Scanner Lineage

Twenty versions. One sealed substrate.

20
Documented versions, February to April 2026
2
Negative controls in the lineage
v13.1.0
Production, locked at SHA256 e5250de8e9de07d6

The scanner that produced the v13.1.0-production-day-zero-full sealed snapshot is the result of a documented development lineage spanning twenty versions across three months. The lineage organises into three eras and one corrective iteration. The v11.1 era produced the historical baseline of 511 institutions used by the first three working papers in the corpus. The v11.1.1 attribution overlay introduced cryptographic provenance at every pipeline stage. The v12.1.1 production scanner was certified through capture-replay validation on an anonymised G-SIB. The v13 corrective iteration tested four targeted interventions: v13.0.0 cleanly falsified one of them, v13.1.0 retained the three confirmed interventions and replaced the falsified one with a minimalist counterpart. Two negative controls in the lineage, the v11.0 edge compression bug and the v12.0 silent-degradation parser failure, each produced directionally predictable output changes upon correction. The instrument responds to known interventions in the predicted direction at the predicted magnitude. That is the structure of a validated measurement.

Figure 1 · Scanner Version Lineage
TWENTY VERSIONS, FEBRUARY TO APRIL 2026 · v13.1.0 LOCKED AT SHA256 e5250de8e9de07d6 v11.1 HISTORICAL BASELINE 511 institutions / 84,379 agents / 344,541 edges ATTRIBUTION cryptographic provenance PRODUCTION + CORRECTIVE ITERATION v13.1.0: 543 inst / 95,876 agents / 626,390 edges v5 Feb foundation v8 Mar context strat. v9 Mar evidence tiers v10 Apr 61% inferred v11.0 Apr compression bug SHA v11.1 Apr baseline (511) v11.1.1 17 Apr attribution SHA v12.1.1 19 Apr capture-replay v13.0.0 20 Apr falsified SHA v13.1.0 20 Apr PRODUCTION v13.2.0 24 Apr incremental historical active production falsified, corrected SHA cryptographic seal applied Twenty versions. Two negative controls. One sealed substrate. MAR®500.com

The timeline shows the scanner versions central to the v13.1.0 production substrate. The v11.1 era produced the historical baseline used by the first three working papers in the corpus; v11.0 is shown as the falsified predecessor whose edge compression bug suppressed halt evidence by a factor of forty-two on a controlled G-SIB scan. The v11.1.1 attribution overlay introduced cryptographic provenance. The v12.1.1 production scanner passed capture-replay validation on an anonymised G-SIB with byte-identical input token counts across four observations. The v13.0.0 scanner was cleanly falsified on a pre-specified prediction; v13.1.0 retained the three confirmed interventions, replaced the falsified one with a minimalist counterpart, and produced confirmations across all six primary metrics. v13.2.0 is an incremental refinement applied during the founding cohort scan window.

Source · Stationary Sea Part 1, Sections 3-5 · Full version history preserved in scanner version registry with prompt text, scoring functions, release hashes Methodology · Stationary Sea Part 1 (Zenodo)
543 institutions · 12 sectors · 95,876 agents · 636,854 governance edges · substrate v13.1.0
Methodology grounded in The Stationary Sea (Part 1: Substrate Construction), Sections 3 through 5 (Scanner Lineage and Capture-Replay Validation).