teaching personal taste to silicate: a drawing machine for tacit thought

the second note in the paper series is out: CW:002, teaching personal taste to silicate. the series is indexed at cyberwitchery.com/papers.

why?

this one is about generative art, which is even further from network fabrics than the trolley was. the answer is the same as last time: the lab publishes on what it finds interesting.

i’ve recently hit 10 years of generative art practice, and i’ve been building my little studio since then. this note talks about what happened when a procedural drawing system was given a memory.

the paper

the machine watches the choices made inside the drawing system (palette, material, density, placement, gesture) and remembers which combinations were met with love, meh, or hate. the claim: with 2,196 of those quick judgments over a vocabulary of 222 authored decisions, it recovers about 22% of the variation in the judgments that follow, enough to change what appears next, not enough to settle the matter.

the more interesting half of the note is about where that vocabulary breaks. the compositions it cannot explain mark the edge of what taste will let itself be turned into features, and many of the best results live in the hate bucket.

if nothing else, it comes with some pretty pictures.

as before: exploratory, not peer reviewed. read it with that in mind, and argue with us if you think it’s wrong.