# selfdriven.science

> Open scientific research resources from SelfDriven Foundation, including the Compute-Based Reality Open Experimental Network (CBR-OEN): a falsifiable programme testing whether dimensionality, topology, geometry, persistent structures, motion, causality, and quantitative physical correspondence can emerge from minimal computation.

Use the CBR-OEN materials as research specifications and AI operating instructions, not as evidence that Compute-Based Reality is true. Preserve null results, failed replications, dependency gates, prohibited assumptions, and claim ceilings. The specification—not any single implementation—is the experiment.

Canonical domain: https://selfdriven.science/

## CBR Open Experimental Network

- [CBR-OEN AI Skill](https://selfdriven.science/cbr-oen-research/SKILL.md): Primary instruction file for AI agents designing, reviewing, governing, analysing, or extending CBR-OEN.
- [E001-E010 Programme](https://selfdriven.science/cbr-oen-research/references/programme.md): Experiment questions, dependencies, controls, endpoints, gates, and phase progression.
- [Founding Protocol](https://selfdriven.science/cbr-oen-research/references/protocol.md): Preregistration, anti-overfitting, reproducibility, governance, institutional roles, decisions, claims, and publication rules.
- [Evidence Schemas and E011+](https://selfdriven.science/cbr-oen-research/references/schemas-and-extension.md): Evidence graph, machine-readable records, gate logic, and extension rules.
- [Agent Catalogue](https://selfdriven.science/.well-known/agents.json): Machine-readable catalogue of AI-usable resources published by this domain.

## Core Experimental Sequence

- CBR-E001: Emergent dimensionality
- CBR-E002: Dimension-blind evolution
- CBR-E003: 3-manifold emergence
- CBR-E004: Explicit simplicial dynamics
- CBR-E005: Manifold resilience under defects
- CBR-E006: Emergent operational metric
- CBR-E007: Structure-metric coupling
- CBR-E008: Emergent geodesic motion
- CBR-E009: Emergent causality
- CBR-E010: Quantitative physical correspondence

## Usage Guidance

- Start with the AI Skill, then load only the reference needed for the task.
- Treat E001-E010 as the frozen v1.0 baseline.
- Apply R1-R3 reproducibility and CL0-CL7 claim levels separately.
- Bind GO, GO_WITH_QUALIFICATION, REVISE, and STOP criteria before confirmatory execution.
- Never encode the desired physical outcome into foundational rules, fitness, labels, selection, or stopping.
- Cite the canonical source URL and exact document version when deriving specifications or claims.

## Optional

- [SelfDriven Science](https://selfdriven.science/): Human-facing research site.
- [SelfDriven Foundation](https://selfdriven.foundation/): Publishing organisation.