# EDRDG-derived data notice

KanjiStudy4All uses selected fields from KANJIDIC2 and uses JMdict to verify Japanese vocabulary spellings and kana readings. EDRDG identifies these files and data derived from them as licensed under Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0).

## Licence and attribution

- EDRDG licence statement: <https://www.edrdg.org/edrdg/licence.html>
- Packaged local snapshot: `third-party/EDRDG-LICENCE.html`
- CC BY-SA 4.0 deed: <https://creativecommons.org/licenses/by-sa/4.0/>
- KANJIDIC project: <https://www.edrdg.org/wiki/index.php/KANJIDIC_Project>
- JMdict/EDICT project: <https://www.edrdg.org/wiki/index.php/JMdict-EDICT_Dictionary_Project>

Commercial use is permitted when the licence conditions are met. The application code may use a separate, including closed-source, licence. Canonical and generated data artefacts containing EDRDG-derived adaptations are marked as CC BY-SA 4.0, and attribution, change indication, provenance and ShareAlike information must remain available.

## Field boundary

Fields based on or checked against EDRDG data:

- `glyph`, KANJIDIC2 source identifier and selected `meaningsEn`;
- `strokeCount`, classical `radical`, `onyomi` and `kunyomi`;
- `commonVocabulary[].written`, `commonVocabulary[].readingKana` and `jmdictEntryId`;
- upstream version, snapshot, hash, access date and transformation entries in `provenance`.

Original KanjiStudy4All editorial fields:

- `difficultyBand` and `difficultyRationale`;
- `meaningsTr`, `usageNotesTr`, vocabulary Turkish meanings and usage labels;
- Japanese example sentences authored for the app, kana renderings and Turkish translations;
- `mnemonicTr`, pedagogical `components`, `confusableKanji`, stable app IDs and review workflow state.

The original editorial fields do not erase or reduce EDRDG rights in upstream-derived fields. The distributed combined data artefact is therefore accompanied by this CC BY-SA 4.0 notice.

## Transformations

KANJIDIC2 records are selected, normalized into the app schema and cross-checked with Unicode Unihan. JMdict entry IDs are resolved to verify Japanese spelling/reading pairs; JMdict English glosses are not translated into Turkish. The build combines separately versioned batches and emits compatibility fields for the legacy quiz.

## SKIP codes

KANJIDIC2 SKIP codes are not used. The parser does not read `q_code` or `qc_type="skip"` values, the schema has no SKIP field, and no canonical or generated record stores or displays a SKIP code. Jack Halpern's additional SKIP conditions therefore do not apply to the current dataset. If SKIP is introduced later, its separate licence and attribution requirements must be reviewed before release.

## Updates

EDRDG requires a procedure for regular data updates. KanjiStudy4All performs a monthly upstream snapshot check as its own operational policy; the monthly cadence is not presented as a universal fixed interval for every application.