how about the best (asymptotically)?
Kolmogorov Complexity. The problem is that it was
already reduced to the halting problem. :-)
And I'm not sure that predictability is closely
related to stetics. White noise and Stockhausen is
hard to predict. The beats in house music and La Monte
Young music are harder.
Markov chains do a great job. pherhaps by applying
hidden markov models to the language dave is building
would be nice because it would always yeld compilable
stuff.
marcel
--- Alberto de Campo <adc_at_khm.de> wrote:
> >On May 19, 2005, at 4:21 AM, Nick Collins wrote:
> >>programming style analysis- artificial agents look
> at a corpus of
> >>your coding style and try to predict the things
> you will do,
> >
> >Linear Predictive Coding!!! Ho ho ho.
> >
> >This is fairly insane. I like it.
>
> Well, _Non_Linear Predictive Coding would also guess
> at your
> next ideas before you have them ;-)
>
> adc
>
> --
> --
> Alberto de Campo
> Bergstrasse 59/33
> A-8020 Graz, Austria
> e-mail : decampo_at_iem.at
> --
>
___________________________________________________________
How much free photo storage do you get? Store your holiday
snaps for FREE with Yahoo! Photos
http://uk.photos.yahoo.com
Received on Wed May 25 2005 - 01:19:51 BST