Open scholarly data

Structured companions to the human-readable pages. Identifiers and relationships retain their provenance in the reviewed corpus.

• publications.json — Publication metadata, full narrative sections, attribution, identifiers, and relationships
• discoveries.json — Discovery statements, full narratives, and evidence-to-publication mappings
• research-areas.json — Six research areas with themes, publications, and discoveries
• themes.json — Research area and theme membership
• connections.json — 129 documented directed publication relationships
• claims.json — Original claim-to-source mappings
• vocabulary.json — Controlled terms and publication memberships
• timeline.json — Chronology and scientific transitions
• publication_inventory.json — Source inventory, including an inventory-only entry; the 63 reviewed records define site coverage
• search_index.json — Original search index
• provenance.json — Source checksums and scope

Conventions

Publication IDs are stable slugs. Discovery IDs follow the original claim IDs. Publication canonical_url fields are absolute; original url fields are site-relative. Research-area URLs are absolute. Graph from/to fields reference publication IDs; substantiated_by references supporting papers. Missing identifiers are null, not inferred.