Public Methodology
This page explains how the Compendium treats sources, evidence tiers, folklore, community leads, disputed claims, and corrections. It is public by design; internal research notes and private review material are kept separate.
AMC gives priority to official records, archival material, contemporaneous newspapers, books with citations, named witnesses, field reports, and clearly identified photographs or scans. Later retellings are useful, but they are not treated the same as earlier evidence.
Reddit posts, forum threads, podcasts, videos, and personal submissions can point toward interesting research trails. They are treated as leads until independent sources are found.
A case file should preserve the mystery while also naming skeptical explanations, mundane explanations, hoax claims, misidentifications, or later corrections when they exist.
Supported by strong primary or near-primary records: official documents, contemporary newspapers, court/police records, medical/scientific papers, archival scans, or multiple named sources.
Supported by multiple independent secondary sources, credible local reporting, or a combination of witness accounts and documented context.
Interesting but source-limited. Often based on later retellings, memoirs, folklore collections, community reports, or single-source testimony.
Weakly sourced, heavily disputed, folkloric, or mainly useful as a research lead. Tier D does not mean “false”; it means “do not overstate this.”
Private research notes, bot search logs, rejected leads, possible hoax notes, duplicate checks, GitHub issue links, submitter emails, and working theories are not displayed here. Those belong in the private review system, not on the public website.