Methods.
The program treats methods as a public commitment, not an appendix. What follows is the working version of the protocols the papers depend on — reproducibility, scoring, the case-recoding gate, the statistical posture toward simulations, and the risks that the program is tracking and not pretending it has solved.
The 49-case recoding gate
The Trust Force Equation paper (P3) currently relies on 49 author-scored cases. Author scoring is a private taxonomy, not a measurement. Until the recoding gate closes, every claim that depends on these cases is contingent and is labeled as such on the Roadmap and Papers pages.
Code, data, and the DOI commitment
Each paper ships with an archived code-and-data DOI deposited at OSF, Zenodo, or a comparable repository. The DOI is referenced from the cover letter, the data-availability statement, and the code-availability statement. The deposit is finalized before submission and is not revised after.
Simulations carry their seeds, their parameter sweeps, and their null models in the same archive as the figures they produced. A reviewer who wants to re-run any plot in any paper can re-run it end to end from the deposit alone.
Simulations are formal exploration, not validation
Stylized simulations probe the boundaries of the formal claims and check that the math behaves as advertised under known perturbations. They do not stand in for empirical data. The papers label them as formal exploration in the manuscript text, the figure captions, and the abstract.
Empirical validation arrives with the DP9 study (P4) at the earliest. Until then, no paper claims field support for any prediction. Reviewers who treat the simulations as evidence are engaging with a paper the program did not write.
Reviewer recruitment is parallel, not sequential
The program is recruiting three to five domain readers across commons governance, signed-network theory, percolation, and institutional repair. Recruitment runs in parallel with submission prep, not after. A list of named readers, with permission, will appear on /research/collaborate as it firms up.
Naming risks publicly is itself a control. The list below names a concrete failure mode and the specific mechanism the program uses to keep that mode in check. The controls are commitments, not aspirations — if a control is dropped, the corresponding risk should be assumed to be active. Operational red lines for KTP are named separately in the canon's limits.