About 關於

🧸 What this is

Eighteen little 3D worlds for children learning English, about 2 to 6 years old. Click anything and hear its word and one sentence about it, then, at the press of a button, a short, friendly conversation about it. There are games to find things and to say their names, and six science worlds with experiments to guess and watch.

They were made in Hong Kong for two brothers, Hank and Cooper, and are shared for every child learning English. Free, with no ads. A family account, to save each child's progress, is optional.

🔒 Privacy 私隱

• Without an account, nothing about your child leaves your device. What has been heard, found and said, and the settings, are kept in this browser only (its local storage). To wipe them, clear this site's data in the browser's settings.
• A family account is optional (on the website). A grown-up makes it with a passkey: no email, no password and no real names. We keep the account's name, your passkeys' public keys, and for each child a nickname, a picture and which words they have heard, found and said, so that it comes back on your other devices. It is kept in the site's database at Cloudflare, which hosts the site. Other families cannot see it, and it is never sold, shared or used for ads. Cookies are used only for signing in (one for a few minutes while you sign in, one that keeps you signed in), and only once you do.
• How long it is kept: until you delete it. On the family account page you can download all of it, or delete a child or the whole account, at any time. Deleting takes it out at once; the database's automatic backups can hold it for up to 30 more days, and then it is gone for good.
• No ads and no tracking across sites.
• We count visits with Cloudflare Web Analytics, which uses no cookies and collects nothing personal: only things like which page was opened and how fast it loaded.
• The worlds never use the microphone or the camera, and the site tells the browser to refuse them.

🙏 Credits 鳴謝

• The 3D worlds are drawn with three.js (MIT licence, © the three.js authors).
• The voices were made with Google Gemini text-to-speech, and every clip was checked by ear with OpenAI's Whisper speech recogniser.
• The words were chosen with the help of Wordbank (Frank, Braginsky, Yurovsky & Marchman: which words thousands of children learn first), age-of-acquisition ratings by Kuperman, Stadthagen-Gonzalez & Brysbaert (2012), and concreteness ratings by Brysbaert, Warriner & Kuperman (2014).
• The screen-time advice follows the World Health Organization (2019) and the American Academy of Pediatrics.