Malmsteen for example, a hugely technically accomplished guitarist, plays and composes neo-classical guitar instrumentals that conform to rules of arpeggiation, chord structure...etc. It's an amazing thing to listen to, but it holds little of the emotion and imagery that can be made when composing without your first thought being about musical rules. Imprecision is one of the things that can make music the most beautiful.