Source Transparency Watch¶
Are the organisations behind global digital-rights data adopting primary-source, open-data transparency — and when did they do it? This page tracks peer sources (the aggregators behind our scorecard indicators, plus related rights and data-transparency bodies) for four signals:
-
Datasets
Machine-readable downloads — CSV / Excel / JSON, "get the data" exports.
-
Document & legal files
Links to the actual primary documents — per-country law texts, document-link manifests.
-
APIs / endpoints
Public APIs or machine-readable data endpoints.
-
Structure / portals
New open-data sections, data portals, and URL/permalink restructures.
Adoption timeline
By source
How to read this
Each entry is the first archived appearance of a transparency signal on that source's site — a reproducible, bot-friendly proxy for "first offered publicly," drawn from the Internet Archive's capture history. A ★ marks signals that are on-topic for that source (e.g. a data-protection dataset on UNCTAD, rather than an unrelated dashboard).
Caveats — read before citing
Web-archive coverage is incomplete, so a first-seen date is an upper bound on adoption: an organisation may have offered something earlier without it being archived, and the absence of a signal is not proof it never existed. Treat these as indicative leads, and verify the underlying page before citing. This watch is regenerated from the live archive — see methodology below.
Why this matters¶
Our project publishes its scorecard data, scoring rules, and sources openly. This watch holds the wider field to the same standard — surfacing when peer organisations move from closed summaries toward downloadable data, primary-document access, and open APIs. It also doubles as an early-warning system for our own pipeline: when a source restructures its URLs or relocates its data, that shows up here first.
Methodology¶
The watch is generated by utils/build_transparency_watch_data.py, which queries the
Internet Archive Wayback CDX API
for each monitored domain, matches captured URLs against transparency-signal patterns
(datasets, document manifests, API paths, open-data sections), filters out analytics/asset
noise, and records the earliest archived appearance of each — preferring on-topic examples.
To refresh it:
This writes docs/transparency-watch/data/transparency.json, which this page renders. Add or adjust monitored sources in the SOURCES list at the top of that script.