Cohesive Systems logoCOHESIVE SYSTEMS

Value

A value is pure structured data. It is the concrete information used to read, write, transmit, compare, validate, transform, or carry state.

Values are identity-free, version-free, lineage-free, and timeless. They do not carry subject, observer, source, occurrence, or provenance. Those belong to contextual concepts such as observation, event, entity state, state records, histories, and lineage metadata.

A value may be:

  • Empty or absent, such as null or undefined.
  • Scalar, such as boolean, integer, decimal, string, temporal value, or bytes.
  • Composite, such as array, vector, map, dictionary, record, object, or another structured aggregate.

A value may have a shape or representation. The shape describes the logical structure expected by a model or operation. The representation describes how the value is physically stored or accessed, such as dictionary-backed fields, ordinal record layouts, sparse buffers, packed presence bits, columnar storage, or struct-of-arrays storage.

Representation is not part of semantic identity. Two values with the same logical content are the same value even if one is dictionary-backed and another is layout-buffer-backed. Optimized representations must preserve logical equality, validation, serialization, and transformation behavior.

Values relate to the other semantic concepts as follows:

  • State is condition; value is the concrete information used to read or write that condition.
  • Observation is a value with context: subject, shape, source, version, observer, or lineage.
  • Event is a value with occurrence.
  • Command is an input event interpreted as an attempted transition.
  • Query is an input event interpreted as a request to observe, compute, or return information.
  • Entity state is an entity-scoped observation: a value attributed to an entity identity at a version.
  • A transition consumes values and observations in a transition context and may produce an accepted endogenous event or state change.

Values may be complete, partial, or projected only relative to a declared shape, model boundary, or operation. Completeness is not intrinsic to the value itself.

Related concepts: shape, state, observation, event, command, query, transition, entity, observable, boundaries, equivalence vs equality, naturality.