Roadmap

This page is the public version of the project’s translation prioritisation. It reconciles three pressures:

• Strategic value — exposure, public interest, philological legitimacy.
• Production state — what is signed off, what is in flight, what is queued.
• Reviewer bandwidth — the project’s single human reviewer bottleneck.

The prioritisation favours finishing what is started over starting new work, and completing a lens-relevant cluster over breadth for its own sake.

This is the roadmap — the ordering and the next picks. The acquisition strategy that determines what enters the queue at all (traditions, source-licence posture, Edition vs Translation track) lives at Library Acquisition Program.

Production state (2026-08)#

28 books shipped or in flight. Every drafted chapter is signed off as stable. Distribution by tradition:

Production glossary. v2.73.0, 791 terms, six source languages (Hebrew, Sumerian, Akkadian, Ugaritic, Arabic, Greek). Distribution: 710 direct / 74 inferred / 7 speculative. The speculative count being deliberately small — at fewer than one percent of entries — is itself a discipline statement. Notably, the entire 1042-line Theogony shipped without adding a single term to the central glossary: all Greek-side decisions live in the book’s overlay (42 entries, 0 speculative), with cross-corpus convergences recorded as labeled comparative apparatus rather than fused refIds.

Per-translation overlay glossaries. Active for 26 books (every shipped translation except Genesis-WoH and the Mormon Edition track). The mechanism is described in Glossary system.

Multi-language fanout. Operational; runs after a book reaches stable. See Translations.

Next picks (2026-08)#

After Genesis-WoH closed out (the original “Phase 0”) and the pipeline proved it generalises across Hebrew, Greek, Akkadian, Sumerian, Arabic, Ge’ez, and Ugaritic, the phase-gated language of earlier roadmaps no longer fits production reality. The active model is named-pick + verified queue, not numbered phases.

Shipped since the 2026-06 picks: both prior top picks closed out — Isaiah’s initial scope {6, 24–27, 53} and Atrahasis I–III — plus the Sumerian King List (completing the Mesopotamian pilot pair) and Hesiod’s Theogony, the program’s first classical-Greek book, now a complete composition (1022 lines + the 929a–t variant passage, translated and signed off in four sequential blocks, 2026-08).

Top two picks#

1. Gospel of Thomas — opens the Coptic pilot. Source layer is already acquired (gospel-of-thomas-woh, CC BY via Coptic Scriptorium); logion structure keeps chapters small and the sign-off cadence fast; heavy canon engagement (logia 18, 19, 50, 84–85 on origins, images, and the “place of light”).
2. Vidēvdād Fargard 2 — opens the Avestan pilot (vendidad-woh, Geldner PD source already staged). Yima’s vara — the strongest engineered-survival narrative outside the flood cluster, and the bridge into the Persian material the comparative corpus keeps citing.

The 2026-06 acquisition queue#

72 further books are registered in catalog.json with status planned. Each carries a priority field (5 = pilot wave / cluster completion, 6 = build-out, 7 = long tail), an originalLang, and a versionSource recording the openly-licensed source edition the translation will be built from. The selection logic and the per-tradition acquisition tables are at the Library Acquisition Program.

The pilot wave by tradition:

Each pilot proves a language and unlocks build-out within its tradition. Each build-out is gated on the pilot’s sign-off.

Cluster completion within already-shipped traditions#

The corpus benefits more from closing existing clusters than from opening new ones the project lacks the bandwidth to support. Each shipped tradition has a small completion shelf:

• Mesopotamian: Etana, Enmerkar and Aratta, Enuma Elish (upgrade from Edition stub), Descent of Inanna / Descent of Ishtar, Anzu, Erra and Ishum, Instructions of Shuruppak.
• Levantine/Ugaritic: rest of the Baal Cycle (KTU 1.1–1.2, 1.5–1.6), Aqhat, Rephaim texts, Kirta, Shachar and Shalim.
• Hebrew Bible: Job 39 (closes the planned {1–2, 38–42} scope — recommended quick-win); Exodus extension toward 1–20.
• New Testament: Revelation 5 if the throne-vision-sequence hole in {1, 4, 6} is unintentional (editor’s call).
• Pseudepigrapha: more Jubilees; Enoch Parables 37–71 (deferred — Ge’ez competence risk).

Resource reality and what would change the plan#

The phase ordering assumes a single named human reviewer (currently Zara Zinsfuss) with finite throughput. If reviewer bandwidth expands — adding a named second human reviewer with source-language competence — multiple pilots can run in parallel and the queue loosens substantially.

If a credentialed scholar of a specific tradition joins the project, that tradition can move earlier in the queue. A named Bahá’í scholar with Persian competence could front-load Kitáb-i-Íqán; a named Akkadianist could parallelise Atrahasis and Enuma Elish.

If a multi-model drafting pipeline becomes operational (Claude + GPT in parallel, with the Editor receiving both drafts), the Translator stage gains a quality multiplier without reviewer bandwidth growth. This is the most likely near-term expansion.

What does not enter the queue (and why)#

Some texts on the corpus’s reference shelf are deliberately excluded, not merely deferred. Recording the exclusion is part of the project’s discipline.

Earlier roadmaps treated Book of Mormon material, Kojiki, and Egyptian Pyramid Texts as deferred. That position was reversed in 2026-06 — the LDS Restoration Scriptures are now shipped under the Edition track; Kojiki ch. 1–6 enters the queue in the Oomoto context (where its kami names are load-bearing); Egyptian work begins with the Heavenly Cow and the Pyramid Texts ascent spells.

Cross-references#

• The catalogue of texts the Raëlian canon engages — the source pool the roadmap selects from — is at the Reference corpus of the Raëlian canon wiki entry on the main site.
• The current production glossary state lives at data-content/i18n/translation-glossary.json in the data-content repository.
• The pilot-wave source acquisitions live at data-library/{slug}-woh/source-{lang}-1.json in the data-library repository — immutable provenance per file.
• The selection logic and exhaustive per-tradition acquisition tables are at Library Acquisition Program.