Explore the archive
Ranked views and lists across the whole archive, which the /browse doesn't include.
- Most programs mentioned - Documents referencing the most surveillance programs.
- Most people mentioned - Documents whose metadata names the most individuals.
- Most countries mentioned - Documents spanning the most countries.
- Longest documents - Documents with the most pages.
- Portion markings - The marking on each classified portion, read out of the OCRs (levels, compartments, and how often).
- Classification patterns - The marking mix across document years, and an agency x classification heat table.
- Publishers by year - When each publisher released documents, and how much (i.e. documents and stories per publisher and year).
- Technical IDs - Every document id, PDF filename, size and SHA-256 checksum.
- Genres - How many documents of each kind the archive holds (slides, reports, wikis, manuals, guides, newsletters).
- Assemblies - Sets of pages this archive put together from documents and copies published separately, so they can be read in one place.
- Acronyms & codewords - Every ALL-CAPS token across the corpus, auto-extracted (e.g. acronyms, program names, compartments, org designators).