# Tergwame
Tergwame (ትርጓሜ) is Geʿez for 'translation'.

> Greek, Coptic, Ethiopic, Syriac, Armenian, Georgian and Latin Christian literature (c. 30-1500 CE):
> a translation-coverage register, evidence-backed gap dossiers,
> verified electronic-text sources, Harvard Hollis records, and
> machine-assisted English translations with passage-level commentary.

Counts: 8,509 works catalogued; 531 curated gap
dossiers (382 with no English translation located,
138 partial/older only, 41 full controls);
14,869 manuscript and print records; 76 digital sources checked;
562 new texts identified.
Translations: 65 works, 328,809 English words
(machine-draft; no named human review).

Translation has begun. The first Coptic tranche runs on DeepSeek V4 Flash
(`deepseek/deepseek-v4-flash:floor`) and publishes source, English and
commentary side by side. Every record is labelled machine-draft.

## Key pages
- Overview: https://naxiai.com/christian
- Translations: https://naxiai.com/christian/works
- Gap register: https://naxiai.com/christian/gaps
- Catalogue: https://naxiai.com/christian/registry
- Holdings: https://naxiai.com/christian/holdings
- Scan register: https://naxiai.com/christian/scans
- Models and cost: https://naxiai.com/christian/models
- Sources and translation series: https://naxiai.com/christian/sources
- Hollis citations: https://naxiai.com/christian/citations
- Method and runbook: https://naxiai.com/christian/method

## Data
- Registry API: https://naxiai.com/christian/api/registry
- Gaps API: https://naxiai.com/christian/api/gaps
- Scan register API: https://naxiai.com/christian/api/scans
- Models and cost API: https://naxiai.com/christian/api/models
- Per-dossier JSON: https://naxiai.com/christian/api/gaps/{id}
- Models and cost API: https://naxiai.com/christian/api/models
- Works API: https://naxiai.com/christian/api/works
- Sources API: https://naxiai.com/christian/api/sources
- Citations API: https://naxiai.com/christian/api/citations

## Important caveats
- 'no English translation located' is a catalogue-and-series search result, not
  proof that no English translation exists, and a French/German/Italian/Latin
  rendering does not close the gap. Each dossier names what was checked and
  lists any non-English translation found, with its relation to the text.
- 'unknown' means the source catalogue has no translation field; it is not a gap.
- Licences differ by source: First1KGreek and Perseus are CC BY-SA 4.0; Coptic
  Scriptorium is mixed (NC and academic-only documents are catalogued, text
  withheld); Beta Masaheft files say CC BY-SA 4.0 while the site says NC-SA;
  PEMM and HMML metadata are CC BY 4.0; HMML images are not reusable.
- The catalogue counts works, not editions or manuscripts; holdings are separate.

## Sitemap
- https://naxiai.com/christian/sitemap.xml