Privacy Policy
Kanji Daily · last updated 23 August 2026
Kanji Daily does not collect personal information from anyone. This page explains what that means in practice, and what the project does handle.
What this is
Kanji Daily ("the project") is an automated publishing tool operated by Robert Scott, an individual based in United States of America. It generates a short educational video about one Japanese character each day and uploads it to the operator's own YouTube, Instagram and TikTok accounts. There is nothing to sign up for, no account to create, and no service that other people log into.
Information we do not collect
The project does not collect, store, sell or share:
- names, email addresses, phone numbers or postal addresses of visitors or viewers;
- account information belonging to anyone other than the operator;
- viewing history, watch time or engagement data about individual people;
- location data, device identifiers or advertising identifiers.
This website serves static pages only. It sets no cookies, runs no analytics, embeds no third-party scripts or fonts, and has no contact form. Your visit here is not logged by us beyond whatever standard, short-lived request logs the static host keeps for operational purposes.
Information the project does handle
To upload a video to a platform, the software has to be authorised to post to the operator's own account on that platform. That authorisation is the only personal data involved, and it belongs to the operator:
- Platform access and refresh tokens for the operator's YouTube, Instagram and TikTok accounts, plus the numeric account identifiers those platforms require in an upload request.
- Basic account information returned by the platforms during authorisation — for example the channel or creator name, and the privacy settings an account is permitted to use. TikTok requires this to be requested before every post.
These credentials are stored as encrypted secrets in the project's automation environment, are used only to publish the project's own videos, and are never transmitted to anyone else. They are not used to read viewer data, and the project requests the narrowest permissions each platform offers for uploading.
Where the video content comes from
Lesson content is generated from public dictionary data — KANJIDIC2, JMdict and KanjiVG — together with original music and artwork. No personal data of any kind is used to produce a video.
Third parties
The project sends data to a small number of services, all of them in the course of making and posting a video:
- YouTube / Google, Instagram / Meta, TikTok — receive the finished video file and its caption, so that it can be published. Their own privacy policies govern what happens after that.
- A text-to-speech provider (Google Cloud Text-to-Speech or OpenAI) — receives the Japanese and English lesson text to be read aloud. This text is dictionary material; it contains no personal information.
- GitHub Actions — runs the automation on a schedule and holds the encrypted credentials described above.
No data is sold, rented, or shared with advertisers or data brokers, because none is collected in the first place.
Google user data and Limited Use
The project's use of information received from Google APIs adheres to the Google API Services User Data Policy, including its Limited Use requirements. Concretely: the YouTube authorisation granted to this project is used solely to upload the project's own videos to the operator's own channel. That data is not transferred to others except as necessary to provide that function, is not used for advertising, and is not read by humans except with permission, for security purposes, or where required by law. The project uses the YouTube API Services, and is therefore also bound by the YouTube Terms of Service; Google's privacy practices are described in the Google Privacy Policy.
Retention
Credentials are kept only while the project is publishing, and are deleted when access is revoked or the project stops. Published videos remain on the platforms until the operator deletes them. The lesson data used to build the videos comes from public dictionaries and is kept indefinitely, because it contains nothing personal.
Removing your data
Since the project holds no data about anyone but its operator, in almost every case there is nothing to delete. If you believe otherwise — or if you appear in a video and want it taken down — see Data Deletion or write to the address below, and it will be handled promptly.
Children
The videos are educational and suitable for general audiences, but the project is not directed at children under 13 and knowingly collects no information from them — or, again, from anyone.
Changes
If this policy changes, the revised version will be posted here with a new date. The project handles so little data that material changes are unlikely; if one ever amounts to collecting personal information, it will be stated plainly at the top of this page.