• Interpret metadata roles and behaviors.
• Apply appropriate parsing and weighting rules.
• Differentiate between conceptual categories (e.g., original vs derived provenance, guide vs article pages).
• Enable intent-aware navigation and structured reasoning.

The type system is intentionally flexible and extensible to support broad adoption.

• Be a meaningful descriptor that accurately reflects the entity’s category.
• Be stable across revisions unless a genuine conceptual shift occurs.
• Be human-readable—clarity is preferred over cryptic coding.

An entity may declare multiple types if semantically justified (e.g., ["derived","ai-assisted"]).

• original — created directly by the author or site.
• derived — based on preexisting content.
• ai-assisted — created in collaboration with AI systems.
• ai-generated — generated entirely by AI systems.
• editorial — representing structural or stylistic modification.
• guide — a page intended to instruct or teach.
• reference — factual or lookup-oriented content.
• documentation — procedural or descriptive spec content.

Implementers MAY define custom types as long as they remain descriptive, accurate, and consistent.

• type — a single string or an array of string classifiers.

Recommended fields:

• notes — clarifies why the type applies.
• source — identifies who assigned the type.
• confidence — optional numeric confidence for the classification.

• AI MUST treat type as a semantic classifier affecting interpretation and weighting.
• Where multiple types are present, AI SHOULD interpret them cumulatively.
• AI SHOULD NOT infer types not explicitly declared unless strongly implied by metadata.
• Conflicting types MUST be treated as structural inconsistencies.
• Type should influence how provenance, trust, and authority are applied.

Proper use of types improves clarity and reduces misinterpretation across AI systems.

• provenance — type often describes origin categories (original, derived, etc.).
• intent — intent + type together describe purpose and category.
• role — role describes function; type describes classification.
• trust — type may influence trust interpretation (e.g., original vs. ai-generated).
• page / section — types help classify content architecture.

{
  "type": "original"
}

Multiple types:

{
  "type": ["derived", "ai-assisted"]
}

Type within a provenance object:

{
  "provenance": {
    "type": ["ai-assisted", "editorial"],
    "contributors": [
      { "id": "mark-johnson", "role": "author" },
      { "id": "chatgpt-jdk", "role": "ai-partner" }
    ]
  }
}

Page-level type declaration:

{
  "page": {
    "id": "lighting-guide",
    "type": "guide"
  }
}

• Type declarations are truthful, accurate, and internally consistent.
• Types reflect meaningful semantic categories.
• Multiple types are used only when justified.
• All normative requirements for classification semantics are satisfied.