Keep the whole idea pool
Grouped, searchable proposals. No proof or novelty claim is required to be archived.
Browse all hypotheses →31 October 2026 · Studies 1 & 2
A place for every question.
A clear label for every claim.
Explore 4,200 supplied hypotheses in two collections. Follow their source-selected leads and five bounded mathematical analyses, with original wording, counterexamples and files preserved.
A reason to look closer
Supplied feasibility shortlists, including 45 deeper dossiers. Evidence ranges from bibliographic leads to assessed sources; proposed effects remain unproved.
Explore the leads →What follows from a model
Written conditional proofs, reproducible finite checks, assumptions and failure cases. No physical experiments are claimed.
Read the model results →One pool, nested selections. The leads, deeper dossiers and model results are subsets of the hypotheses, not extra ideas. The archive preserves unsuccessful, replaced and unselected proposals. Highlighting controls attention; it does not decide what is retained.
Two supplied studies
| Collection | Hypotheses | Leads | Deeper dossiers | Model results |
|---|---|---|---|---|
| Study 1 | 2,700 | 300 | 30 | 3 |
| Study 2 | 1,500 | 150 | 15 | 2 |
Study 1’s source directory is named “3000”, but its supplied IDs run H0001–H2700: the actual total is 2,700. Its 2,200 related experimental variants remain labeled as family members. Study 2 preserves N0001–N1500.
Read it at the right level
Background literature can make an idea plausible without showing its predicted effect. A written mathematical proof can settle a specified model without verifying that real materials obey it. These distinctions stay beside the claims.
Download every original file, full catalogs and provenance →