This project automatically processes thousands of scanned document pages using AI-powered OCR to:
Extract and preserve all text (printed and handwritten)
Identify and index entities (people, organizations, locations, dates)
Reconstruct multi-page documents from individual scans
Provide a searchable web interface to explore the archive
This is a public service project. All documents are from public releases. This archive makes them more accessible and searchable.
Someone back this all up in that huge Minecraft book server.