Actors, roles and assignments

This page is published under the terms of the licence summarized in the footnote.

All free-to-read materials at http://avancier.website are paid for out of income from Avancier’s training courses and methods licences.

If you find the web site helpful, please spread the word and link to avancier.website in whichever social media you use.

 

 

This paper picks up from where “social groups and social systems” left off

Contents

On actors in systems and meta systems. 1

On the assignments of actors to roles. 1

Abstraction of roles from actors. 3

Questions about actors. 3

Questions about UML use case diagrams. 4

On human activity systems as a special case. 4

 

On actors in systems and meta systems

The previous paper said the members of one social group can act in more than one social system.

Moreover, the members of one social group can act both in a social system and in a meta system that directs that social system.

(This is a principle of agile development methods, and surfaces also in the paper on hierarchies and networks)

 

The term “actor” can be used at both system and meta system levels

There are actors (or employees) who are hired to play roles in operational systems.

There are also actors (or architects) who define the operational systems.

It is possible for an actor to play both roles.

Some systems thinkers promote the idea that employee actors should act also as architect actors.

In other words, a person acting as an employee actor can step outside the operational system to act as an architect actor in the meta system.

The actor ought to be clear which role they are playing at any one time.

This paper is about actors hired to play roles in operational systems.

On the assignments of actors to roles

What are the core elements in a human social system? This table presents three views.

As seen by

The elements of an organisation are

related by

Common man

Human actors

authority or command and control relationships

Weber and Boulding

Assignments of humans to roles

messages containing information

Luhmann

Communication acts

references to the same “code”.

 

The view of Weber and Boulding is the one taken here.

A core element in a system’s description is role.

A core element in a system’s operation is not such much actor as the assignment of an actor to a role.

One role can be played by several actors, and one actor can play several roles, which is to say there are three concepts:

 

Roles

Enterprise architecture views a business as a social system whose roles are formalised.

A role is a type that defines what actors in that role do, and perhaps abilities required for that.

Roles are defined in terms of services offered and activities performed.

One role can be played by several actors, and one actor can play several roles.

The entity that relates one actor to one role is an assignment

 

Assignments

An assignment is the allocation of an actor to a role.

An instance of a role type is not a human, but rather, a human assignment to that role.

Of course, assigning human actors to roles is different from assigning machines to roles.

In the operation of a system, computer actors behave in an orderly, prescribed, way.

By contrast human actors are likely to do more, less or different from what is defined in any role given to them.

And in practice, a business depends on humans doing things that are not explicitly defined role definitions.

 

Actors

An actor in a system is an individual or object with the ability to perform one or more defined roles.

We design software components to perform required processes, and expect no more from them.

Of course, we don’t design human actors, we hire them (and sometimes training them).

And every human actor arrives with an immense range of abilities, outside what is required for any business role

Though hired to play one role, a human may choose to play a different role, or change their role or invent a new role.

Also, a human is amore needy than a computer; it is not sufficient to plug them in to a supply of electricity.

A human actor demands attention, in various ways, from managers, colleagues and others.

 

The table below paints a picture of a spectrum from behaviour-oriented system theory to people-oriented systems thinking.

Enterprise architecture & system theory                                       Systems thinking & organisation architecture

Processes involve roles which are instantiated in assignments performed by actors employed in organisation units

Abstraction of roles from actors

The systems of interest here are composed of business processes performed by human and computer actors.

Yet most of what actually happens in the operational system is never described by business system architects.

The architects represents only some behaviour of the operational system, that which has to be understood or assured.

All architects rely on the successful operation of computers’ circuit boards and humans’ internal organs (which are described separately by computer manufacturers and biologists).

 

Architects describe processes in which they abstract the roles played by human and computer actors.

So if one actor replaces another actor playing the same role, it makes no difference to the system.

 

This idealist view of a business system has a possibly shocking implication.

Most of what makes a human human, and a computer a computer, lies outside of any system as it is described.

Architects do not mention:

·         electricity supply, circuit boards, heating, lighting and ventilation

·         the muscles, blood, lymph systems and other organs humans need to keep them alive

·         almost all human hopes, dreams, frustrations, skills and knowledge.

 

It is untrue that abstracting roles from actors diminishes the part that people play in business success.

The reverse is true, that what people can do in the real world diminishes the application of system theory.

Since not only are people vital to business success, but much of what they do is way beyond what architects can describe.

 

Business systems descriptions only lightly describe human and computer roles – the actions they perform or services they provide.

These specified actions and services cover a small part of what computers are capable of, and a tiny, tiny part of what humans are capable of.

 

Enterprise architects (knowingly or not) presume humans will do far more than any role or process description declares.

So, it is vital to business success that humans are well motivated and managed.

However, this motivation and management of humans is not described in any enterprise architecture repository I am aware of.

This is the province of Human Resources, Business and Project Managers, Organisation Design, Business Change Consultants and so on.

Much of it is ad hoc, and much is part of the infinite complexity of the real machine – way beyond any business system description.

Questions about actors

There are human and computers actors both inside and outside any business in operation.

Business system descriptions usually feature the roles actors play, rather than the individual actors.

 

Q) What does it take for an actor to be able to play a role, outside or inside a system?

It takes whatever abilities the role requires.

The role definition for a human actor takes for granted a vast range of basic human abilities (breathing, thinking, talking).

The role definition for a software actor takes for granted the more limited basic abilities of a computer.

 

Q) What does it take for an external actor to interact with a system’?  

The ability to supply inputs and/or consume outputs.

Where the inputs and outputs are data flows, actors need the ability to code and decode those signals.

 

Q) How do we choose where to draw system boundaries around things outside of our system of concern? 

As the focus of our interest and attention shifts, so the system boundary may shift.

Given a system boundary, you can regard things in the wider environment as discrete external entities and activities.

There is no obligation to consider how those external entities and activities may be related in a wider system.

Questions about UML use case diagrams

A use case diagram shows processes, performed by a system, in which external actors are involved.

 

Q)  What is an actor in a UML use case diagram? 

The actor is usually a role; it is an actor type rather than actor instance.

A line connecting a role/actor to a use case says that the former is somehow involved with the process of the latter.

The involvement is often to supply input and/or consume output information.

The involvement can be specified more exactly by <<stereotyping>> the line between role/actor and use case.

  

Q) Why draw a whole person and not a disembodied hand in a trivial ATM use case?  

A human actor in a use case diagram is never ever a whole person.

It is only that tiny, tiny part of a human actor that plays the specified role in the specified use case.

In the ATM, it is not the hand, is that tiny, tiny part of the person whose intention is to make a particular use of the ATM.

 

Q) Are we making unspoken assumptions about intention and intentionality? 

You are supposed name a use case after the intention (check balance, withdraw money etc.) of actors in specified roles.

On human activity systems as a special case

Enterprise architecture views a socio-technical organisation as an ordered system.

That is, as a collection of actors assigned to play roles in repeatable processes to maintain system state and produce desired effects.

Do not think of a human as an instance of a role, rather, a human assignment is an instance of a role.

Of course, assigning human actors to roles is different from assigning machines to roles.

In the operation of a system, computer actors behave in an orderly, prescribed, way.

By contrast human actors do more, and less, and different, than is prescribed in any role.

In practice, social organisations depend on humans doing things that have not been explicitly defined in assignments to roles.

Human nature and flexibility

We don’t design human actors, we hire and train actors to play designed roles.

But in the case of human actors we get more than we designed for.

We get the amazing general-purpose components that biological evolution has created.

So every human has an immense range of abilities, countless more than are required for any business role

A human is more generally able and multi-talented than is required by any role we design.

In fact, a human is probably more complex than any system we design – let alone any role in it.

And being self-aware, a human can choose which of their diverse talents to exercise, regardless of any role they are given.

 

Of course, human actors may act outside any role given to them, or refuse to act as expected.

The system describer may allow for this in “exception paths”.

The system describer decides what range of choices and activities to include in the system and what to exclude.

 

One individual may play the same, different or competing roles in several social groups.

Though hired to play one role, a human may decide to play a different role, change their role or invent a new role.

 

By the way, a human actor is also demanding of attention, in various ways, from managers, colleagues and others.

(It is not sufficient to plug them in to a supply of electricity.)

Relying on human nature and flexibility

All businesses rely on human actors behaving in ways that are not defined in any system description.

It is assumed that intelligent people, being aware of an organisation’s goals, can do what is wanted in situations that arise.

In practice, business owners depend on the ability of humans to:

·         assess the rules in a role description

·         fills gaps in an incomplete description

·         interpret, improve or correct loose or badly-written rules

·         tailor general rules according to specific circumstances

·         ignore or defy given rules to meet what they perceive the "real" requirements to be.

 

Because of this, human role describers get away with informal, vague, even erroneous definition.

Much of what humans do is undocumented and unpredictable from any system description.

Individual actors may change the rules, processes and roles of the system they work in, and even change the goals.

Some may even pursue their own agenda, in conflict with the rules and purposes of the system that employs them.

Does the actor shape the role? Or vice-versa?

In a human activity system, the granularity of the human actors is a given.

So, the granularity of employable actors shapes the granularity of role descriptions.

In a computer activity system, the granularity of software actors is a matter of design

So, the granularity of roles shapes the granularity of the actors.

 

 

Copyright conditions

Creative Commons Attribution-No Derivative Works Licence 2.0             03/01/2016 20:57

Attribution: You may copy, distribute and display this copyrighted work only if you clearly credit “Avancier Limited: http://avancier.website” before the start and include this footnote at the end.

No Derivative Works: You may copy, distribute, display only complete and verbatim copies of this page, not derivative works based upon it.

For more information about the licence, see http://creativecommons.org