The original note, as written at the time
A chat application that will not accept to send a message if the user on the other side doesn’t respond to an online and available ping.
Ongoing development
A chat application that refuses to send a message until the person on the other side responds to an availability ping.
It reads as hostile and it is actually a design against a specific rudeness: the assumption that because a message can be delivered instantly, it must be read instantly.
The current default is that every message is an interruption and every sender assumes presence. A ping asks a question the recipient can answer with a state rather than a sentence: I am here, I am not. If they are not, the message waits, unsent, and the sender knows the difference between a message that has been ignored and a message that has not yet been received.
That distinction is the whole value. Most of the anxiety in messaging comes from the gap between delivered and read being invisible, so people fill it with stories. Making availability explicit removes the stories.
The obvious cost is friction, and it is a real cost: nine times out of ten you want to send the thing and get on with your day. So the ping must be cheap and it must be optional per contact. A colleague you work with hourly should never ping. A cousin you have not spoken to in a year probably should.
The deeper idea is that synchronous and asynchronous are different media and we have collapsed them into one. Chat treated everything as immediate, and then invented a hundred small rituals — typing indicators, read receipts, last seen — to recover the information the collapse destroyed.
The ping is the honest version of all of them.