Now let's combine the motivations behind event-driven architectures and event sourcing.

Event-driven architectures communicate by atomic events. It makes sense to piggyback on this approach and build the system using event sourcing, by using the events as the system's source of truth. Doing so combines the benefits of both approaches, enabling horizontally scalable, event-sourced systems.

The question is how to model event-driven applications that base their domain model on events? And how to efficiently calculate and return the current state of domain entities?

The Command Query Responsibility Segregation (CQRS) principle describes how to model these applications. It is a consequence of event-driven architectures and is based on event sourcing.