What is the Difference Between Identity and Entity?

In the context of Domain-Driven Design (DDD), identity is something that is inherent to an entity. It is something that uniquely identifies them from all other entities. For example, two people are considered different regardless of the attributes they possess. This can be confusing because when a director logs in, they are authenticated and become an entity.

Authentication is the act of informing that a specific identity is involved, so entities should also be identities by default. A director can have one or more identities (but only one at a time). An identity can have associated identity policies.

Identity resolution

is an Entity Identity and Information Management (EIIM) configuration where the input is identity information of the entity and the output is the identifier of the Entity Identity Structure (EIS) that represents the entity.

The knowledge structure created to represent a master data object is called an EIS. The storage and management of identity information and the persistence of entity identifiers are the added value that EIIM brings to Entity Resolution (ER). These systems have a reduction scheme, which produces a smaller list of “candidate identities” that are likely equivalent to the input reference. For systems that manage a large number of identities, it is not practical or necessary to compare the input reference with each identity.

Identity information is a collection of pairs of attributes and values that describe the characteristics of the entity, characteristics that serve to distinguish one entity from another. An emergency system can do this by storing the identity information it accumulates during processing, a kind of “intelligent fusion and purge”. The transmission of identity information is achieved by conserving (storing) the EIS that represents the entity. Other ER systems are designed to capture and store identity information as they implement an external view of identity.

Identity management in an emergency system occurs when the system is preloaded with identity information, as in the case of identity resolution systems, or when it retains all or part of the identity information of the entity from the references it resolves in the case of identity capture systems. If generously interpreted as “No entity that, in principle, is not identifiable”, this slogan is acceptable. While Entity Resolution Framework (SERF) focuses primarily on resolution efficiency, EIIM focuses on maintaining the integrity of the entity's identity through incremental resolution (update) cycles, including manual updating (affirmation). The entity's identity information will change and Master Data Management (MDM) system must have sufficient functionality to account for these changes.

As long as the identity identifier does not change, the same link value will be assigned every time a reference to that entity is presented to the system. Just like merge and purge process does not retain identity information, link identifiers it assigns to set of references it processes are transitory. Identity capture systems start with a clean slate; when they process your first entity reference, identity information from that reference is saved as first identity in system and an identity identifier is assigned to it.