Generally, it is beneficial when mixed-initiative interfaces learn and adapt to individual users.

Because users' goals and situations change over time, the system is never "ready."

The feasible communication acts and their effects are conditioned by the state of the partner.

The paradoxical effect of hyperarticulation is that despite trying to improve understanding, it can make speech recognition worse.

When an automated action is taken, it is important to consider the timing, as incorrectly timed automated actions can distract the user.

If there is ambiguity about what the user wants and wrong automation might harm the user, the system should ask for more information or not carry out the command.

Since the system will be unlikely to always automate functions successfully, it is important that users can directly trigger and terminate functions.

If the system is uncertain about the user's intent, the system should ask the user after having considered the cost of interrupting the user.

If a system operates under a high uncertainty of the user's goals, the system should perform less automation to avoid interrupting the user with poor suggestions.

When there is a misunderstanding about the context of the dialogue, errors may happen, and the partners must recover from them.

If the supervising user wants to intervene, the gulf of evaluation becomes relevant.

The current understanding is that users form internal models that predict how their actions produce perceived outputs, and they learn to minimize prediction errors. This explains why people explore interfaces (to develop better internal models) and why, eventually, they no longer need to compare outcomes against goals.

Mixed-initiative interaction is the idea of organizing interaction in dialogue where both the computer and the human can take initiative. Unlike in the case of an FSM, the computing system can take action without a command from the user; the initiative is mixed.

Code-switching refers to a switch in language to match the capabilities of the communication partner.

A mode refers to the variation in the interpretation of a user's input according to an internal state. In a modeless dialogue, all inputs are possible in all states and their interpretation is always the same.

Formally, an FSM is a tuple (Σ, S,s0, δ, F), where: Σ is the input, that is, a finite set of symbols; S is a finite set of states or modes; s0 ∈ S is the initial state; δ is the state transition function δ : S × Σ→S; F is the set of final states, that is, a subset of S.

Gulf of evaluation: This gulf refers to knowing how a perceived change in the computer has moved it closer to the intended goal state. For example, imagine setting the temperature of an intelligent thermostat and not perceiving an immediate effect. How can you tell if your command had the desired effect on the system?

Gulf of execution: This gulf is about knowing what to do to bring about a desired state change in the computer. For example, what should you do to get a piece of text copied to the clipboard and pasted in a specific location?

The key idea in the dialogue view of interaction is the organization of communication as a series of turns. Dialogue evolves through communication turns between two or more partners. In one turn, an appropriate communication act is made by one partner based on the communication context. The act aims to get the other partner to do or understand something. This understanding then forms the context within which the other partner takes their turn.

prompt engineering—the search for prompts that give the output the user finds adequate for the task.

Visibility is a handy related concept in design that underlies direct manipulation interfaces.

Affordance, which we discussed in Chapter 3, refers to how well users can interpret what actions are possible with a widget.

Mapping requires the user to figure out how to accomplish a goal with an interface.

An FSM is a model of discrete computation applicable to dialogues. In computer science, an FSM is a special case of a Turing machine that reads but does not write on the tape.

Communication repair refers to the "work of restoring shared understanding" when conversational partners misunderstand each other.