Skip to content

The Board · Evidence stale

Forecast ledger: no current finding

The source is stale. Palimpsest keeps the availability and evidence record visible without publishing retained values as current news.

What the record says

Availability. No current finding is published for Forecast ledger because the source status is stale.

This report is scoped to forecast-ledger. It does not merge unlike instruments or infer a cause from co-movement.

How it was measured

strictly prequential (Dawid 1984): each reading is forecast from only its own past, never refit with hindsight. Point forecast = previous value (random walk). Band = adaptive conformal quantile of past absolute one-step changes, updated by alpha_{t+1} = alpha_t + gamma_t*(alpha - err_t) with decaying gamma (ACI, arXiv:2106.00170; decaying steps, arXiv:2402.01139), valid under arbitrary distribution shift. Scored by the Weighted Interval Score (arXiv:2005.12881), a proper rule, against the same forecast using a fixed quantile. Lower WIS is better.

What this cannot establish

  • Current finding withheld: freshness deadline has passed
  • Calibration describes aggregate historical coverage and does not guarantee that any one next reading falls inside its band.

Instrument brief

availability-brief · Evidence stale

Palimpsest withholds a current finding for forecast-ledger: the source status is stale. This is an availability brief, not a measurement.

Key number

  • withheld headline metric · withheld because the source is not live

Current number and denominator

  • No current number is published for forecast-ledger because the newsroom status is stale. The retained metric is withheld.
  • No current finding is published for Forecast ledger because the source status is stale.

Same-edition board context

  • Same-edition elevated layers: none declared. Other live lead or high signal ids: event-flags, ddti, undertext, ooni-gfw, in-path-interference, censored-planet, inside-view, ioda-outages, vantage-fusion, erasure-observatory, wayback, app-storefront, apple-censorship, cny-fix-gap, data-darkness, generative-firewall.
  • No editorial_priority or MAD score is treated as a finding on this reading.

What this reading does not show

  • Current finding withheld: freshness deadline has passed
  • The brief does not assign motive, identify a person, merge unlike denominators, or treat a missing reading as a zero.
  • strictly prequential (Dawid 1984): each reading is forecast from only its own past, never refit with hindsight. Point forecast = previous value (random walk). Band = adaptive conformal quantile of past absolute one-step changes, updated by alpha_{t+1} = alpha_t + gamma_t*(alpha - err_t) with decaying gamma (ACI, arXiv:2106.00170; decaying steps, arXiv:2402.01139), valid under arbitrary distribution shift. Scored by the Weighted Interval Score (arXiv:2005.12881), a proper rule, against the same forecast using a fixed quantile. Lower WIS is better.

Elevated peers this edition

event-flags (command), ddti (attention), undertext (attention), ooni-gfw (network), in-path-interference (network), censored-planet (network), inside-view (network), ioda-outages (network), vantage-fusion (network), erasure-observatory (erasure), wayback (erasure), app-storefront (erasure), apple-censorship (erasure), cny-fix-gap (economy), data-darkness (economy), generative-firewall (models)

Counterreadings

  • A current or withheld instrument value is not independent corroboration of any publisher article, and it is not a cause.
  • Co-movement with another live signal on this edition is layer context, not a shared rate or a coordinated action.

Limitations

  • Current finding withheld: freshness deadline has passed
  • The companion copies newsroom templates only; it adds no interview, article body, or generative prose.

Analysis instrumentv-dab350b05d45120c0c31d68a · structured assessment · reading companion

Read the evidence

The exact source reading is forecast-ledger-latest.json. The structured version of this dispatch is published beside the article.