Graduation gates
These eight conditions must close before this instrument moves from Keeper-curated statements only to open member submission. Until they all close, every statement on this site comes from an MP Pack, an Unmai desk, Magalir Avai, or the Case Organ — never from anonymous member submission.
All eight gates are open this era. No timeline promise. Methodology before promise.
- 1 · DPIA published
A UK GDPR Article 35 Data Protection Impact Assessment for the cookie-pseudonym vote path, the cohort fields, and the DP counter, published in full on this site.
Status · open - 2 · Independent privacy review
A named academic privacy reviewer (UK or EU) audits the DPIA and the ε budget calculation against the documented threat model.
Status · open - 3 · Math audit
An independent verification of the PCA + k-means + Laplace-DP implementation against the polis-community/red-dwarf reference. Pass criteria: clusters agree on a published vote corpus within stated tolerance.
Status · open - 4 · Anti-coordination defences
Rate-limit per cohort cell, statement-quality scoring, Keeper veto throughput sufficient to review every new member-submitted statement within 72 hours.
Status · open - 5 · Two-Keeper sign-off protocol
Every member-submitted statement requires sign-off from two unrelated Keepers before it appears for voting.
Status · open - 6 · Member proposal queue UX
A statement-submission surface that explicitly states the rules, the rejection criteria, and the cooling-off period before re-submission.
Status · open - 7 · Legal review
Documented review of the instrument against UK Terrorism Act 2000 §12, UK DPA 2018, and GDPR Article 9 considerations (no special-category data is ever collected; this is the legal proof of that).
Status · open - 8 · Academic mirror partnership
At least one independent academic mirror (Zenodo + a university partner) co-witnesses the closure of each era-week reading, so the reading exists outside this site's control.
Status · open
The gates mirror the pattern set by Unmai's graduation gates. Methodology, not promise.
The Belarus 2026 Coordination Council precedent is instructive but not adopted: their instrument was an election with declared winners. Karuthu Vellam is an opinion-mapping instrument with published readings. Different output, different gates.