Sprint goal: close Phase 3 Atlas (unified nav chain, brand corps, content refinement) and establish Phase 4 foundations (bookmark system, location-culture resolution, character creation skeleton). 15 tickets assigned across server (7), client (5), copy (3). Briefings written for all four teams. DB backup updated. Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
5.2 KiB
5.2 KiB
Sprint 36: Forge — Copy Tasks
Goal: Close Phase 3 Atlas (unified nav chain, brand corps, content refinement) and establish Phase 4 foundations (bookmark system, location-culture resolution, character creation skeleton).
Branch: sprint-36/copy
Agents: Mellanie (author), Paula (narrative), Gestalt (systems)
New Tickets
| # | Title | Blocked by |
|---|---|---|
| #849 | Core-world atlas cohesion pass — hand-refine high-visibility systems | — |
| #838 | Review and hand-refine generated atlas content across all inhabited bodies | — |
| #828 | Author 120-170 notable brand corps across all 8 categories | — |
Use tooling/db/ticket show <id> for full details.
Key Decisions
decisions/economics.md— D-189 (brand layer architecture — 8 categories), D-190 (brand volume calibration), D-185 (brands are not commodities), D-182 (TOML source of truth)decisions/architecture.md— D-191 (atlas pipeline, markers.json pixel-space format, cultural corridor naming palettes)decisions/content.md— D-128 (culture implicit in starting location — corridor identity matters for brand names and atlas names)
Notes
#849 — Core-world atlas cohesion pass — hand-refine high-visibility systems
- The high-traffic systems in the Reach (Gateway/Sirius, Groombridge/Lendel, Van Maanen's Star, Lalande, etc.) were generated by the Sprint 35 Gemma pipeline. These are the systems players will see most — they need white-glove treatment above the baseline #838 pass.
- Deliverable: edited
markers.jsonfiles for each priority system. Files live undertooling/planet-gen/generate/(path is<system_id>/<body_id>/markers.json). - D-191 §8: markers.json is in pixel-space
[row, col]format against a512 × 256grid. City positions, road polylines, and named features all use this format. Do not convert to lat/lon — that is a display-time derivation. - Corridor naming palettes per D-191: north_reach (Anglo-Saxon), south_reach (Iberian/Portuguese), east_reach (East Asian), west_reach (Germanic/Nordic), inner_orbit (institutional Latin/Anglo). Cohesion pass should check that names match the corridor palette and that no Earth-echoes slip through the blocklist.
- This is a child of #838 and should share notes on quality standards.
#838 — Review and hand-refine generated atlas content across all inhabited bodies
- Baseline refinement pass across all 273 inhabited bodies. Blockers (#832, #833) are already done from Sprint 35.
- Check city placements: not in water (ocean/river cells in the heightmap), not on steep mountain slopes, reasonable quadrant distribution.
- Check names: culturally appropriate for corridor, no Earth IP echoes, no dedup collisions.
- Check infrastructure: road and rail paths should follow terrain logic — rivers and valleys are cheap corridors, mountains are expensive. Paths that cross mountains without a pass are a generation artifact.
- Deliverable: corrected
markers.jsonfiles committed totooling/planet-gen/generate/. Document any systematic issues found (patterns the generator repeats) in a refinement notes file so the generator can be patched.
#828 — Author 120-170 notable brand corps across all 8 categories
- Source of truth format: TOML entries for
content/brands/brands.toml(or equivalent — check with server team on exact file location used by #829 generate_brands pipeline). - D-189: 8 categories with census targets:
- terroir: 15-20 (planetary origin goods — food, wine, materials)
- heritage_craft: 10-15 (old-world craftsmanship — tools, fabric, instruments)
- tech_premium: 20-25 (consumer electronics, implants, vehicles)
- cultural: 15-20 (entertainment, fashion, media)
- service_premium: 10-15 (hospitality, professional services)
- commodity_branded: 15-20 (branded commodity goods — fuel, water, staples)
- design_heritage: 10-15 (luxury goods, art, high design)
- institutional: 10-15 (education, healthcare, finance brands)
- D-190: these are notable brands, not minor ones. They are the top-of-mind touchstones in the Reach — the ones players will encounter in dialogue, on signage, in product descriptions. Minor brands (~10K) are generated by #829.
- D-185: brand corps are commodity demand nodes — they consume raw/intermediate goods from the economy simulation. Brand entries should specify their primary inputs (which commodity categories they consume).
- D-182: TOML is the source of truth. Each brand entry needs: name, category, founding_system (or corridor), price_tier, tagline (optional but useful for copy team), primary_inputs (list of commodity categories).
- Coordinate with server team (#829) to confirm the exact TOML schema before starting bulk authoring.
Dependency Chain
#849 (core-world cohesion) — standalone, start immediately
#838 (baseline refine pass) — standalone, start immediately; #849 is a deeper pass on the same files
#828 (brand corps authoring) — standalone; coordinate with server on schema before bulk writing
PR Workflow
tea pr create --repo jpmschweitzer/settled-reach --login schweitz \
--title "content(atlas): sprint 36 — atlas refinement and brand corps" \
--description "body" \
--base main --head sprint-36/copy