Endgame — Atlas of `rejestry` (procedural editorial-commit instances)

D-023: Endgame — Atlas of `rejestry` (procedural editorial-commit instances)

Pick 11 from the 2026-05-21 vision verification session. D-013 §Endgame named the four-Spisy arc but never operationalised it as gameplay. D-016 ratified the tree math but left the 20%-gap milestone unnamed. D-022 §3 explicitly reserved the “atlas-as-endgame-gating” semantic for “the forthcoming D-023 atlas-endgame ratification” — this is that ratification. D-023 stitches D-013 + D-016 + D-022 into a single endgame surface: the Atlas of rejestry.

The rejestr semantic alignment is the load-bearing call. PoE’s atlas / map system is the right mechanical reference, but borrowing PoE’s terminology directly would be a lore violation — the data layer has already authored 10+ touches of rejestr as a metaphysical concept (Cech działa w dwóch rejestrach; Quest 004 literal title Połowa rejestru; D-013 calls itself a rejestr; D-016 §Player-facing surface frames Walker level AS strona N rejestru per D-014 IA conventions). The endgame surface continues that grammar rather than importing a competing one. Same engine, two registers — lore rejestr = editorial recordkeeping artifact, gameplay rejestr = playable editorial-commit instance.

1. Unlock = ~144 tree points spent

The Atlas opens when the Walker has spent ~144 tree points (per D-016 math: ~80% reach — saturation of the design-mandated reachable tree). NOT a hard gate at exactly 144: opening at 140-148 within a small envelope is mechanics-designer’s B-level tuning. The canonical anchor is “endgame opens when meaningful tree allocation is done, not earlier.”

D-016 already canonised the 50-pt milestone as the first writing-rights (Spis Wszechrzeczy access per D-013 §Endgame). D-023 names the endgame milestone as ~144 pts — the point at which the Walker stops reading the world (50 pts = becomes a Pisarz, reader-with-pen) and starts writing new chapters into it (~144 pts = becomes an author of new Spis-arc content).

2. Each `rejestr` = a playable editorial-commit instance

A rejestr is a procedurally-modified region-instance the Walker walks. Structure:

Field	Description
Base region	One of the 6 canonical regions (per D-006 Pillar 4) re-rendered as an instance variant
Modifiers	Per-rejestr modifiers (B-level pool authored by mechanics-designer) — leak-tier shift, faction interference, time pressure, weather, encounter density, etc.
Walking goal	Steps required to complete the rejestr — typically a fraction of a daily budget; the Walker physically walks through it
Combat encounters	D-020 `tropy` queue fires within the rejestr with elevated badge distribution
Reward	Higher-tier materials (above D-014 §3 weekly cap baseline), Spis writing-rights extension, possible new keystone unlock
Difficulty signature	Atlas-tier badge — TBA, derived from D-020 badge taxonomy as W-level extension

Walking + combat are the rejestr’s medium. A rejestr is not consumed by clicking; it is consumed by walking it with combat resolved along the way. This IS the literal “writing into Spis Wszechrzeczy” — the Walker’s footfalls + editorial commits (D-013) ARE the new chapter.

3. Atlas = meta-view of accessible `rejestry`

The Atlas (atlas; PL and EN identical — no transcreation needed) is the Walker’s meta-view of:

Currently open rejestry (instances unlocked but not yet completed)
Rejestr modifier composition (visible BEFORE commitment, per D-020 §4 ethos — the Walker chooses risk/reward at the Atlas selection screen)
Spis access map: which of the four Spisy (Spis Wszechrzeczy, Spis Czasów, Spis Słów, Spis Pustek per D-013 v2.0) the Walker has earned writing-rights into
Completed-rejestr archive (Spis page count, for player legacy view)

Atlas integration with the D-022 Map: open for ui-designer. Two viable shapes — (a) Atlas IS a zoomed-out diegetic layer over the D-022 Map (rejestr selections fly out from regions on the same map), (b) Atlas is a separate Home-accessed surface that opens INTO the Map widget for execution. D-023 does not lock either; ui-designer + mechanics-designer joint call. Both shapes preserve D-022 §2 nav posture (no new top-level tab).

4. Reward = materials AND writing-rights expansion

Two reward axes:

Material drops — higher-tier materials above the D-014 §3 weekly cap baseline. Specific drop math is B-level (mechanics-designer); D-023 only ratifies that rejestry drop above-baseline materials. Otherwise the endgame is engagement without progression and contradicts the compounding-scarcity curve D-018 anchored.
Spis writing-rights expansion — completing rejestry chains extends the Walker’s writing-rights from Spis Wszechrzeczy (50-pt milestone, D-013 §Endgame) to the other three Spisy. Specific gating (which Spis at which rejestr-completion threshold) is open D-level, deferred to a future ratification when v2.0 four-Spisy arc gets gameplay scope authored. D-023 ratifies the SHAPE (rejestry → writing-rights expansion); the specific thresholds are explicitly NOT locked.

5. Terminology disambiguation (D-017 carve-out for D-023)

rejestr / rejestry in D-023 (and downstream endgame surfaces) means playable editorial-commit instance — a procedural region-instance the Walker walks + fights to commit a new chapter to a Spis. This is the same metaphysical engine as the existing lore rejestr (the editorial register the Cech keeps, the rejestr Quest 004 tracks, the rejestr D-013 IS) — but distinct scope: lore rejestr = recordkeeping artifact (a thing the Walker reads / writes a line into / contributes to); gameplay rejestr = playable instance (a thing the Walker walks + fights to complete).

The rejestr glossary entry per D-017 four-phase pick MUST capture this scope-disambiguation explicitly in transcreationNote (same carve-out shape as D-019 §2 class-vs-rank, D-020 §5 gameplay-vs-process). Two registers, one engine.

atlas is a separate glossary entry — handle endgame.atlas or similar (narrative-designer four-phase pick). PL and EN identical; transcreation note records the Walker’s meta-view scope.

What D-023 does NOT decide (delegated)

Specific modifier pool for rejestry — B-level (mechanics-designer).
Material drop tier math (which materials, at what rates above weekly cap) — B-level (mechanics-designer + tech-architect anti-cheat sanity).
Atlas UI shape (zoomed-out map layer vs separate surface) — ui-designer + mechanics-designer joint.
Schema for Rejestr, RejestrModifier, AtlasState — W-level (tech-architect); likely coupled to D-022 MapState / MapPin.
Whether rejestry are time-gated (cooldown after completion), unlimited (instant re-run), or quota-based (N per week per Spis) — B-level (mechanics-designer); default assumption: weekly-cooldown to align with D-014 §3 harvest economy.
Whether different factions surface different rejestr trees (e.g. Cech rejestry vs Wolnokupiectwo rejestry) — B-level (mechanics-designer + narrative-designer).
Specific Spis-access thresholds (which Spis unlocks at which rejestr-completion count) — open D-level, deferred to a future ratification when v2.0 four-Spisy arc gets gameplay scope.
Co-op rejestry (whether multiplayer Walkers can run the same rejestr together) — deferred to D-025 multiplayer ratification.
Endgame keystones beyond the 6 leak-handler set (D-014 §2) — open. D-013 §Endgame line “Endgame keystones = Spis-access gateways” implies new keystones; D-023 does not author them.
Offline-rendering policy for Atlas state during D-009 §2 7-day offline cap — W-level (tech-architect + mobile-developer joint, parallel to D-022 §What-not-decide).

Rejected alternatives

“Atlas” without rejestr framing (pure PoE port). Rejected — fights existing D-013 / D-014 lore consistently using rejestr; ignores the 10+ quest / faction / NPC files already authored against the rejestr engine; loses the metaphysical alignment that makes the endgame a continuation of the early-game grammar rather than a genre-imported coda.
Procedural rejestry that ARE the existing entity-level rejestry (e.g. Quest 004’s Połowa rejestru upgrades into an endgame instance). Rejected — collapses authored quest content into endgame churn, prevents quest closure as a meaningful narrative event, double-loads quest 004’s specific dramatic weight.
No endgame instance system, just continued keystone unlocks past ~144 pts. Rejected — the 20% inaccessibility gap D-016 ratified is by design; there’s no canonical room for “keep allocating after 144.” Endgame must be a separate progression surface, not a tree extension.
Endgame unlock at exactly 50 pts (the D-013 first writing-rights milestone). Rejected — collapses mid-game milestone (Walker becomes a Pisarz, reader-with-pen) into endgame (Walker becomes an author of new chapters). The two milestones are deliberately layered; conflating them loses the progression arc.

Reasoning

D-023 is the operational closure of D-013 §Endgame v2.0. D-013 named the four-Spisy arc but never said how the Walker earns extended writing-rights; D-016 ratified the tree math but left the 20%-gap milestone unnamed; D-022 reserved the atlas semantic. D-023 stitches all four (D-013 + D-014 + D-016 + D-022) into a single endgame surface and aligns the gameplay terminology to the lore that’s already been authored.

The ~144-pt unlock threshold honors D-016’s design tension: the 20% inaccessibility gap exists so that finishing the tree is a real payoff. If endgame opened earlier, the rest of the tree would feel like an optional grind. ~144 pts means the Walker has earned the endgame by saturating the canonical early/mid game — the ~80% reach is exactly the full-game-completion marker that D-023 then converts into a new gameplay axis.

Forward-compat with D-025 (multiplayer, HARD-gated) is deliberate — D-023 leaves co-op rejestry open precisely so D-025 can answer it without re-litigating endgame shape. The Spis-threshold deferral (which Spis at which completion count) is the same kind of forward-compat: gameplay scope for Spis Czasów / Spis Słów / Spis Pustek hasn’t been authored, so locking thresholds now would be speculative.

Source: pick 11 from ops/memory/vision-qa-2026-05-21.md. Triage by game-director (2026-05-21 session). Pre-ratification recon per [[feedback-decide-triage-recon]] found rejestr deeply embedded in existing lore (10+ touchpoints across quests/factions/NPCs); rather than naming-collision, this is semantic alignment — D-023 explicitly inherits the existing rejestr engine and disambiguates the new gameplay scope from the existing recordkeeping scope (§5 carve-out).