Design Spec — Indonesia: A Comprehensive History (textbook PDF)

Date: 2026-06-26 Status: Approved (design); ready for implementation planning Deliverable: A university-grade, expert-level history textbook (~1000 pp.) covering the Indonesian archipelago from deep prehistory to 2026, with real cited sources, real public-domain/CC images, original diagrams, and a print-quality PDF build.

1. Goal & audience

Produce a comprehensive survey that a motivated non-specialist can read cover-to-cover and come away genuinely conversant — “expert” in the sense of a well-read generalist, not a credentialed area specialist.

  • Audience: serious general reader / undergraduate survey level. Voice: clear, pedagogical, scholarly but readable. Define terms; explain debates; avoid jargon without glossing it.

  • Scope (decided): broad reading of “founding → today.” Covers prehistory → classical Hindu-Buddhist kingdoms → Islamic sultanates → European encounter & VOC → colonial state → national awakening → Japanese occupation & revolution → Sukarno → New Order → Reformasi → contemporary Indonesia (through the 2024 election / 2026), plus thematic syntheses.

  • Length target: ~1000 printed pages (~400,000–500,000 words) across ~50 chapters at ~18–25 pp. each. Built in phases; the project is stoppable at any phase milestone with a coherent partial book.

2. Build approach (reuse the existing pipeline)

Fork the proven docs/textbook/ pipeline into a new docs/indonesia-history/:

  • Playwright/Chromium offline HTML→PDF renderer (build.mjs), inlined Mermaid for diagrams.

  • Section-fragment HTML files + a sections.json manifest + assets/doc.css.

  • Auto-generated cover, table of contents (from manifest + <h2 id> subsections), and part dividers — already supported by the existing build.mjs.

Required pipeline adaptation — image support. The current style guide forbids external images. This book needs real imagery, so the forked pipeline MUST support embedded images while staying fully offline and self-contained. Approach:

  • Store sourced images under docs/indonesia-history/assets/img/.

  • Embed them in the PDF as base64 data-URIs (preferred — keeps the single output file self-contained and offline), or render Playwright from a file:// base so relative assets/img/... references resolve. Implementer picks one in Phase 0; base64 is the default recommendation.

  • Add CSS for figures-with-photos, captions, attribution lines, endnote/citation styling.

Output: docs/indonesia-history/indonesia-a-comprehensive-history.(html|pdf).

Rationale: proven, visually consistent with existing rmh docs, fully offline, already handles TOC/figures/PDF. Alternatives (Pandoc/LaTeX, React-PDF) rejected as higher-risk and stylistically inconsistent with the repo.

3. Citation & accuracy contract (non-negotiable)

  • Citation style: Chicago notes-bibliography. Inline superscript markers in the prose → a numbered endnotes list at the end of each chapter → a single consolidated master bibliography in back matter.

  • No citation from memory. Every cited claim must trace to a source the writing subagent actually fetched during its research pass (record the URL/DOI). Subagents use the deep-research skill / web search + fetch to ground claims before writing.

  • Adversarial verification pass. After a chapter is drafted, a separate verification step spot-checks a sample of citations for (a) the source existing and (b) actually supporting the claim. Any chapter with thin/unverifiable sourcing is flagged in its handback, not silently shipped.

  • Seed scholarship. A starter bibliography of standard works seeds research and is extended per chapter, e.g.: Ricklefs, A History of Modern Indonesia since c. 1200; Vickers, A History of Modern Indonesia; Reid, Southeast Asia in the Age of Commerce; Taylor, Indonesia: Peoples and Histories; Elson, The Idea of Indonesia; Cribb, Historical Atlas of Indonesia. (Treated as leads to verify, not as quotable-from-memory.)

  • Honesty on contested history. Where scholarship disagrees (e.g., extent of Majapahit, the 1965 killings, 1998, Papua), present the debate and attribute positions; do not assert one contested narrative as settled fact.

5. Coherence harness (cross-chapter consistency)

Because ~50 chapters are written by many subagents, a shared, append-only harness keeps the book consistent. Every writing subagent reads these first and appends to them:

  • conventions.md — transliteration/spelling rules (e.g., Srivijaya, Majapahit, Sukarno vs Soekarno, Yogyakarta), date/era conventions, periodization, naming of figures/places, citation formatting rules.

  • master-timeline.md — growing canonical chronology; chapters reconcile their dates here.

  • glossary-ledger.md — running glossary / names ledger feeding the back-matter glossary.

  • bibliography.md — seed + accumulating verified sources feeding the master bibliography.

This mirrors the established VerseCraft cohesion-harness pattern in this repo.

6. Chapter structure (per chapter)

Following an adapted version of the existing STYLE_GUIDE.md:

  1. Learning objectives box.

  2. Lead paragraph (<p class="lead">).

  3. Narrative body with <h2 id="kebab-id"> subsections (ids feed the TOC), <h3>/<h4> nesting.

  4. 2–5 figures (real images and/or Mermaid diagrams), each with a numbered <figcaption> + attribution where applicable.

  5. Key-term callouts where concepts are first introduced.

  6. Chapter summary box.

  7. Review & Discussion Questions (6–10) — lightweight, NOT the full exam/worked-solution apparatus of the rmh codebase textbook (decided: history survey doesn’t need exams; this keeps scope focused).

  8. Endnotes — numbered citations for the chapter.

7. Chapter outline (~50 chapters)

Front matter: Cover · Preface (how to use, periodization, transliteration) · Introduction: The Idea of “Indonesia.”

Part I — Land, Peoples, Deep Time

  1. The Archipelago: Geography, Geology, Volcanoes, Monsoons & the Sea (Wallace Line)

  2. Human Origins: Homo erectus (Sangiran/Trinil), Homo floresiensis (Liang Bua)

  3. The Austronesian Expansion: Language, Migration, Neolithic Societies, Megaliths, Dong Son

  4. Early Trade & the Coming of Indian Influence (the “Indianization” debate)

Part II — Classical Hindu-Buddhist Era (4th–15th c.) 5. First Kingdoms: Kutai, Tarumanagara & the Earliest Inscriptions 6. Srivijaya: The Maritime Empire of Sumatra 7. Central Java’s Golden Age: Sailendra, Mataram, Borobudur & Prambanan 8. Eastern Java: Kediri & Singhasari 9. Majapahit: The Great Archipelagic Empire (Gajah Mada, Nagarakretagama) 10. Bali & Cultural-Religious Continuity

Part III — Islam & the Sultanates (13th–17th c.) 11. The Coming of Islam (Sufism, trade, Pasai, the Wali Songo) 12. Malacca & the Maritime Muslim World 13. The Sultanates: Aceh, Banten, Demak, Islamic Mataram, Makassar, Ternate/Tidore 14. The Spice Trade & the Early Global Economy

Part IV — Encounter & the VOC (16th–18th c.) 15. The Portuguese & First European Intrusion 16. The Dutch VOC: A Company-State (Batavia, Coen, monopoly) 17. VOC Expansion, the Banda Massacre, Wars & Decline (bankruptcy 1799)

Part V — The Colonial State (19th c.) 18. From Company to Crown: The Netherlands East Indies 19. The Cultivation System (Cultuurstelsel) & Its Consequences 20. The Java War, Padri War & Regional Resistance 21. The Aceh War & the Completion of Conquest 22. Liberal Era, Plantation Economy & the Ethical Policy 23. Colonial Society: Race, Class & Daily Life

Part VI — National Awakening (1900–1942) 24. Origins of Nationalism (Budi Utomo, Sarekat Islam, education) 25. Ideologies in Contest: Islam, Communism (PKI), Secular Nationalism 26. Sukarno, Hatta & the Movement; Sumpah Pemuda 1928

Part VII — Occupation, Revolution & Sukarno (1942–1965) 27. The Japanese Occupation 1942–45 28. Proclamation & the National Revolution 1945–49 29. Parliamentary Democracy & Its Crises 1950–57 30. Guided Democracy, Konfrontasi & Regional Rebellions

Part VIII — The New Order (1965–1998) 31. 1965: The Coup, the Mass Killings & Suharto’s Rise 32. The New Order State: Developmentalism, Pancasila, Dwifungsi 33. Economic Transformation & Its Discontents 34. East Timor, Aceh & Papua: Integration & Insurgency 35. Society, Religion & Culture under the New Order

Part IX — Reformasi & Contemporary Indonesia (1998–2026) 36. The 1997–98 Crisis & the Fall of Suharto 37. Democratization, Decentralization & Reformasi 38. Reform-Era Presidents: Habibie, Wahid, Megawati, SBY 39. Jokowi, Prabowo & the 2024 Election 40. Contemporary Economy, Infrastructure & the New Capital (Nusantara) 41. Religion, Islam & Pluralism Today 42. Separatism, Human Rights & Unresolved Pasts 43. Indonesia in the World: ASEAN, China & Geopolitics

Part X — Thematic Syntheses (span all eras) 44. Cultural History: Language, Literature, Wayang, Batik, Architecture 45. Religious History across the Archipelago 46. Economic History: From Spices to Palm Oil & Nickel 47. The Chinese Indonesians & Other Diasporas 48. Women, Gender & Family Through Indonesian History 49. Environmental History: Forests, Volcanoes & Climate 50. Historiography: How We Know — Sources, Debates & Contested Pasts

Back matter: Master Chronology · Glossary & Transliteration Guide · Dynastic/Presidential Tables · Maps Appendix · Consolidated Bibliography · Image Credits & Licenses · Key-Terms Index.

8. Phasing (built in waves; stop at any milestone)

  • Phase 0 — Scaffolding (done directly, not via subagents): fork pipeline to docs/indonesia-history/; implement image support; write conventions.md, seed bibliography.md, master-timeline.md, glossary-ledger.md; author cover + preface + introduction; extend doc.css (figures/photos/attribution/endnotes); set up sections.json.

  • Phase 1 — Parts I–II (ch 1–10): prehistory → classical kingdoms.

  • Phase 2 — Parts III–V (ch 11–23): Islam → colonialism.

  • Phase 3 — Parts VI–VII (ch 24–30): nationalism → Sukarno.

  • Phase 4 — Parts VIII–IX (ch 31–43): New Order → contemporary.

  • Phase 5 — Part X + back matter (ch 44–50 + bibliography, glossary, maps, timeline, credits, index).

Per-chapter subagent loop (every chapter): deep web research + source-fetch → record verified citations → source PD/CC images (license-checked) → write the HTML fragment per §6 → append to the shared ledger/timeline/bibliography → adversarial citation-verification spot-check → handback noting any thin sourcing.

Each phase is a wave of parallel chapter-writing subagents, reviewed before moving to the next phase.

9. Risks & mitigations

  • Scale/cost: ~50 research-heavy chapters = many large subagent runs over multiple sessions and significant token spend. Mitigation: phasing; each phase delivers a coherent partial book; stoppable at any milestone.

  • Citation integrity (top risk): subagents can fabricate plausible-but-fake citations. Mitigation: fetch-then-cite contract (record URLs/DOIs), adversarial verification pass, explicit flagging of thin sourcing. Spot-checks won’t catch everything — residual risk acknowledged.

  • Cross-chapter inconsistency (spellings, dates, overlapping coverage): Mitigation: the coherence harness (§5) read/appended by every subagent.

  • Image licensing: Mitigation: Commons-only, license-verified, full credits appendix.

  • Contested history framed as fact: Mitigation: attribute positions, present debates (§3).

10. Decisions captured

  • Scope: prehistory → 2026 (broad reading). ✓

  • Citations: Chicago notes-bibliography, fetch-then-cite, verified. ✓

  • Images: Wikimedia Commons PD/CC only, downloaded + attributed. ✓

  • Length: target ~1000 pp., phased. ✓

  • Format: fork existing docs/textbook/ Playwright PDF pipeline + add image support. ✓

  • Exercises: lightweight Review & Discussion Questions only; no exams/worked-solution appendices. ✓