How people try to read the Voynich manuscript, with live toys, in English and Thai.
https://nanobotco.github.io/voynich/ · https://nanobotco.github.io/voynich/th/
tools/prep.pyturns the sources intools/data/intodocs/data/:vms.json(every page of René Zandbergen's ZL transliteration, v3b, CC0: folio, illustration, Currier language, scribe, lines), and 14,000-word Latin and English samples from Project Gutenberg.tools/copy_text.pyholds all copy (EN + TH);tools/build.pywritesdocs/index.html,docs/th/index.html,llms.txt,sitemap.xml,robots.txtandicon.svg.docs/vms.jsis the engine: EVA glyphs drawn as pen strokes, text statistics (Zipf slope, word lengths, h1/h2 entropy, repeats), a slot grammar, and the text machines (letter swap, verbose cipher, a Rugg-style grille, self-citation after Timm & Schinner).docs/app.jswires the page.?cardat 1200×630 renders the share card (docs/card.jpg);?t=fixes the hero's moment.- Manuscript scans and the Voynich photograph come from Wikimedia Commons (
tools/credits.json), public domain.
Text CC BY 4.0, NaNoBotCo. Code MIT.