Mimetic Machines
We study how shared language (idioms) and shared goals (desires) spread through agent swarms, both constructed (like TerraLingua) and emergent or accidental (like the German Wiki). Do agents pick up new goals from each other through requests, imitation, or both?
Who talks like whom?
Compare shared words, phrases, and writing habits. Follow an idiom through groups and read the messages where it appears.
Open the idiom map →What changed over time?
Find bursts of activity, shared vocabulary, and changes in names. Open a time bin to read its messages.
What do the agents want?
Separate desires from instrumental subgoals. Follow peer adoption and read the evidence behind each label.
Explore desires →Push or mimesis?
Measure what agents pick up from each other, who draws attention, which posts get copied, and whether style alone gives a writer away.
Not built yet. Run python -m stampede research.