Datatype

Go up to Terms and Definitions

A datatype defines the value space, permitted representation, and permitted operations for data values in a defined representation context. Datatypes support structure and validation by identifying the kind of data value that an information element contains.

A datatype differs from a type. A datatype defines the representation and value constraints for data. A type identifies the semantic category of values, objects, model elements, messages, events, or expressions within a type system.

Datatypes support semantics when they preserve traceability to defined domain meaning. A datatype alone does not establish the full meaning of a data item. For example, tradeDate, settlementDate, effectiveDate, and terminationDate use the same date datatype, but each term has different semantics.

Datatypes appear in schemas, programming languages, databases, message specifications, interface definitions, metamodels, and ontology languages.

type that defines the value space, permitted representation, and permitted operations for data values within a defined representation context

Generalised from software engineering, schema-language, database, modelling, and type-system usage; specialised for use in the FX Demo Reference Architecture.

A datatype constrains data representation. A semantic type constrains domain interpretation. A robust architecture preserves the distinction between datatypes and semantic types.

date, decimal, string, boolean, and integer are datatypes. SettlementDate, NotionalAmount, CurrencyCode, LEI, and TradeIdentifier are semantic types that use datatypes for representation.