Credits & Open Data Attribution
LingoGist uses several open data sets for character stroke animations and reference materials. We're grateful to the projects below — each is credited individually with its license, source, and (where applicable) a link to download the derivative data we ship.
KanjiVG
Used for: Japanese hiragana, katakana, and kanji stroke-order data
Original source: https://kanjivg.tagaini.net/
License: Creative Commons BY-SA 3.0
Our derivative Japanese stroke-data subset is available under CC BY-SA 3.0 via the data-export endpoint linked below.
animCJK (parsimonhi)
Used for: Simplified Chinese (zh) and Traditional Chinese / Cantonese (yue) stroke-skeleton centerlines, derived from the Arphic PL KaitiM GB / Big5 fonts
Original source: https://github.com/parsimonhi/animCJK
License: Arphic Public License (APL)
Our derivative Simplified Chinese (zh) and Traditional Chinese / Cantonese (yue) stroke-data subsets are available under APL via the data-export endpoints linked below.
IAPR-TC11 Indian Devanagari Character Dataset
Used for: Hindi (hi) Devanagari handwriting samples, source for the LG-1920 import
Original source: http://tc11.cvc.uab.es/
License: IAPR-TC11 dataset terms (research / non-commercial use, with attribution)
Our derivative Hindi stroke-data subset preserves the dataset attribution and is available via the data-export endpoint linked below.
Noto Sans Devanagari
Used for: Reference glyphs for the LG-1926 font-similarity picker that selects the best handwriting sample per Devanagari character
Original source: https://fonts.google.com/noto/specimen/Noto+Sans+Devanagari
License: SIL Open Font License 1.1
The Noto font is bundled with the importer pipeline only — no font glyph data is shipped to clients.
DM Sans
Used for: Latin-script text in share images and other server-rendered SVG output (LG-1937)
Original source: https://fonts.google.com/specimen/DM+Sans
License: SIL Open Font License 1.1
Bundled unmodified; the OFL allows embedding without redistribution-as-data obligations.
Derivative stroke data — download
The character stroke-skeleton data we derive from the sources above is available as JSON, one language per endpoint. These exports satisfy the "make available on the same terms" clause of the share-alike licenses listed above.
- Japanese stroke data (CC BY-SA 3.0)
- Simplified Chinese stroke data (APL)
- Traditional Chinese / Cantonese stroke data (APL)
- Hindi stroke data (IAPR-TC11 dataset terms)
News content
News-article cards are sourced from the public RSS feeds of the publishers below. We do not republish the original articles. Each headline and summary is rewritten by AI into a graded version in your target language for educational purposes; the original article is linked from the card and remains the property of its publisher.
Open-source dependencies
LingoGist's server is built on top of permissively-licensed open-source npm packages (MIT, Apache-2.0, ISC, BSD, MPL-2.0). The full attribution list is at /public/NOTICE.txt — generated mechanically from our production dependency tree (LG-1941).
Other acknowledgements
LingoGist uses Lucide icons (ISC license), the Anthropic Claude API for AI content generation, Google Cloud Text-to-Speech for audio, and Firebase Authentication. None of these ship third-party data inside our service that requires the redistribution mechanism above.