This folder is a living research workspace for persistent investigation into artificial gravity, inertial control, spacetime engineering, and adjacent physics that could eventually inform technologies often imagined in science fiction.
Important
This repository contains theoretical analysis and numerical experiments, not an experimental demonstration of artificial gravity, inertial control, reactionless propulsion, spacetime engineering, or faster-than-light travel. A solver artifact is evidence about the stated mathematical model only.
The standing premise is ambitious but disciplined:
- Artificial gravity may be a key enabling capability for deep-space habitation, high-acceleration travel, and more speculative propulsion concepts.
- Faster-than-light travel remains unsupported by known experimentally validated physics, so every idea must be tracked with its assumptions, evidence, failure modes, and testability.
- Failure is expected. Useful negative results should be preserved because they map the boundary of the possible.
- The work should combine orthodox physics, fringe-adjacent hypotheses only when clearly labeled, engineering constraints, historical dead ends, and new opportunities from modern theory and experiment.
- Research_Charter.md: mission, standards, scope, and rules of evidence.
- Nightly_Runbook.md: exact workflow for each midnight automation run.
- Daily_Log.md: chronological record of nightly work.
- Hypotheses_Register.md: ranked ideas, conjectures, and falsification status.
- Sources_and_Notes.md: citations, summaries, and source quality notes.
- Experiment_Backlog.md: calculations, simulations, reading tasks, and possible experiments.
- Boundary_Map.md: known limits, no-go results, and places where assumptions may be revisited.
- Open_Questions.md: unresolved questions for future runs.
The models/ directory contains reproducible bounding calculations
and numerical experiments. The maintained verification target is Python 3.14.
python3.14 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -r requirements-research.txt
python -m unittest discover -s tests -vAccepted .npz checkpoints and deliberately retained work snapshots are
integrity-checked research artifacts tracked with Git LFS. After cloning, run
git lfs install and git lfs pull. Their independent full-file digests are
recorded in models/checkpoints/SHA256SUMS.
Work snapshots preserve exact prior solver states so later agents can inspect
or resume the recorded path; a work suffix is not an acceptance or physics
claim. Disposable .partial.npz and .tmp.npz files remain ignored.
Each night at midnight, Codex should spend at least 20 minutes and no more than one hour doing research and thinking, then update this folder with:
- What was studied.
- What was learned.
- Which ideas became more or less plausible.
- What failed or reached a boundary.
- The next best question to pursue.
The 20-minute floor is a real work requirement, not just a scheduling label, and the intended use of the nightly window is deeper than a single quick calculation. If the first calculation or source pass resolves quickly, continue within the same narrow focus by adding sensitivity checks, alternate models, primary-source review, failure-mode analysis, comparison against existing boundaries, and a sharper next experiment or hypothesis. Do not stop after a short result simply because the result is decisive. A normal unblocked run should usually keep working toward the 45-55 minute range before switching fully to documentation.
The goal is not to prove a desired conclusion. The goal is to keep pushing intelligently at the boundary between known physics, engineering possibility, and speculative opportunity.
No open-source license has been granted for this repository. Public visibility does not by itself grant permission to copy, modify, or redistribute its contents.