For any given Use Case we need to have a list of all the different types of players. In the initial stages of the life cycle of a use case it is less relevant whether these players are called Stakeholders, Actors, Roles, User Types or Systems, we simply need to know what terms the actual users, the real customer(s) of the use case, are using for them in their daily practice.
We call them Personas.
For example, in the use case "Legal Entity Management" you would have Personas like:
- Data Owner
- Signator Power Pursuent to Commercial Register
- Board Member
- Legal Council
Many of these Personas are involved in multiple use cases and are grouped together in persona taxonomies and defined in ontologies with "machine readable meaning" that can be used to let algorithms "understand" the actual context and act upon it.
In traditional use case modelling people use the term Actor rather than Persona. The term Actor is also used in a TOGAF Metamodel and combined with the term Role whereby an actor assumes a role to perform a task.
The Use Case Tree Method does not make that distinction, primarily to keep things as simple as possible when capturing requirements --- primarily as stories --- but also because the term Persona combines both concepts into one whereby:
- Actors assuming a Role is fully automated, context dependent, model-, rule- and policy-driven
- Personas are Concepts tied to Ontology-defined Classes that can inherit from other Persona types
- Personas are not just the "users" (or systems) of the use case but also any other party in the related data-models (or EKG models/ontologies). For instance, your user can have the persona "Legal Entity Maintainer" but since the legal entity can have a Director as well, the Director is also a Persona, even when that Director might never be an active user of the system.
A subset of the model, as a UML Class Diagram, around Persona.