Original subreddit exports, user-history downloads, and vendor research files.
The project needs a visible boundary between raw material and derived interpretation.
This route makes the ledger explicit: what is cleaned archive, what is user intake, what is external-event context, and what is still pending.
Narrow the current section by spike month and flair without losing the route context.
Artifact ledger for the public-facing build.
caption
The site should make it obvious which layers are raw exports, which are cleaned views, and which are derived interpretations. That boundary is what keeps the project defensible.
Current public surface scale.
caption
This is the public accounting layer: archive volume, language review, source registry size, and the depth of the normalized event ledger.
The site keeps raw material, cleaned material, and interpretation in distinct lanes.
Normalized NDJSON, merged event registries, Hebrew review queue, and source inventory.
Spike correlation, narrative clustering, contradiction scoring, and rhetoric watchlists.
caption
This is the project’s main integrity rule. A public-facing reader should be able to tell where direct archive facts stop and where the investigative synthesis begins.