Pre-release. Until the official launch, content may change, including published files.
datamodels.jp

Source: https://datamodels.jp/en/LICENSE-CONTENT · Text CC BY 4.0

Licences for model content ​

Machine-readable files are CC0 1.0 (no conditions), prose is CC BY 4.0, code is Apache-2.0. Decided on 2026-09-28 (#83).

Which licence applies ​

WhatLicence
Machine-readable files: JSON Schemas (schema.json, /schema/…), JSON-LD contexts (context.jsonld, /context/…), RDFS vocabularies (/vocab/…), examples (examples/, /examples/…), catalog.yaml, mapping files (mapping/*.yaml, /mapping/…), catalog.json and the adapter files (/adapters/…)CC0 1.0 Universal
Prose: notes.yaml and README.md in the model folders, and the pages of https://datamodels.jp (model pages, subject pages, guides)CC BY 4.0
Code: everything outside models/, including site/, scripts/, the adapter code and CIApache-2.0, see LICENSE

A licence applies to a whole file. Text inside a machine-readable file, such as the titles and descriptions in a schema or the labels in a vocabulary, is CC0 there like the rest of the file.

The reason for the split: attribution works where people read, and it does not survive a schema pasted into another repository. So credit is asked only for prose, where it can actually be given.

Credit and the canonical URL ​

  • Prose (CC BY 4.0): credit "datamodels.jp" with a link to the page you used, and say whether you changed it. Every page shows its URL at the top.
  • Machine-readable files (CC0): no conditions. Each file already names its canonical URL: $id in a schema, the @context URL, the type and attribute IRIs. Please keep them when you copy a file. This is a request, not a condition: it lets anyone who finds a copy see that a newer version exists.

Parts from other sources ​

  • Copied from a CC BY source: a file that contains content copied from a source under CC BY, such as Smart Data Models or the Digital Agency's GIF core data model schema, stays CC BY 4.0 as a whole, with that source's attribution. This holds for machine-readable files too. The LICENSE.md of the model folder names each such file and its source.
  • Today no model contains copied upstream content. Upstream standards are mapped, not copied: the mapping/*.yaml files record the correspondence, and IRIs are reused. A reused IRI is a name, not copied content.
  • GIF: the repository (https://github.com/digital-go-jp/GIF) is CC0 1.0, and its core data model schema GIF_Core_DataModel.xsd declares CC BY 4.0. The mapping files name GIF fields to record the correspondence. #65 asks the Digital Agency which licence applies.
  • 自治体標準オープンデータセット: material from the Digital Agency's municipal standard open datasets (公共データ利用規約（第1.0版）, PDL1.0, compatible with CC BY 4.0) is credited to its source in the model's mapping files.

Change ​

Until 2026-09-28 all model content was licensed under CC BY 4.0. The split above applies to every published version, including files published before that date.