protocol.json

{
  "schema": "oph-1011-data-release-contract-v1",
  "canonical_target": "FZ-10",
  "owning_science_issue": 546,
  "comparison_issue": 1011,
  "state": "NOT_READY",
  "prediction_source": "evidence/custody/falsification/frozen_targets/fz10_2026-07-28/frozen_target_koide_conditional_tau_2026-07-28.md",
  "rule": "FZ10-R01 in listed-bullet order: FAIL, COMPATIBLE, otherwise INCONCLUSIVE",
  "reference_models": ["same balanced Koide relation", "free tau mass after electron and muon inputs"],
  "dataset": null,
  "registration_evidence": null,
  "selection": "Earliest first-public release after public registration of this dataset contract of a dedicated experimental charged-tau mass measurement; tie by canonical identifier. Select from metadata before reading value or uncertainty. No world averages, theory estimates, reprints or updated versions of pre-registration results.",
  "admission": "After nomination, retain this candidate even if ineligible. Require public source bytes and hashes, first-public timestamp evidence, event/run overlap inventory against exposed measurements, exposure declarations from comparison participants, mass convention, all calibrations and systematic/covariance conventions, and the official combined standard uncertainty. Missing admission inputs yield NOT_READY, not a replacement candidate.",
  "unexpected_exposure": "Metadata fields do not guarantee blindness: a title or search snippet may contain a result. Record the person and exposed release, retain that release in chronological nomination, and block its admission. Do not hide the exposure or nominate a later release instead.",
  "nuisance_policy": "Use the collaboration's prereported result, calibration and combined standard uncertainty unchanged. Do not fit nuisance parameters to FZ-10, pool results, convert asymmetric intervals silently, assume independent systematics, or reinterpret a confidence limit as a standard uncertainty. Keep the historical FZ-10 electron/muon inputs, center and window fixed; no recalibration to later CODATA editions.",
  "statistic": "Absolute distance from the canonical frozen center divided by the reported standard uncertainty; retain a separate full-window robustness assessment. No global chance probability or OPH-versus-Koide evidence ratio.",
  "multiplicity": "One nominated release and one charged-tau mass observable; one primary verdict. Component errors and window sensitivity are diagnostics, not additional attempts.",
  "stopping": "Stop after that release, including failure, insufficient precision or blocked admission. Do not wait for a favorable update or swap experiments. Publish all outcomes and metadata exclusions. Corrections retain the original outcome and require a separately versioned assessment.",
  "outcomes": ["NOT_READY", "FAIL", "COMPATIBLE", "INCONCLUSIVE"],
  "resource": "Existing primary public release only; local arithmetic. No hardware, paid data, cloud job or automatic polling is authorized.",
  "blockers": [
    "No eligible untouched measurement identified by the bounded metadata discovery; this is not a claim that no unindexed or private result exists.",
    "No selected dataset, run-overlap/exposure inventory or experiment-specific uncertainty/nuisance dossier is available.",
    "This execution contract has no claimed external registration evidence yet; historical FZ-10 custody alone does not freeze a new dataset selection."
  ]
}