Privacy

What we collect, what we do not, and what happens when we are ordered to remove something.

**We ask for one email address, and that is the only thing we hold that could identify you.** It is used for exactly two things: proving at signup that there is a person behind a new account, and sending you a code if you lose your password. It is never shown to another user, never attached to anything you write, and never given or sold to anyone.

We would rather not hold it at all — an address is the one field that ties a handle to a real person — and for a long time this page said we did not. What changed is that recovery codes alone lock out anyone who loses them, and inviting people to a place they can be permanently locked out of is its own kind of unkindness.

You can take it back off. Removing the address from your account leaves you with recovery codes, exactly as it was before, and leaves us with nothing.

We do not send anything you did not ask for. No newsletter, no digest, no "we miss you". Two kinds of message exist: your signup code and your password code. Both are plain text with no images, so there is no pixel to tell us when you opened one, or whether you did.

We do not sell data, and there is not much to sell. We store what you wrote, when you wrote it, and a count of how many times each crib has been viewed. Views are recorded against a salted hash of a browser session, not against you.

We do not track you elsewhere. No third-party scripts, no analytics services, no advertising pixels. The fonts are served from this server rather than from anyone else's, so loading a page tells nobody but us that you were here.

Cookies. Two. One remembers you are signed in. One holds a random session identifier used to keep the response tiers in a stable order and to avoid counting the same reader twice. Neither identifies you to anybody.

On free speech, honestly

An earlier version of this page said cribate was "based on absolute free speech". That was not true and could not be, and saying it would have been a promise we broke the first time we removed anything.

Here is the actual position.

We remove very little. There is no engagement algorithm deciding what you see, and no rule against being wrong, unpopular or annoying.

When we do remove something, we say so in the place it used to be. Removed content leaves a marker, not a gap. The marker names who removed it — the author, this platform, or a government — the reason given, the countries it applies to, and a cryptographic hash of what was there.

When a government orders a removal, we publish the reason they gave us. Verbatim, where the law allows it. We will comply with lawful orders, because the alternative is a person going to prison rather than a principle being upheld. What we will not do is comply quietly.

Removals are scoped where they can be. An order from one country restricts the content in that country, not everywhere.

The promise we can keep is not that nothing will ever be removed. It is that nothing will ever be removed *silently*, and that you will always be able to see what was taken and who asked.