The Design of Everyday Things

Chapter 4 focuses on the use of constraints in design. Designers can guide user behavior by combining “knowledge in the world,” such as cultural, logical, and semantic constraints, with physical constraints or explicit design cues, all of which help users determine how to use products.

Four Kinds of Constrains: Physical, Cultural, Semantic, and Logical

This section describes four types of constraints, all of which limit possible operations. The first, physical constraints, are most effective when they are easy to see and interpret. Norman uses the example of batteries to explain this point. Cylindrical batteries lack sufficient constraints because they fit inside battery compartments in two directions (one correct, the other incorrect). They also lack signifiers, making it difficult to determine the proper orientation. Norman recommends designing batteries that make orientation irrelevant, inventing contacts that allow batteries to be inserted in either direction, or creating batteries that only fit into compartments in one way. He argues that the current design persists because of a “legacy problem”—namely, that too many manufacturers have adopted it and made it the standard.

Culture presents constraints by limiting possible actions in certain social settings. Violating cultural norms, such as facing the rear of an elevator, makes people uncomfortable. Cultural conventions are a form of cultural constraint associated with how people behave. These conventions are culturally-specific and can change over time.

Semantic constraints rely on the meanings of situations to guide possible actions. For example, a person must sit facing forward to use a motorcycle. This type of constraint relies on external knowledge. Like cultural constraints, semantic constraints can change with time.

Logical constraints rely on deduction. If one finds a leftover part after repairing a leaky faucet, for instance, one can deduce that the part should have been installed. Natural mappings also provide logical constraints by underscoring the logical relationship between the spatial and functional layout of components. For example, if two light switches control two lights, the left switch should operate the left light, while the right switch should operate the right light.

Applying Affordances, Signifiers, and Constraints to Everyday Objects

This section focuses on the importance of visual cues in the design of everyday objects. Returning to the example of doors, Norman argues that a well-designed door must have visible signifiers telling users how to operate it. Proper hardware, such as a knob or a plate, not only allows the door to operate smoothly, but also indicates how to operate it. In Norman’s view, modern cabinet doors with no handles stress aesthetics over functionality.

Norman presents constraints as necessary features of good design. A row of identical switches that operate different lights lacks constraints, forcing users to memorize which switch controls which light. Natural mappings do away with this problem. Norman describes applying this technique to his home, mounting a miniature floorplan of his living room on a plate, orienting it to match the room, and placing each switch in the area it controlled. He also angled the plate to make it easier to see.

Other solutions, such as touchscreens, allow excellent natural mapping, but lack the physical affordances of conventional switches. As Norman notes, a screen requires specific hand gestures to operate, not just a free elbow. Similarly, activity-centered controls, such as switches programmed for videos or lectures, lack flexibility and only work if the programming closely matches the requirements.

Constraints That Force the Desired Behavior

“Forcing functions” are physical constraints that prevent progress or inappropriate behavior. Having a key, for example, is the forcing function associated with starting a car. Safety engineers have devised three types of forcing functions: 1) interlocks, or mechanisms that force operations to occur in the proper sequence; 2) lock-ins, which prevent users from prematurely stopping operations; and 3) lockouts, which prevent actions or events from occurring.

As Norman observes, forcing functions can hinder normal usage, prompting some people to disable them. Good design minimizes the nuisance value of forcing functions, while retaining their safety features. For example, designers place gates blocking stairwells on ground floors to prevent people from mistakenly going to the basement during a fire. However, the gates are generally easy to open and do not impede normal usage.

Conventions, Constraints, and Affordances

This section addresses the gap between perception and understanding. Affordances, or potential actions, are only discoverable if they are perceivable, but perceptibility does not guarantee understanding. Designers rely on cultural conventions to bridge this gap. A doorknob, for instance, has the perceived affordance of being graspable, but it is the knowledge that doorknobs open and close doors that makes it functional. It is the interpretation of a perceived affordance, then, that is a cultural convention.

Cultural conventions function as constraints. For example, eating conventions vary between cultures, with some using utensils and others fingers and bread. Moreover, what is considered proper in one culture may be impolite in another. Conventions can change, but not without resistance. For example, designers met with pushback when they created the destination-control elevator system, which groups users going to the same floor. Keypads are located outside the elevators, while arrows direct people to the elevator that will most efficiently reach their floor. This system is faster than conventional elevators because it minimizes stops, yet developers were slow to adopt it. Designers meet with resistance whenever they violate conventions, even if new systems are superior. Changing conventions requires learning new skills and investing in new materials. In some cases, however, the benefits of a new system outweigh the difficulty of change.

The Faucet: A Case History of Design

Norman uses case studies to explain and analyze design problems throughout his book. In this section, he describes the unique challenges of designing faucets.

Water enters a faucet through two pipes, one hot, the other cold, creating a direct conflict with flow control. Designers have developed several solutions to this problem, including separate controls for the hot and cold water, controlling the temperature with one control and the rate of flow with another, and having a single control for the temperature and flow rate. Each solution presents design problems, some of which are culturally-specific. For example, regardless of their function, knobs in the United States must be turned counterclockwise to loosen and clockwise to tighten, but the opposite is true in England (151).

Mapping problems also exist. In cases with two knobs, for instance, users must determine which one controls the hot water and which one controls the cold water. Even single-spout, single-control faucets have mapping problems related to the dimensions of control affecting temperature and flow rate, and the direction of control affecting temperature. In other words, single-control faucets prioritize aesthetics, while neglecting functionality by lacking visible affordances, signifiers, and discoverability.

Norman recommends standardization to eliminate or minimize these problems. Although he acknowledges that standardization can hinder future development by stifling creativity and innovation, standardization can also provide the foundation for new ideas. Over time, innovations become the standard and serve as the base for the next innovation.

Using Sound as Signifiers

Designers rely on sounds to help people use products. Sound signifiers include the whistle of a teakettle when water boils and the roar of a muffler with a hole. Sound signifiers must be detectable even when users are inattentive. They must also convey information about the actions taking place. As useful as sound signifiers are, however, they are not without shortcomings. Sounds are often intrusive and can distract as easily as they can aid. In addition, sounds are difficult to keep private and can disrupt others.

Norman explains that sound is an important signifier of presence, making soundless products potentially dangerous. Electric cars, for instance, are so quiet that pedestrians do not hear them coming. Marketers have experimented with adding artificial sounds to electric vehicles. Porsche added a throaty growl to its electric car prototype, while Nissan experimented with tweeting birds for its hybrids (158). Advocates for people who are blind have lobbied for “skeuomorphic designs”—that is, making electric cars sound like gasoline cars.

Designers continue to experiment with automobile sounds as lawmakers and industry specialists grapple with standardization protocols. As Norman notes, sound signifiers must meet several criteria: 1) They must alert people to the presence of cars; 2) They must allow people to determine where cars are located and provide a rough idea of speed and directionality; 3) They must not be annoying; 4) They must be standardized.