This is a commons, not a project. There are no maintainers, no roadmap, no deliverables.
If you find something here worth picking up — to formalize, test, extend, or disprove — please do. That's the point.
Formalizing the framework:
- Dynamical systems analysis of the attractor geometry claims
- Information-theoretic grounding of "compatibility bandwidth"
- Formal definition of the relay ↔ diffusion regime boundary
Empirical work:
- Probing Hyperloop's internal representations across loop iterations (do they form coherent trajectories?)
- Toy model implementation of the two-regime propagation model
- Testing whether loop depth correlates with diffusion regime emergence
Connecting to existing literature:
- Biological analogues — does the pattern lifecycle map onto anything in neural development or evolutionary biology?
- Relationship to existing work on attractor networks, reservoir computing, or dynamical systems approaches to cognition
- Connection to evolvability literature (Nugget 8 suggests this)
Critique:
- Identifying where the framework makes predictions that could be falsified
- Finding where "theory of how intelligence works" and "theory of how we'd want it to work" diverge
Open an issue or PR. No formal process. If you're building on this, a note in your own work pointing back here would be appreciated but is not required.
Original exploration by Zee, April 2026. Developed in dialogue with Thea (ChatGPT, OpenAI) and Claude (Anthropic). The ideas belong to whoever can make something useful of them.