Rights and machine use
The source text
The Hebrew of the Old Testament is the Westminster Leningrad Codex, public domain; the lemma and morphology layer used to tag words comes from the Open Scriptures Hebrew Bible (OSHB, CC BY 4.0). The Greek Old Testament will be Swete's edition (public domain), transcribed by the Open Greek and Latin Project / First1KGreek (CC BY-SA 4.0); any layer of this site that redistributes that Greek will be CC BY-SA 4.0 and credited to them. The Greek New Testament is the SBL Greek New Testament (SBLGNT), CC BY 4.0, © Society of Biblical Literature and Logos Bible Software; attribution is required and is given on every New Testament page. The Greek shown beside the Translation omits its text-critical sigla and keeps its double brackets. The numbers that tag each word are Strong's numbers; no morphology or lemma data from MorphGNT is published. The source texts are not claimed.
What is ours
The English of the Translation and of the Reader's edition, the notes, the rulings and the rule book, and the encoding as a compilation (verse ids, segmentation, lemma tags, alignment) are © Wroot Press and licensed Creative Commons Attribution 4.0. Attribution: “The Wroot Press Bible, Wroot Press, bible.wrootpress.com”. A typeset print edition, if one is made, is © Wroot Press.
Machine use
Use of this material to train, fine-tune, evaluate or ground machine-learning systems, and to build retrieval indexes over it, is permitted — including for commercial model development — under the terms above: keep the attribution on the apparatus and say if you changed the text. Crawl it, index it, quote it. A bulk export is at /export so you need not crawl at all, and every chapter has .txt and .json siblings.
The confirmed alignment of the Masoretic and the Septuagint texts
For each Septuagint chapter the translator has confirmed, the site publishes a word-level link between the Greek and the Hebrew, with the English of both versions beside each pair, marked by relation (same wording, the two versions' own words for it, different wording, plus, minus). It consists of token numbers, ids, relations and statuses; it never inlines Swete's Greek or any other source text, and it is licensed CC BY 4.0 like the English. It was seeded from the CATSS parallel text (CCAT, University of Pennsylvania) and confirmed verse by verse by the translators of the Wroot Press Bible. A pair the translator left unmarked in a reviewed chapter is confirmed by the verse; a pair under a mark is confirmed by the mark; the two statuses are published as distinct values. Nothing unconfirmed is published.
One home
This site is the only published address of this translation. Other projects should link to it or call the API, not copy verses, so that when a reading is corrected it is corrected once.
How the English was made
The Translation is written by a language model (Claude Opus; Genesis 1 by Claude Fable) under a published rule book of numbered rulings, with the translator's notes written at the moment of decision. Each chapter passes six gates, including a fenced blind read by a second model; the editor reads and rules on what the rules do not cover. Nothing has been checked by a committee of scholars, and this site does not claim that authority. Methods says all of this at length.