<< Prev | - Up - | Next >> |
The main part of the dialogue specifications is a graph that consists of the following types of nodes (we discuss the characteristics of most of them in the next sections):
The main part of the dialogue specifications is a graph that consists of the following types of nodes (we discuss the characteristics of most of them in the next sections):
Exactly one start node in which the dialogue starts.
End nodes: if the dialogue goes into such a state, it ends.
Output nodes, from which an output is sent to a device (See Section ``Output nodes'').
Input nodes, which wait for input from a device and then go to a new state (See Section ``Input nodes'').
Set-variable nodes, which set the value of a variable (See Section ``Variable nodes'').
Condition nodes, which branch depending on a condition (See Section ``Condition nodes'').
Test-variable nodes, which branch depending on the value of a variable (See Section ``Variable nodes'').
Subgraph nodes, which represent a complex subtask that is used only in one place in the dialogue. (See Section ``Subgraph and Procedure nodes'').
Procedure nodes, which represent a complex subtask that can be called several times, possibly with different parameters. (See Section ``Subgraph and Procedure nodes'').
<< Prev | - Up - | Next >> |