Event Sourcing
In plain terms
Storing state as a sequence of events rather than just the current snapshot.
Going deeper
Instead of overwriting data, event sourcing records every change as an immutable event, and current state is rebuilt by replaying them. It gives a full audit trail and time-travel, at the cost of complexity.
Under the hood
Append-only events give full auditability and replay but complicate querying (often paired with CQRS) and require careful event schema evolution.
Examples
Banking ledgers, order-state history, Kafka as an event log, Axon framework.
Think of it like...
Like a bank ledger that records every deposit and withdrawal: your balance is derived from the full history, not just a single number.