Chapter 8 of 13 · 3 min read
The first honest checkpoint
Why the v0.1 system was useful even while most person-level evidence remained incomplete.
The Search for 100 Thinkers · Published 1.0 · 2026-09-22 · Moses Sam Paul
The first honest checkpoint was not the finished hundred. It was the moment the system learned to say what it did not know.
IoV100 v0.1 represented all 760 candidates, preserved inherited leads, made blockers explicit and generated reproducible exports. Most candidates had not received a complete evidence review. That distinction mattered more than the comfort of declaring the spreadsheet done.
The table below records the later v1.0 state so readers can see how far the evidence process moved after that first correction.
| Layer | Recorded state | What it does not establish |
|---|---|---|
| Discovery universe | 760 candidates | Complete person-level verification |
| Research ontology | 22 capabilities; 30 questions | Verified coverage of every capability or question |
| IoV100 v1.0 portfolio | 100 evidence-audited candidates | A ranking, accepted invitations or endorsements |
| Priority set | 25 provisional candidates | A panel of advisers |
| Inherited eligibility | 53 Advance; 16 Hold; 691 Pending | Fresh independent eligibility judgements |
| Minimum Stage 4A packets | 112 across the universe; 100/100 in the revised portfolio | Publication-grade deep verification |
| Full Stage 4B evidence dossiers | 100/100 portfolio members, including 25/25 Priority candidates | Empirical validation of an IoV construct |
| Comparative challenge | 71/71 challenged seats reviewed | A final ranking |
| Controlled revisions | 4 evidence-comparable changes in total; portfolio remains 100 | Participation or endorsement |
| Inspected evidence coverage | 22/22 capabilities; 30/30 questions | Adequate depth, independence or valid numeric scores |
| Contact addendum | 56 work-email fields; 93 profile sources; 7 blocked | Complete outreach readiness or intellectual merit |
| Contact repair | 33 fields checked; 17 withdrawn | Independent reinspection of every legacy field |
Read the validation label carefully
The workflow validates records and the imported checksum inventory. The contact audit checks provenance within its stated scope. Neither validator can prove that the scientific argument is correct or that a hundred people are intellectually independent.
The five early calibration records remain scoped reviews. They should not be promoted by a change of prose. The portfolio now has a minimum packet and a full three-artifact Stage 4B evidence dossier for every seat. The sole remaining artifact blocker was resolved through a controlled, evidence-comparable replacement: Ellis Ballard entered the portfolio and Vipasha Goyal moved to the reserve queue without an eligibility rejection.
The final Empty-Chair Test found no unexplained seat. Outside-lineage scrutiny is recorded for 100/100 dossiers. Eight implementation and adversarial tests passed against the canonical synthetic casebook. Those tests examine implementation invariants; they are not observed empirical validation of wellbeing, value or causality. Intellectual independence is still not inferred from an absent coauthor link.
IoV100 v1.0 is an evidence-audited, unranked intellectual portfolio. It does not imply participation or endorsement. Empirical validation of IoV remains open.
That is a useful checkpoint. It is not a victory lap. At this stage the confetti would mostly obstruct the audit trail.