Methodology

How entries, interviews and sources get into the museum, and how much you should trust each kind. Short version: we link to where material lives, we say what we have and haven’t opened, and we fix errors at the source.

Entries

• A person curates the list of Wikipedia pages that belong here; the text is read live from Wikipedia and is not reviewed or rewritten by us.
• Dates and coordinates come from Wikidata; images and their licenses from Wikimedia Commons.
• Connections between entries are the links between the underlying Wikipedia articles.
• Because this is live, an entry can change when Wikipedia changes. Contested claims are Wikipedia’s to resolve; use the “Improve this entry” link.

Interviews, archives and primary sources

• Each record in the catalog links to the holder; we don’t host the recordings or documents.
• Interview / talk: a recording or text of a person speaking, with interviewer and date when known.
• Archive / collection: a library or archive holding relevant material; usually a finding aid, often access by appointment.
• Primary source: material from the events themselves (period press, film, recordings, posters, letters, official records), as distinct from later retellings.
• Records found from search listings but that our automated checker couldn’t read (bot walls or script-rendered pages) say so on the card.
• A weekly script re-checks every link; failures are marked “link unverified” until fixed or removed.
• Dates marked as approximate are inferred (for example from a broadcast or post date) and say so in the notes.
• “Possible recordings on the Internet Archive” are automated, unvetted search matches. They are leads, not endorsements.

Where the record stands

These numbers are counted from our data each time the site is built, so they can’t drift from the catalog.

• 88 of 171 entries have at least one catalogued interview, archive or primary source. The rest say “No sources catalogued yet.”
• 141 catalog records. An automated check on 2026-11-01 could open 139 of them.
• 0 of 141 records have been opened and checked by a person. That work is under way.
• 0 records have a saved web archive copy.
• 4 searches for sources on entries that have none are logged, so we can say “searched, nothing found” rather than imply nobody looked.

What we don’t do (yet)

• We don’t independently verify the accuracy of what is said in an interview or source.
• We don’t host copyrighted audio, video or images without permission.
• We don’t yet mark every record as opened by a human reviewer; that is on the roadmap.

Spot a mistake?

See Rights, corrections & takedowns. For encyclopedic text, the fastest fix is on Wikipedia or Wikidata, which improves the page for everyone.