Simply put, contracting muscles cause your body to move. These skeletal muscles are connected to your bones by nonelastic, fibrous tissue called tendons. Your muscles move parts of your skeleton by pulling, via the tendons, on skeletal bones to make them move. Muscles can only pull, not push, and thus can't push themselves back into their original positions. Only another muscle pulling it back into its initial shape can do that.

Figure 2-1. Muscles pull on the bones of the skeleton, causing movement. Courtesy Gray's Anatomy.

Thus, there are two kinds of muscles: agonists and antagonists. Agonists (sometimes called prime movers) are the muscles that generate most movement. The biceps of the upper arm is an example of an agonist: your arm curls because the biceps contracts. An antagonist returns the limb to its original, initial position. As its name suggests, it acts in opposition to the agonist muscle. The triceps is an example of an antagonist. It returns the arm from the curled position into which the biceps moves.

This type of pairing of biceps and triceps is called an antagonistic pair. The pair is made up of a flexor muscle, which closes the joint by decreasing the angle between the two bones, and an extensor muscle, which does the opposite, opening the joint by increasing the angle between the two bones.

Joints are what allow humans to move at all. A joint is the location at which two or more bones make contact, held together by ligaments, which are short bands of flexible, fibrous tissue. Ligaments can actually prevent or limit certain movements of the joint.

Joints can be categorized functionally by the amount and type of movement they engender. For our purposes, we're mostly concerned with what are called diarthroses. Diarthrodial joints are those that permit a variety of movements and are flexible thanks to a pocket of lubricating synovial fluid in the space between the bones. There are six types of diarthroses:

These joints are the fulcrums on which all the major movements of the body occur. The only other joints we should be concerned with are those between the vertebrae in the spine, connected by pads of cartilage, which can move slightly, allowing for bends and twists of the whole torso.

The medical and kinesiology communities have technical terms for the types of movement the human body is capable of. Frequently, these movements come in pairs: first, there is the execution of the movement, and then the opposite movement that returns the body or limb to a neutral state. The human body is capable of the following broad movements:

In addition to these major, broad movements, there are specialized movements that only the hands and feet can perform. These, too, often come in pairs:

The last major movement is one that combines several movements—flexion, extension, adduction, and abduction—into one. Circumduction is the circular movement of a body part such as the wrist, arm, or even eye. Rotating the foot at the ankle is an example of circumduction.

We can use these technical terms to more precisely describe nonstandard movements in order to document those movements (see Chapter 5).

In 1955, Henry Dreyfuss wrote a seminal book on industrial design called Designing for People (Allworth Press). In it, as well as in a book he coauthored with Alvin R. Tilley, The Measure of Man and Woman: Human Factors in Design (Wiley), Dreyfuss demonstrated how the human body should be considered when designing products. To do so, he created two composite "people," Joe and Josephine, who illustrated the common dimensions of human physiology.

Designers need to be aware of the limits of the human body when creating interfaces that are controlled by it. The simple rule of thumb (pun intended) is this:

Almost anyone (see DESIGNING TOUCH TARGETS later in this chapter) can tap a button on a touchscreen. But not everyone is going to be able to wave her left hand while balanced on her right leg! Care should be taken so that anyone who will use the system can follow essential commands. If the users are the general populace, on a system they don't use often (such as a public kiosk) the interactive gestures should be simpler than those used by expert teen gamers on their home gaming system. Motor skills decay as we get older, making it more difficult to perform physical tasks, especially those requiring fine motor skills in the hands.

Designers of gestural interfaces need to know the same things Henry Dreyfuss did: the general capabilities of the average human.

Adult fingers typically have a diameter of 16 mm to 20 mm (0.6 inches to 0.8 inches).^[23] Children's and teens' fingers may be smaller, whereas disabled, elderly, or obese people may have misshapen or larger fingers.

For pushing buttons or touching screens, usually the pad of the finger is used instead of the tip. Fingertips are narrow, only 8–10 mm (0.3–0.4 inches) wide. Because of this small surface area, humans usually push buttons at an acute angle using the pad of the finger, not straight on using the tip of the finger. Finger pads are wider than fingertips, but narrower than the full finger, typically 10–14 mm (0.4–0.55 inches).

Fingernails are a blessing and a curse. Long fingernails, although unlikely to scratch a touchscreen, can make it difficult to perform certain gestures. Made of keratin, they are hard and smooth enough to mimic a stylus and are thus more accurate than a fingertip or finger pad. However, fingernails may not provide enough contact with some screens to trigger a touch event, especially capacitive touchscreens. This is especially true of fake fingernails.

Gloves, too, can make it difficult to use gestural interfaces. Not only do they usually inhibit hand movement and increase finger size, but they also may not trigger capacitive touchscreens because they do not conduct electricity. In climates where cold weather often necessitates wearing gloves, care should be taken when deploying gestural interfaces.

Although the exact percentage is unknown, studies have suggested that 7% to 10% of adults are left-handed, meaning they favor their left hand for tasks such as writing and, one can surmise, would do so as well with interactive gestures. Although obviously most gestural interfaces should be optimized for the 90% of right-handers, southpaws should not have to be unduly burdened, such as having to reach across the screen to the righthand side constantly to perform actions. A means to flip the controls or adjust the layout for lefties can be a good design practice.

Fingers also have natural oils and/or can get slippery, which can make it tricky, if not impossible, to manipulate things. Touchscreens, too, can get oily and worn, making them difficult to use. Finger oil (and dirt) also means fingerprints and smudges. Patterns or bright backgrounds can help hide them. Black—seemingly the default choice for many manufacturers—makes fingerprints even more noticeable.

Another ergonomic problem arises with gestural interfaces in that, because there is typically no wrist support, using one for a long period of time can be tiring on the hand and fingers. This is especially true with touchscreens, because often, significant pressure is required to engage them, and the screens, being made of glass, don't bend or give.

Fingers are also just messier and more inaccurate than cursors are. As reported in the New York Times Magazine:^[24]

Designers and developers thus need to remember that human fingers are an imperfect input device and account for this "flaw."

Finally, because, unlike a mouse cursor, our finger pads don't float transparently in space (unless they're Ghost Fingers—see Chapter 3), the rest of the finger, the hand, and the arm will likely cover up some part of the interface while the user is touching it, especially the part of the screen immediately below what the user is interacting with. Thus, it is a good practice to keep the following warning in mind.

Figure 2-8. An illustration of the difference between designing for a cursor and designing for a hand. The hand covers up not only the labels on the slider, but also the display showing the color that is being created.

Placing menus and controls at the bottom of the screen instead of in their traditional place at the top is helpful to prevent screen coverage.

Hands can also interfere with the technology of touchscreens, causing poor readings and accidental touch events if users rest their palms on the screen while performing a task. For some touchscreens, the solution to this is straightforward: mount the touchscreen vertically so that users cannot rest their palms on its surface. This won't, of course, work with mobile devices, but it is a solution for kiosks and displays.

All of these human factors have some interface implications, especially when designing touch targets.

Obviously, some gestural interfaces will be extremely difficult for those with physical challenges to operate, especially for those with limited hand movement. Although no guidelines exist yet, to be accessible to the widest possible user base, touch targets will have to be large (perhaps 150% of the typical size, or 1.5 cm square) and gestures simple and limited (taps, waves, proximity alone). Anything beyond some basic movements (tap, wave, flick, press) that use single fingers or the whole hand as one entity may be challenging for some disabled users. Even patterns such as Pinch to Shrink can be tricky.

On the positive side, some gestural interfaces might be more accessible than most keyboard/mouse systems, requiring only gross motor skills or facial recognition to engage them.

The body is an amazingly flexible tool with a wide variety of gestures. The hands alone have the potential for hundreds of possible configurations that could be used to trigger a system response. It is just up to designers to employ them correctly and not to overstrain users.

The next two chapters examine gestural interfaces that combine some of these physical gestures with system behaviors to form a pattern. Chapter 3 examines those patterns for direct manipulation, and Chapter 4 details those for indirect manipulation.