Options

The optional entity kind prefix is a directive that suggests how the entity might handle its state and received messages. In the example above, we expect the “Printer” entity to be a physical device. An “user” entity in of the same name could be expected to be a person who prints.

The options available suggest how the entity might handle its state and message input:

• kind - indicates the intended kind of entity in an argument to the kind entity.
• event sourced - indicates that event sourcing is used in the persistence of the entity’s states, storing a log of change events
• value - indicates that only the current value of the entity’s state is persisted, recording no history off the change events
• aggregate - indicates this entity is an ‘aggregate root entity’ as defined by DDD.
• persistent - indicates that this entity should persist its state to stable storage.
• consistent - indicates that this entity tends towards Consistency as defined by the CAP theorem and therefore uses sharding and the single writer principle to ensure a consistent view on state changes
• available - indicates that this entity tends towards Availability as defined by the CAP theorem and therefore replicates its entity state to multiple locations or even across data centers using CRDTs (conflict-free replicated data types).
• grpc - indicates that the entity should be accessible via gRPC api invocations
• mq - indicates that commands and queries may be sent by a message queue such as Kafka, Pulsar, or Google Pub Sub

Entities use event sourcing to keep track of the entire history of changes to the entity’s state.