In short
A noticeboard that belongs to the people using it
Imagine the noticeboard outside a market. People pin up what they can offer and what they need. Someone lends a hand, someone else remembers the favour, and the whole thing runs on the fact that everybody roughly knows who everybody is.
This is that noticeboard, on your phone. The difference is that there is no company behind it. Nobody owns it, nobody can switch it off, and nobody is reading over your shoulder to sell you something.
It does three things
One
People help each other, and the help is counted
You post what you can offer or what you need. When someone helps you, both phones sign it, and the hour is recorded. An hour is an hour: a lawyer’s hour and a cleaner’s hour are worth exactly the same here. It is a time bank, not a market.
Two
A group can work out what it actually thinks
Somebody asks a real question. Everyone writes short statements and agrees or disagrees with each other’s. There is no reply button, so there is nothing to argue with. What the app then shows you is not the most popular opinion but the ones people on opposite sides both agreed with.
Three
You can tell who is who without any ID
Nobody signs up. Your phone makes you a private stamp that only it holds. Standing comes from people who have met you in person saying so, which means it is earned from your neighbours rather than granted by an office.
Why it is built oddly
It was made for places where the internet is expensive and comes and goes, so it assumes you are offline most of the time. Everything lives on your phone and works with the data off. When you do get a signal it swaps a small file with a web address, and that is the entire network: files on a shared shelf, not a company holding your data.
It was also made for places where people who need each other do not always trust each other, across language, faith or family lines. That is why the voting screen hides who wrote what, why hiding a post needs objections from people on both sides of an argument rather than a loud majority.
Why the Collective
Three sets of ideas built this, and they do not entirely agree with each other. Where they conflict, the disagreement is usually visible in the code.
Plurality
Technology that works across difference, not around it
Plurality, by Audrey Tang and Glen Weyl, argues that most social technology does one of two bad things with human difference: it flattens it, pretending we are all the same kind of user, or it farms it, because outrage between groups is profitable. It proposes a third option. Build tools that let groups who do not agree cooperate anyway, without requiring them to stop disagreeing.
The deliberation screen is the most direct borrowing. It follows the approach used in Taiwan, where people write short statements and agree or disagree with each other’s, and the system surfaces the statements that people on opposite sides both accepted. There is no reply button here for the same reason: replies reward the sharpest thing you can say to someone who already disagrees with you.
The trust model comes from the same place. You are not one credential issued by an authority. You are the sum of people who have met you and said so, which is a different thing in every neighbourhood you walk into.
Mutual aid
Solidarity, not charity
Mutual aid is an old idea with a precise meaning, and it is not a synonym for helping people. Charity runs downhill: someone with more decides who deserves it, and the person receiving is a case to be assessed. Mutual aid runs sideways. Everyone gives and everyone receives, nobody applies, and nobody is means-tested.
That is why the ledger here always sums to zero. There are no donors and no recipients, only people who are currently ahead and people who are currently behind, and those swap around. It is why an hour is an hour whoever works it, and why both people have to sign an exchange: nobody is given anything, and nobody is owed anything until they agree they are.
Effective altruism, partly
Measure honestly. Do not appoint anyone to decide
Effective altruism’s good habit is refusing to let good intentions stand in for good results. Ask whether the thing works, count it, and be willing to say plainly when it does not. That habit is all over this project: every claim it makes about itself has a test behind it, and the parts that do not work are written down rather than quietly dropped.
What is deliberately not taken is the rest of it. Effective altruism tends to end with a small number of clever people deciding where everyone’s resources should go, on the strength of their own analysis. This has no such seat, and could not have one. There is no central pot, no allocator, and no committee that ranks whose need matters most. Your neighbours decide what they owe each other, and no spreadsheet overrules them.
Where mutual aid and effective altruism disagree, this follows mutual aid. Where Plurality and either of them disagree, it follows Plurality, because the hardest problem here was never how to distribute help efficiently. It was how to get people who have reason to distrust each other to cooperate at all.
What it cannot do
If you lose your twelve words, it is gone. There is no company to email and no password to reset. That is the cost of nobody holding your account.
It speaks one language. Everything here is in English. A commons meant to bridge divides, in a country with hundreds of languages, currently reaches the people who read one of them.
There is nobody to appeal to. If your neighbours hide something you wrote, there is no head office to overrule them. Nothing is ever deleted and you can always read it, but there is no referee.
It is empty until people join. A noticeboard with nobody at it is a blank wall. Everything here depends on real people you actually know turning up.
The one-sentence version: it is a noticeboard and a record of favours that belongs to the people using it, works without the internet, and has nobody in the middle.