jpmschweitzerandClaude Sonnet 4.6
58d2e3b695
server.md: four warnings from Troblum — generate_npc() requires a live bevy World (stub routine generation in Phase 1), cultural text assembly is a new code path not a one-liner, DayPhase alias collision in generator.rs, schema negotiation takes rounds. joint.md: confidence 15% note at top. Intra-zone variance test added (rural seed 42 vs rural seed 43 — coherence within type, variance across seeds). Pass conditions restructured into three explicit comparisons: cross-type, intra-type, culture. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
4.7 KiB
Sprint 25: Emerge — Joint
Confidence: 15%. This sprint is exploratory — we learn whether the approach works, not whether we can ship it. Declare what we learned, not victory.
Goal: Prove the generator can extrapolate from minimal input — a rural Krenn area AND an industrial Krenn zone from zone type and culture profile alone, no per-location spec. Two zone types, one culture, side-by-side comparison.
Pre-Sprint
| Item | Owner | Blocks |
|---|---|---|
| Tyre shares example YAML schema from #611 with Miri | Tyre | #609, #610 |
| Agree on file paths for zone spec and culture YAML | Tyre + Miri | #609, #610, #612 |
The critical coordination handoff: #611 defines the Rust structs and writes example YAML showing the schema. That YAML goes to Miri immediately. Copy fills the schema with real content. Server builds Phase 1 of #612 with hardcoded stubs in parallel — does not wait for copy.
Sprint Completion Proof — Throwaway Render
This is not a test suite. It is an eyeball test.
Commands:
cargo run --bin generator-spike -- --zone rural --seed 42
cargo run --bin generator-spike -- --zone industrial --seed 42
cargo run --bin generator-spike -- --zone rural --seed 43
Expected output per invocation:
- Zone type and summary (block count, social site list)
- NPC roster: name, role, 2-3 traits, one observable behavior
- Relationship pairs: "A knows B as [type] ([valence])"
Pass condition — three comparisons:
Cross-type (rural vs industrial, seed 42):
- Different output shape — social site mix, NPC role distribution, density all differ
- Both feel Krenn — shared naming conventions and cultural markers
- Zone taxonomy does visible work (outputs distinguishable by zone type alone)
Intra-type (rural seed 42 vs rural seed 43):
- Both recognizably rural — same zone shape, similar role distribution
- Different people — different names, traits, relationship pairs
- Tests coherence within a type and variance across seeds
Culture (both zone types):
- Culture profile does visible work — output feels Krenn, not generic space-village
- NPC relationships are readable from text — not inferred, read
Fail condition:
- Both outputs look the same with different labels (zone taxonomy not doing work)
- Neither feels Krenn (culture profile not doing work)
- NPC traits unreadable from text (D-129 legibility gate not met)
This sprint discovers the right spec — it does not implement a known one. If the output is light and not fully deep, that is expected. The proof is in the pudding: run it, read it, judge it.
Tickets by Team
| Team | # | Title |
|---|---|---|
| server | #611 | NpcBlueprint struct design |
| copy | #609 | Zone identity spec |
| copy | #610 | Krenn culture profile |
| server | #612 | Template assembly generator (includes NPC pipeline + stdout output) |
Dependency Chain
#611 (structs + schema) ──> #609 (zone spec YAML) ──┐
──> #610 (culture YAML) ──┴──> #612 Phase 2 (integrate real YAML)
#611 ──────────────────────────────────────────────────> #612 Phase 1 (hardcoded stubs, build in parallel)
#611 first. #609 and #610 in parallel after schema arrives. #612 builds in two phases — Phase 1 with stubs runs alongside copy, Phase 2 integrates real YAML.
Open Questions
| ID | Question | Blocking |
|---|---|---|
| Q-WTF-033 | AI templating: Claude API, local ollama, or manual for v0.2? | #623 (deferred) |
| Q-WTF-039 | Character creation screen: portrait render or tile-scale preview? | Sprint 26+ |
| Q-WTF-040 | Do creation choices trace into the generated apartment? | Sprint 26+ |
Q-WTF-033 does not block this sprint. Design #612 so AI templating slots in later without a rewrite — the culture profile's speech markers are already the prompt constraint structure.
Deferred to Future Sprints
- Tycoon bookmark (#605, #614-617) — needs generator output proven first
- Character creation screen (#606, #618-620) — client work, post-generator
- NPC legibility systems (#607, #621-622) — downstream of generator
- AI content templating (#623) — Q-WTF-033 unresolved
- World feel systems (#608, #624-626) — downstream of everything
- Visual rendering of generated world — out of scope for this spike
What This Sprint Does NOT Do
- No client work (no rendering of generated content)
- No new content beyond the zone spec and culture profile YAML files
- No AI templating implementation (design for it, don't build it)
- No playable session from the generated world
- No ECS entity spawning into a running bevy world (#613 absorbed into #612 as stdout-only proof)