Problem
docs/content/finding-data/downloading.md describes a v4 archive under "Archive contents" — main.json plus score and count CSVs. The annotation CSVs, mapped-variant JSON, and VA-Spec NDJSON shipped in v5 on 2026-06-25 and are undocumented. The page also describes Zenodo as the only route to bulk data.
Proposed behavior
The page documents the stable URL as the primary route for automated consumers, keeps Zenodo as the citable versioned route, and describes the archive as it now exists.
Acceptance criteria
- "Archive contents" lists every member the export currently writes, including the annotation CSVs, VRS NDJSON, mapped-variant JSON, and VA-Spec NDJSON.
- The stable URL and its cadence are documented, along with how to reach a prior dated archive.
- The archive extension change from
.zip.gz to .zip is stated.
- The CC0-only restriction is stated plainly, with the API named as the route to more restrictively licensed score sets.
- The page notes that the archive path is rate-limited and that a 429 means retry rather than a permanent block.
- Zenodo is described as the citable versioned snapshot rather than as the only bulk route.
Problem
docs/content/finding-data/downloading.mddescribes a v4 archive under "Archive contents" —main.jsonplus score and count CSVs. The annotation CSVs, mapped-variant JSON, and VA-Spec NDJSON shipped in v5 on 2026-06-25 and are undocumented. The page also describes Zenodo as the only route to bulk data.Proposed behavior
The page documents the stable URL as the primary route for automated consumers, keeps Zenodo as the citable versioned route, and describes the archive as it now exists.
Acceptance criteria
.zip.gzto.zipis stated.