For running a model, we need an infrastructure. Every process in the model needs its own facility on this infrastructure. We call this the processes' engine which is 'running' on the infrastructure.

The functionality of the engine is provided by the 'backbone' at the processes' 'integration'point. This functionality is stated by the backbone functions. It is this functionality that is the outcome of the Proto-6 investigations.

There are two more things needed before the process can 'work'.

First we must be able to 'see' the process. The process must be 'visible' from outside the model. Therefore we use an interface to the model. There must be an interface for a process as long as we want to 'see' it. A process without interface is useless, it doesn't communicate with its environment.
The interface must of course be known by the backbone, where it is enabled by the infrastructure. And it acts as an interface through its interface functions.

An engine with an interface might be quite interesting, but it doesn't do anything. So, in the second place, we need something to run the engine. In the Interactive Model Architecture this is done by 'programmed' tasks, which can be sent to an engine as a workorder. In this manner, a workorder can be sent from engine to engine ordering them to perform the included tasks.
The running task must of course be known by the backbone, where it is enabled by the infrastructure. And also, a task has its typical functionality through task functions.