Privacy promise: what we actually store
We claim a game privacy policy short enough to read. This is the log behind it: which rows exist in our database, how long they live, and the three things we refuse to collect.
We claim a game privacy policy short enough to read. This is the log behind it: which rows exist in our database, how long they live, and the three things we refuse to collect. No marketing language, just the list. The play page is open if you came here for a game.
The rows we keep
- Match records: board size, ruleset, move list, result, timestamp. Without them the replay link does not open.
- Rating history: one row per game per player, holding the rating before and after.
- Guest id: a random string in your own browser storage. It is how your record follows you without an account.
- Aggregate counters: games played per day, used for the leaderboard window. Nothing is broken down per person.
That is the whole schema. There is no email column, no name column, no phone column and no advertising identifier, because we never ask for any of them.
The three things we refuse to collect
- Email and phone number. There is no signup form, so there is nowhere to type one.
- Analytics that follow you across other sites. Page counts are aggregated on our own server.
- Chat content from the match page. If we ever add one, it gets the same treatment as the move list: stored to make the feature work, and for nothing else.
How long things live
Match records stay until you delete them or until the record behind a guest id goes quiet for a long stretch; the exact window is on the privacy page. Rating history is kept as long as the ladder needs it, because dropping a row would change the ratings of everyone that player met. Server access logs are the shortest lived, counted in days, and exist so we can see when something breaks.
What a no data collection game still has to do
Some data has to exist for the product to run at all. A rated game needs a record of the result, or the rating is fiction. A replay link needs the move list, or there is nothing to replay. The line we draw is purpose: if a row is not needed to make a feature work, we do not create it. That is a narrower rule than collecting everything and promising to be careful, and it is far easier to audit.
The other half is enforcement. Anti-cheat runs on the server, which means every move gets checked against the rules. That check reads the board state, not who you are, and writes a pass or fail flag beside the game. Nobody reads it by hand unless a result is disputed.
Where the gaps are
Two, and we will say them plainly. IP addresses stay in server logs for a short window, because there is no other way to stop a flood of requests. And our host, like every host, can see traffic metadata at the network level, which is outside what we can change. Both are written on the privacy page rather than buried.
We went through the reasoning earlier in no ads, no tracking, just games. This is the follow up: the claim first, then the list of what is actually on disk.
FAQ
Does YiBoard store my name or email?
No. There is no signup, so we never receive either one. Your record is tied to a random guest id kept in your own browser.
Can I delete my games?
Yes. Deleting from the profile page removes the match records and the guest id. Rating changes already applied to other players are not reversed, for the same reason a chess result is not changed after the fact.
Do you sell data?
No. There is nothing in the schema worth buying, and no advertising identifier to sell it against.
Is the privacy policy different from what this post says?
It should not be. The policy page is the binding version. This post exists to say the same thing in plain words.
Related
- No Ads, No Tracking, Just Games
2026-08-16
- Spectator and Commentator Plans
2026-10-21
- Mobile Game Optimization: A Board Built for One Hand
2026-10-20