Learning Paths

Two structured tracks through the documentation. Pick the one that matches your goal.

Track 1: Narrative AI Developer

Build a narrative sifting and scoring pipeline for games or simulations.

Total time: ~2.5 hours

#	Page	Time	What you learn
1	Getting Started	10 min	Build and evaluate your first pattern in batch and incremental mode.
2	What is Sifting	10 min	The concept of sifting and why it matters across domains.
3	Patterns from First Principles	15 min	Mental model: stages, clauses, variables, joins, negation.
4	Sifting by Example	15 min	The same pattern across 4 domains, with interactive playgrounds.
5	Pattern Cookbook	20 min	Worked recipes for common pattern types.
6	Scoring and Surprise	15 min	Information theory for ranking matches by rarity.
7	Incremental Integration	15 min	Wire fabula into a simulation loop.
8	Narrative Quality	15 min	Threads, tension, pivots, and composite scoring for MCTS.
9	Scoring Matches	15 min	Full observe → score → rank workflow.
10	Forking for MCTS	15 min	Clone the engine, speculate, score, select.

After completing this track, you can detect narrative patterns in a running simulation, rank them by surprise, and use scoring to drive an AI director.

Track 2: Engine Integrator

Integrate fabula into an existing system as a temporal pattern detection engine.

Total time: ~90 minutes

#	Page	Time	What you learn
1	Getting Started	10 min	Build and evaluate your first pattern in batch and incremental mode.
2	Sifting by Example	10 min	The same pattern concept across 4 domains.
3	How the Engine Works	15 min	The 4-phase incremental algorithm, forking, deduplication.
4	Incremental Integration	15 min	Wire fabula into your event loop.
5	DSL in Rust	10 min	Parse, compile, and evaluate DSL patterns in Rust.
6	Debugging Patterns	10 min	Troubleshoot unmatched patterns with gap analysis.
7	Custom Adapter	15 min	Implement DataSource for your graph store.
8	Design Decisions	10 min	Why fabula is built the way it is — tradeoffs and alternatives.

After completing this track, you can integrate fabula into any event-producing system, define patterns in Rust or DSL, debug failures, and write a custom graph adapter.

Choosing a track

Pick Track 1 if you're building a game, simulation, or interactive narrative system and want to detect and score narrative events.

Pick Track 2 if you're integrating fabula into an existing system (monitoring, compliance, process mining) and need to understand the engine's behavior and integration points.

Both tracks start with Getting Started. You can switch tracks at any point -- the pages are self-contained.