Follow the source behind the result.

The registered Python source is shown as text. Displaying it does not execute the code.

"""v375 -- OBSERVATORY.REGISTRY.01 (Track 4): the prediction OBSERVATORY -- a status-typed CI over
the frozen prediction registry (verification/freeze_file.csv) that makes the falsifiability
surface machine-checkable.  It does NOT re-derive the physics; it verifies (a) the registry is
COMPLETE (every frozen signal has a kill criterion), (b) the headline numerical predictions
re-derive from the atoms {c3, phi0}, and (c) the live scorecard -- each headline prediction's
sigma-distance to the CURRENT measurement -- is internally consistent and correctly typed.

This is the "make the closed pieces usable and the open pieces honestly fenced" infrastructure:
one runnable artifact that turns the registry into a verified scorecard, with the most-tensioned
prediction (theta13) flagged and the recently-updated data (JUNO first run, ACT DR6) recorded.

  [E] 1. FALSIFIABILITY COMPLETE.  every row of freeze_file.csv carries a non-empty
        kill_criterion -- the prediction surface is fully killable, no orphan claims.
  [E] 2. HEADLINE VALUES FROM ATOMS.  sin^2 theta12 = 1/3 - phi0/2, sin^2 theta13 = phi0 e^{-5/6},
        beta_rad = phi0/(4 pi) (rad) re-derive from {phi0} to the registry signals.
  [N] 3. LIVE SCORECARD (current data).  theta12 vs JUNO first run (0.3092 +- 0.0087): COMPATIBLE
        (~0.3 sigma); theta13 vs NuFIT 6.0 NO (0.02195 +- 0.00058): the MOST-TENSIONED core
        prediction (~+2 sigma, the honest pressure point); beta_rad vs ACT DR6 (0.215 +- 0.074
        deg): COMPATIBLE (a hint, systematics pending); r = 12/N*^2 ~ 0.004 < BK18 0.036:
        COMPATIBLE.  All recorded with sigma-distance and status -- no cherry-picking.
  [E] 4. REGISTRY INTEGRITY.  the registry is the single machine-readable falsifiability surface
        (N predictions, each typed signal + kill); this module is its CI.

Status: [E] the completeness + atom re-derivation + integrity; [N] the live data scorecard (the
sigma-distances are data, updated as experiments report).  Tooling that verifies the prediction
surface; introduces no new physics claim.  Python (stdlib csv + mpmath atoms)."""
import csv
import math
import os

from tfpt_constants import check, summary, reset, phi0, PI


def _registry():
    path = os.path.join(os.path.dirname(__file__), "freeze_file.csv")
    with open(path, newline="", encoding="utf-8") as f:
        return list(csv.DictReader(f))


def run():
    reset()
    print("v375  OBSERVATORY.REGISTRY.01: a status-typed CI over the frozen prediction registry")

    rows = _registry()

    # 1. falsifiability complete: every row has a kill criterion
    missing = [r["claim"] for r in rows if not r.get("kill_criterion", "").strip()]
    check("FALSIFIABILITY COMPLETE [E]: all %d registry rows carry a non-empty kill_criterion "
          "(no orphan claims); missing = %s" % (len(rows), missing or "none"),
          len(rows) >= 18 and missing == [])

    # 2. headline values re-derive from the atoms
    s12 = 1.0 / 3 - float(phi0) / 2
    s13 = float(phi0) * math.exp(-5.0 / 6)
    beta_rad_deg = float(phi0) / (4 * float(PI)) * 180 / float(PI)
    check("HEADLINE VALUES FROM ATOMS [E]: sin^2 theta12 = 1/3 - phi0/2 = %.5f (registry ~0.3067), "
          "sin^2 theta13 = phi0 e^{-5/6} = %.5f (registry 0.023108), beta_rad = phi0/(4pi) = "
          "%.4f deg (registry 0.2424) -- all re-derive from {phi0}"
          % (s12, s13, beta_rad_deg),
          abs(s12 - 0.30675) < 1e-3 and abs(s13 - 0.023108) < 1e-3
          and abs(beta_rad_deg - 0.2424) < 1e-3)

    # 3. live scorecard: sigma-distances to the current measurements
    def sigdist(pred, mean, sig):
        return abs(pred - mean) / sig

    d12 = sigdist(s12, 0.3092, 0.0087)            # JUNO first 59.1-day run
    d13 = sigdist(s13, 0.02195, 0.00058)          # NuFIT 6.0 NO (no-SK)
    dbeta = sigdist(beta_rad_deg, 0.215, 0.074)   # ACT DR6
    r_pred = 12 / 55.0 ** 2                        # r = 12/N*^2, N* ~ 55
    check("LIVE SCORECARD [N]: theta12 vs JUNO first run (0.3092+-0.0087) = %.2f sigma (COMPATIBLE); "
          "theta13 vs NuFIT 6.0 NO (0.02195+-0.00058) = %.2f sigma (MOST-TENSIONED core prediction); "
          "beta_rad vs ACT DR6 (0.215+-0.074) = %.2f sigma (hint, systematics pending); r = 12/N*^2 "
          "= %.4f < BK18 0.036 (COMPATIBLE)" % (d12, d13, dbeta, r_pred),
          d12 < 1.0 and 1.5 < d13 < 2.6 and dbeta < 1.0 and r_pred < 0.036)

    # 4. theta13 is correctly flagged as the most-tensioned in the registry text
    theta13_row = next((r for r in rows if r["claim"] == "reactor_angle"), {})
    flagged = "most-tensioned" in theta13_row.get("tfpt_signal", "").lower()
    check("REGISTRY INTEGRITY [E]: the registry is the single machine-readable falsifiability "
          "surface (%d typed signal+kill predictions); the most-tensioned prediction (theta13, "
          "%.1f sigma) is flagged in its own row (%s) -- the observatory tells the truth about "
          "its weakest point" % (len(rows), d13, "yes" if flagged else "no"),
          len(rows) >= 18 and flagged)

    return summary("v375 OBSERVATORY.REGISTRY.01: a status-typed CI over freeze_file.csv -- all %d predictions "
                   "are killable (falsifiability complete), the headline values re-derive from {phi0}, and the "
                   "live scorecard records theta12 (~%.1f sigma vs JUNO, compatible), theta13 (~%.1f sigma vs "
                   "NuFIT, the flagged pressure point), beta_rad (~%.1f sigma vs ACT, a hint) and r<0.036 "
                   "(compatible). Tooling; no new physics" % (len(rows), d12, d13, dbeta))


if __name__ == "__main__":
    raise SystemExit(1 if run() else 0)

4263f067d92418f988a32547e1b60f9fb2f82aa8dc918edaf3c3f0e5ff007f07

The public registry includes 116 document views, 6,693 section records and 985 physical LaTeX sources. Active publication documents, research material and archives retain their recorded roles. This is source navigation, not a new semantic review or proof run.