Lifecycle Ontology
Everything has a lifecycle: use cases, stories, concepts, personas, outcomes, data products and datasets. This ontology says what a lifecycle is, which lifecycle a thing follows and which state the thing is in. It depends on none of the ontologies that describe those things. A lifecycle has a name, states and permitted transitions, optionally grouped into phases. Its states and transitions form a directed graph; it does not require a separate RDF named graph. Multiple lifecycles can share an RDF graph, with inLifecycle expressing membership independently of graph placement. State order controls presentation, not which moves are permitted. There can be any number of lifecycles. This ontology defines none: the well-known lifecycles are a dataset (dataset/well-known-lifecycles.ttl), and anyone can define more. The registry uses a separate namespace for each lifecycle to distinguish shared state names such as Draft and Deprecated. This is an IRI naming convention, not a graph-storage requirement; custom lifecycles can use another IRI naming convention. Which lifecycles a thing follows is decided in this order: 1. The complete set of lifecycles the thing names itself, with followsLifecycle. 2. Otherwise, the union of lifecycles that apply to its classes (including superclasses), with appliesTo. A thing has one inState history head per effective lifecycle. Each head is a StateTransition occurrence whose currentState belongs to that lifecycle. priorStateTransition links earlier occurrences; the lifecycle is derived from each state, and optional provenance is recorded on the occurrence.
- Classes
- 5
- Properties
- 12