Skip to content

Minimal Projection Schema, Runtime Fallback, and Math Pilot

Generated artifact. Do not edit manually.

Generated by: scripts/build_compendia.py Regenerate with: python3 scripts/build_compendia.py --compendium dev-handbook --web-and-tex Source of truth: docs/dev/source/

Language / Sprache: English (EN) · Deutsch (DE) | PDF: English PDF · Deutsches PDF

Compendium: §1 Documentation Guidelines · §2 Landscape JSON Reference · §3 Minimal Schema & Fallback · §4 Authoring Editors · §5 Graph Editor · §6 Flashcard Editor · §7 Node Ordering · §8 Liquibase · §9 Auto-Update (SSE)

3. Minimal Projection Schema, Runtime Fallback, and Math Pilot

To evolve existing landscapes into the layered architecture (General Goal System) without breaking unmigrated curricula, SkillPilot introduces four optional additive JSON structures paired with deterministic runtime fallback rules.

3.1. Additive Schema Fields and Fallback Precedence

{
  "programUnits": [
    { "id": "sek2-e", "kind": "phase", "label": "E-Phase", "order": 10 }
  ],
  "goalPlacements": [
    {
      "goalId": "<uuid>",
      "unitId": "sek2-e",
      "relation": "primary",
      "context": { "jurisdiction": "DE-HE", "stage": "SekII" }
    }
  ],
  "competencyCatalog": [
    { "id": "K1", "label": "Mathematisch argumentieren", "dimension": "process" }
  ]
}

Goals additionally carry optional competencyRefs: ["K1", ...]. Runtime precedence follows strict non-destructive rules:

  1. Composition view first: If a reviewed .view.json matches the resolved learner scope, compile the default tree from that composition view plus canonical contains.
  2. Explicit placement projection second: Otherwise, when programUnits and matching primary goalPlacements exist, project the program tree directly from those arrays.
  3. Legacy contains / phase fallback third: Landscapes without placement or view artifacts continue rendering their authored contains hierarchy and dimensionTags.phase badges unchanged.
  4. Unchanged mastery & frontier: Mastery storage and requires frontier calculation always operate on canonical atomic LearningGoal IDs, never on programUnits or view structure nodes.

3.2. Entry-Scope Resolution and Canonical Mathematics Pilot

Entry-scope resolution executes in five deterministic steps: (1) select base root curriculum, (2) bind jurisdiction / duration / stage / course profile (Level 2), (3) resolve matching .view.json or placement context, (4) close required atomic prerequisites \(L = \mathrm{Closure}(S)\) on the unpruned canonical graph, and (5) apply optional Level 3 focus filters. curricula/DE/Gymnasium/canonical/DE_DEU_S_GYM_CANONICAL_MATHEMATIK.de.json serves as the reference pilot proving additive programUnits, goalPlacements, competencyCatalog, and state-scoped .view.json compilation side by side.