A data-only archive of the Cherokee New Testament: page scans of the printed edition paired with verse-level transcriptions and their English source lines. There is nothing to build or run here — the repository is a dataset.
This data descends from the Cherokee New Testament Online project,
initiated by Timothy Legg in 2006: the printed edition was digitized on a
flatbed scanner and the full New Testament transcribed by hand over about
three years, with two complete proofreading passes. The project ran
2006–2019; this repository was seeded from Legg's own 2011 data archive
(cntdata/cnt_data.tar.gz). See docs/LEGG.md for the full
acknowledgment of his work. The original site (cherokeenewtestament.org) is
no longer reachable; the project content is preserved in the
CherokeeNewTestament.org repository
and mirrored at cherokeedictionary.net/cnt/.
| Path | Contents |
|---|---|
cnt/ |
Page-scan images, one PNG per page, named BBCCVV.png |
cntdata/ |
Verse-level transcription text, English source lines, chapter HTML pages, and auxiliary working files |
.issues/ |
Local issue store (separate issues-data branch) |
.opencode/ |
Agent deck (configuration, skills, tooling) |
Every data file is named by a six-digit key BBCCVV — book, chapter, and
verse, two digits each:
BB— book number,01–27(Matthew through Revelation, in canonical order)CC— chapter number within the bookVV— verse number within the chapter
So 010101 is Matthew chapter 1, verse 1.
For each verse key BBCCVV:
| File | Contents |
|---|---|
cnt/BBCCVV.png |
Scan of the printed page containing that verse |
cntdata/BBCCVV.txt |
The verse in Cherokee, written in Latin transliteration — not Cherokee syllabary. Searching for syllabary characters will not match these files. |
cntdata/BBCCVV.eng |
The English source line for that verse |
Book-header files use BB0000: cntdata/BB0000.txt carries the book-title
line. There is no .eng book-header form.
Chapter pages are rendered as cntdata/BBCC.html (chapter-level key with the
verse digits dropped). These HTML files reference the scan images by bare
filename (e.g. src="010000.png") — the PNGs themselves live in cnt/, not
cntdata/, so the HTML will not resolve its images in place.
Besides the per-verse data, cntdata/ holds auxiliary working files that are
not per-verse data:
book.txt,combined.txt,output.txt— working/aggregate text filesproofread*.txt,proofread_list*.txt— proofreading listscnt_data.tar.gz,website.tar.gz— archived snapshots
See docs/DATA.md for per-file-type detail and the data conventions observed in the transcriptions.