jpmschweitzerandClaude Sonnet 4.6
7d7aec9cec
Generator spike sprint. 5 tickets across copy and server teams: - #609 zone identity spec (copy) - #610 Krenn culture profile (copy) - #611 NpcBlueprint struct design (server) - #612 Template assembly generator (server) - #613 NPC generation pipeline (server) Sprint proof: throwaway render — rural Krenn village from minimal input (zone type + culture profile, no per-location spec). Closed #586 (tile data model epic — child #594 done). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
101 lines
4.8 KiB
Markdown
101 lines
4.8 KiB
Markdown
# Sprint 25: Emerge — Joint
|
|
|
|
**Goal:** Prove the generator can extrapolate from minimal input — a rural Krenn village from zone type and culture profile alone, no per-location spec.
|
|
|
|
## Pre-Sprint
|
|
|
|
| Item | Owner | Blocks |
|
|
|------|-------|--------|
|
|
| Coordinate output format for zone identity spec (#609) | Miri + Tyre | #611 |
|
|
| Coordinate output format for culture profile (#610) | Miri + Tyre | #611, #612 |
|
|
| Confirm `NpcBlueprint` file location and module wiring | Tyre | #612, #613 |
|
|
|
|
The critical coordination question: does copy deliver YAML files (which the generator reads at runtime) or Markdown design documents (which Tyre reads and encodes into Rust)? Either works — decide at sprint start. YAML is preferred if Tyre can define a schema upfront; Markdown if the profile needs to be free-form first and codified later.
|
|
|
|
## Sprint Completion Proof — Throwaway Render
|
|
|
|
**This is not a test suite. It is an eyeball test.**
|
|
|
|
The proof is a single standalone invocation that feeds the generator minimal parameters and inspects the output for shape and coherence.
|
|
|
|
**Input (the minimum the generator must accept):**
|
|
```
|
|
zone_type: "rural"
|
|
region: "outside Sova urban zone"
|
|
system: "Krenn"
|
|
seed: 42
|
|
```
|
|
|
|
Plus the Krenn culture profile (#610) and zone identity taxonomy (#609) — loaded from disk, not hardcoded.
|
|
|
|
**NO location-specific spec.** No "the village has a tavern." No bespoke social site list. The generator extrapolates entirely from zone type + culture.
|
|
|
|
**Command (to be written as part of #612):**
|
|
```bash
|
|
cargo run --bin generator-spike -- --zone rural --seed 42
|
|
```
|
|
|
|
**Expected output (printed to stdout):**
|
|
1. `DistrictSkeleton` summary: zone type, block count, social site placements
|
|
2. Generated NPC list: one line per NPC with name, role, 2-3 traits, one observable behavior
|
|
3. A sample relationship pair (NPC A knows NPC B as X)
|
|
|
|
**The pass condition (human judgment, not automated):**
|
|
- Does the output have *shape*? A rural area should feel different from an industrial one.
|
|
- Does it feel like it *belongs in the Krenn System*? Not generic space-village. Krenn-inflected.
|
|
- Can you read the NPC relationships from the output? Not infer — *read*.
|
|
|
|
**The fail condition:**
|
|
- Output could be from any game in any setting (culture profile not doing work)
|
|
- All rural areas produce identical outputs with different names (zone type not doing work)
|
|
- NPC traits are unreadable from text (legibility fails — D-129 gate not met)
|
|
|
|
**Why this framing matters (from the approval discussion):**
|
|
The sprint proof deliberately uses minimal input because the generator's entire value proposition is scaling beyond hand-authored content. If it only works with detailed per-location specs, it fails the brief — because the whole point is the player walking into an unspecified area and it still having shape. The "throwaway render" framing is intentional: this is a rough proof, not a polished demo. Pass/fail is a 5-minute eyeball by the team lead.
|
|
|
|
## Tickets by Team
|
|
|
|
| Team | # | Title |
|
|
|------|---|-------|
|
|
| copy | #609 | Zone identity spec |
|
|
| copy | #610 | Krenn culture profile |
|
|
| server | #611 | NpcBlueprint struct design |
|
|
| server | #612 | Template assembly generator |
|
|
| server | #613 | NPC generation pipeline |
|
|
|
|
## Dependency Chain
|
|
|
|
```
|
|
copy #609 (zone spec) ──┐
|
|
├──> server #611 (NpcBlueprint) ──> #612 (template gen) ──> #613 (NPC pipeline)
|
|
copy #610 (culture) ───┘
|
|
```
|
|
|
|
Copy (#609, #610) runs in parallel and completes first. Server (#611 → #612 → #613) runs sequentially after copy delivers.
|
|
|
|
## Open Questions
|
|
|
|
| ID | Question | Blocking |
|
|
|----|----------|---------|
|
|
| Q-WTF-033 | AI templating: Claude API, local ollama, or manual for v0.2? | #613, #623 |
|
|
| 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 is the most relevant to this sprint. #613 does not need to implement AI templating — that is #623 (deferred). But the pipeline should be designed so AI templating slots in later without a rewrite. Discuss at sprint start.
|
|
|
|
## 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 authored beyond the zone spec and culture profile
|
|
- No AI templating implementation (design for it, don't build it)
|
|
- No playable session from the generated world
|