Log

Goal: Gate user actions (e.g. make a post) based on payment of a token. This is a much larger problem than it seems. Naive idea: User buys credits, check user balance, subtract a token before allowing a post. But: You have to track payments in and credits out in order to keep balances correct. So you need a ledger. A single-entry ledger will break when you encounter double-spending, retries, refunds, and audits. So you need a double-entry ledger. I have looked at outsourcing this. But offering complex custom bookkeeping software to non-enterprise customers is a structurally bad business. Also, your app must make an external API call on every gated user action. This indicates that you need to build and maintain an internal double-entry ledger system. This is why most apps: 1) Charge a subscription or 2) Charge for usage afterwards (no real-time gate) But neither of these models produce the user experience incentives that I think are required for a good version of the future.