Skip to main content

uniteTOlove.ca

Home / Working notes / The tri-seat assessment

Working notes — an internal document, published nearly verbatim. Expect apparatus: decision codes, board references, first-person seat voice. Why we publish our working documents: how we share.

TRI-SEAT ASSESSMENT — Founder · Mobilization · Research, one sitting, 2026-08-07

Cowork (Pro-5x seat), on the operator's omnibus (banked verbatim, OPERATOR_LOG 2026-08-07). Companion surfaces: the walk is FIRE_LIST_2026-08-07.md; the mobilization picture is 2026-08-07_STATE_OF_THE_MISSION.md (same day, first sitting). PC3 is the backbone throughout: start small, prove it, multiply (P15) · nested centers coordinate, never absorb (P23) · as fast as possible, as slow as necessary · design for the ones who inherit. Registry facts cite receipts.

§1 · The three plates, assessed

FOUNDER. The judgment seat's real backlog is small and specific: (1) a PROPOSED-mint ratification batch — a recorded standing decision (human-interface floor) · a recorded standing decision (atlas language dimension) · a recorded standing decision/158/159/160 (trustee 76.7% · kit publish-ready · PII sign-off · seal consolidation — all four already executed under the operator's same-day direct words; ratification is bookkeeping, one screen); (2) QBank dual-instrument reconciliation (the 66Q/12-category v0.1 vs the ~300-Q v2.1 — which wins is undecided, flagged since S3); (3) ward-profile scheme consolidation (one of this library's own project records vs the elections-lane build); (4) renders mirror-drift ruling (CURATOR 08-06 flag [1]; a recorded standing decision still PROPOSED); (5) KEY_PROJECTS §M is stale (scenario "S5 owed" — sealed; harvest "deploy-dark" — live; publish train "not live until S1" — live; OBC "Oct 12–18" — Oct 7–14 ratified): one CURATOR truing paste. Items 1–4 are folded into the CORPUS_REVIVAL boot's Founder screens; item 5 rides any curator lane.

MOBILIZATION. Current as of this morning's review sitting — the full waits-on-the-operator list is STATE_OF_THE_MISSION §3. Unchanged headline: machinery built, zero sends fired, Polis clicks by [send scheduling withheld], huylieu.ca call open (runbook §2.9). The omnibus adds four new mobilization-side projects (one of this library's own project records/0149 + the surface/usability program one of this library's own project records), all staged behind the operator's words.

RESEARCH. The plate the estate has quietly under-tended while the mission layer shipped: (1) the a later round research-question batches are still STAGED, NOT FIRED (batch plan: Q3/Q4/Q5 → Q7/Q8/Q9 → Q10/Q11/Q12, gated on the F14 WIP cap — T+H occupy 2 of 3 slots until their tails close); (2) the T/H tails themselves trace to the operator's two items from the 2026-07-31 oversight sitting — the H mint paste (handed 2026-07-30) and T's rung-claim mint ride the same single-window session; no record shows either fired since — closing them is what unlocks the whole Q-wave; (3) the 499-row verification backlog (468 VERIFY / 31 DISPUTED, ≈6–8 lane-days) awaits burn; (4) Issues Corpus W2 is frozen on the Kimi twin (his return) → Founder adjudication; (5) the estate-wide seal_translate prose-sweep debt (internal notation on live research pages — exactly the operator's surface-review ask, now one of this library's own project records); (6) the renderer heading-anchor gap named in NOW_RESEARCH item 18 was BUILT 2026-08-06 (wiki go-live pass) — the open-items registry entry is stale, as is its EPTC/EPRC merge entry (#7, executed same day); one truing line each, folded into CORPUS_REVIVAL's reconcile step. (7) The 20-program bodies stay parked on the operator's own word ("pending once everything else gets published — I am still excited for this") — honored, not resurfaced; noted that CORPUS_REVIVAL's object map should show them as designed-dormant.

§2 · What we need to do (the one-paragraph verdict)

Two motions, run in parallel. Motion A — finish the launch: the operator's minutes list + the send sitting ([send scheduling withheld]) + the voice batch + the two Ontario publish words; nothing new to build. Motion B — turn the estate from a heroic build into a system that others can run: the six new projects. Their shared spine is PC3: the surface gauntlet and publishing systems make our modules provable; the methods codex and open strategy make them replicable (spore kits, not franchises — nested centers coordinate, never absorb); Paper & Signal equips the crowd's own hands; Corpus Revival re-points the whole research estate at its election-season job (candidates/teams · journalists · civil society) and re-arms the question waves behind it. Fire order and what the operator decides in each: FIRE_LIST_2026-08-07.

§3 · What you haven't thought about (each with a proposed home)

  1. The weekly scoreboard ritual. The canon names the numbers (signatures · rooms we didn't staff · org pins · first self-convened room) but nobody produces them weekly. One standing task + a dated one-pager every Monday. → Publishing-systems boot, deliverable 6.
  2. Inbound will outgrow you within days of the first sends. invite-me, civic-hour, press, corrections, TPL, counsel replies — no triage ritual, no response-time promises, no rapid-response protocol for a hostile media cycle. → new §7 of the open-strategy boot (comms half) + a daily inbox ritual the operator approves once.
  3. Baseline before sends. The analytics token (the operator's §2.5) matters this week specifically: without a quiet-week baseline we can never show the campaign moved anything. Token first, then sends. → the operator's minutes list, re-ordered.
  4. Accessibility and language. The skip-link CSS gap is known; nobody has run a real WCAG/AODA pass, and the 14 language cards recruit translators without a landing path for what they translate first. → usability-audit boot, act 3.
  5. The bench. October needs named humans: synthesis owner (the operator's "decide later"), second Polis moderator, correction triagers. PC3 P20 — guild your people. A one-page BENCH roster object with empty seats visible. → open-strategy boot, deliverable 5.
  6. The post-election instrument. The adoption record needs its successor on Nov 1: the promise tracker was tombstoned; its revival as the adoption-record tracker (term accountability) should be designed before Oct, not after. → CORPUS_REVIVAL object map.
  7. Advance-vote logistics push. In the last two weeks, voting logistics beat persuasion. Turnout v2 (advance dates, ID, proxy) + shareables, armed [send scheduling withheld]. → publish-train backlog row, dated.
  8. Continuity/bus-factor, now urgent. The operator is custodian, sole sender, and sole approver in the highest-tempo month. WP-SEC/our internal work-package record's named-second question stops being theoretical; also a 72-hour "operator offline" runbook. → one screen in the S4-class governance queue; counsel package unaffected.
  9. Our own archives. Wayback sweep of our published pages is staged, unfired; plus export drills for the two D1s against the operator's real key (named Code follow-up from a recorded standing decision). → publishing-systems boot, deliverable 7.
  10. Handles and impersonation doctrine. Reserve unitetolove handles on the platforms the census names before Paper & Signal makes the mark spread; and decide the impersonation response under the identity firewall. → PAPER_AND_SIGNAL boot, act 1 (+1 counsel question).
  11. Demand capture for other jurisdictions. The moment ONT-SOTE publishes, out-of-Toronto readers will ask "do my city." The methods codex needs at least a front-door stub live before that word fires — sequence dependency, cheap to honor. → METHODS_CODEX boot, deliverable 1; noted on the ONT-SOTE menu row.
  12. The counsel package should absorb three new questions before it fires ([send scheduling withheld]): third-party-advertising registration thresholds for print + social at scale (Paper & Signal's whole legal frame) · CASL boundaries for any list mail · the Election Finances Act item already parked at queue §1b (75%-of-Canada expansion touches provincial/federal regimes). → one-line rider on the send bundle; do not delay the send for new research.
  13. The federation schema. If Barrie or Hamilton builds a knowledge-base, interop is a protocol decision (IDs, slugs, CSV schemas, correction flow) best made before the first copy exists — PC3 P23, sporing as protocol, not fork. → METHODS_CODEX deliverable 4.
  14. Your own pace is a design surface. The scenario designed December rest for the city; nothing designs the operator's August–October. The fire list now schedules weeks, not just tasks — and "as fast as possible, as slow as necessary" binds the operator too.

§4 · Checks

Prose-with-pointers; no CSVs touched by this file; leak/vault greps + linkify at the sitting's close (SESSIONS line carries results). Every send stays operator-only; every publish click stays the operator's; nothing in this assessment fires anything.