Compare portals
Choose two to four national portals. Each row shows whether its values can be compared directly.
National portals use different definitions and classification systems, so every comparison is marked Direct, Partial or Context only. Select “Why?” on a row, or a badge, for the reason.
Why snapshot data?Open-data catalogue totals are dynamic measurements rather than fixed statistics. OpenDataScope therefore uses a timestamped research snapshot as its canonical analytical dataset. Later official-page readings are retained as validation evidence and demonstrate normal catalogue drift over time.Research snapshot: 8 October 2026, 22:21–23:03 UTC9 October 2026, 00:21–01:03 in Skopje (UTC+2)Live portals can change after the research snapshot.How the snapshot was taken
| Dimension | North Macedonia | United Kingdom | United States | Netherlands |
|---|---|---|---|---|
| Catalogue size Partial | 768 datasetsNorth Macedonia counts public CKAN datasets. | 59,491 directory recordsThe United Kingdom counts directory records that link to data hosted elsewhere. | 590,168 metadata recordsThe United States counts metadata records, including restricted and non-public access levels. | 20,791 register recordsThe Netherlands counts register records, including non-public and restricted entries. |
| Publishers / organisations Context only | 62 active of 132 listedTop five publishers hold 32.3 % of datasets | 1,176 owners of indexed records (1,044 listed)Top five publishers hold 37.4 % of records | 107 active of 123 listed; federal sources hold 94.6 % of recordsTop five publishers hold 88.6 % of records | 26 source catalogues in the CKAN facet; data owners sit in a separate field |
| Portal type Context only | National catalogueCKAN 2.10.9 | National directory of links to data hosted elsewhereCKAN | National metadata catalogue, federal and non-federal | Register of open and non-open government dataCKAN |
| API availability Direct | CKAN Action API, no key; organisation and group lists require POST | CKAN Action API, no key | Catalog API v4; API key required | CKAN API v3, no key |
| Machine-readable formats Direct | JSON in 0.9 % and CSV in 58.9 % of datasets; 90.78 % of resources machine-readable (project rule)Resource-level indicators are project-defined | Not collected | Not collected | JSON in 45.6 % and CSV in 12.0 % of records; controlled EU vocabulary (40 values) |
| Licensing Partial | 22.3 % carry an open-licence id; 50.8 % 'not specified'; 27.0 % none | 37.6 % of records have any license_id; licences may sit in other fields | Not collected | 89.6 % carry an open-licence idThe register also contains non-public records |
| Metadata standard Context only | DCAT-AP extension installed (mk_dcatap); record exposure not verified | Not reliably available | DCAT-US 3.0 | DCAT-AP-DONL |
| HVD support Not comparable in this selection | Not applicable | Not applicable | Not applicable | Not collectedEarlier readings are validation only, so no main value is shown |
| Metadata activity Direct | 33.6 % of records | 53.0 % of records | Not collected | 81.7 % of records |
| Count definition Context only | Public CKAN packages, anonymous package_search | Public CKAN packages in the data.gov.uk directory | Sum of dataset_count over all catalog API organisationsIncludes non-open records. | package_search result.count (CKAN API)Includes non-open records. |
How to read this comparison
- Units differ: each portal counts datasets, directory records, metadata records or register records, so read the count-definition row before comparing numbers.
- The EU portal re-counts records harvested from national and institutional catalogues; its total overlaps national totals and is never ranked with them.
- Serbia is regional context only and is not part of the core comparison.
- Themes are not compared across national portals: no cross-country theme mapping is used where a reliable taxonomy crosswalk is unavailable.
- Research snapshot: 8 October 2026, 22:21–23:03 UTC. Live portals can change after the research snapshot.
Count definitions and comparability limitations are listed for every portal. Explained in methodology