Skip to the main content

Click for FAST Consulting's home page
FAST Consulting
FAST Consulting Home Web UIs Desktop UIs Services Websites Colleagues About Us Lesson Plans Amusements
FAST Consulting

Introduction
Background Color & Pattern
Pick the Background First
Foreground Color & Pattern
Avoiding Problems with Adjacent Colors
References
Resources
Definitions
FAST Consulting

You are here: Home ~ Color & Pattern

Color and Pattern

Adapted from The GUI Style Guide by Susan Fowler
and Victor Stanwick, 1995, Academic Press.

Introduction: Why Use Color?

Color is more interesting than black and white. The advertising industry confirms that color is strikingly successful. According to Starch Tested Copy, a newsletter of the market-research firm Starch INRA Hooper, the average one-page full-color ad in a business publication earned "noticed" scores 45 percent higher than the average one-page black and white ad (Starch 1992).

Color can alert users to problems or to changes in system states quickly. Network troubleshooting software often uses red for overloads and crashes, yellow for problem spots, and green for normal situations. (The hotter, brighter colors are used for the most important information.) A system monitoring pH levels might use yellow for acidic solutions, dark blue for alkaline solutions, and shades of green for the ranges in between.

Color coding shows relationships quickly. Assigning "false" colors to particular types of information adds extra dimensions to the underlying picture. Pseudocolors are used to indicate temperatures and types of reflected light (including invisible infrared) in satellite images and weather maps; to indicate hot and cold spots in medical CAT and MRI scans; and to separate layers--plumbing, electrical, indoor surfaces, outdoor surfaces--in architectural drawings (Banks 1992, 144).

However, even in mundane situations, color can be used to link information. For example, accounting systems often show overdrafts and negative amounts in red, making it easy to spot all of the problem points at the same time.

Color coding shows differences quickly. In the same way that color-coding helps users see relationships, colors help users separate the unrelated items.

When searching, for example, a user can scan for the hits (the found text) much more quickly if they appear in a different color from the normal text. (Note that you should reverse the video rather than simply adding color because colored text is harder to read than black or white text.)

However, before you color-code your software, note that there are four coding rules (Banks 1992, 146-147):

  • Color coding is useful only if the user knows the code. Red for port (left) and green for starboard (right) will make sense to boat owners and airplane pilots, but not to car drivers.

  • The advantage of color increases as clutter increases. In an uncluttered display, color adds nothing to performance. In a complicated, high-density display (60 items), however, color can reduce search time by 90 percent.

  • Average search time increases linearly as the number of items using the same color increases (by 0.13 seconds per extra three items). In other words, you lose some of your color advantage if too many items have the same color.

  • If you're looking for something of a particular color, having items with other colors on the same display has no effect on search time if the other colors are sufficiently different from the target color. In other words, since red is very different from yellow, no user will pick a yellow triangle when she's looking for a red triangle. However, she might mistake an orange triangle for a red one since orange and red are too close, especially for red-blind individuals (see Color Confusion).

Why Use Pattern?

"If you generate a cash-flow graph for [security] FH 1080, it is difficult to delineate the PACs. Even worse, when you print it out, it looks like three black cows on a dark night." Neill Reilly, director of sales, EJV Partners, New York.

Despite color's appeal, color is not enough. In fact, color should be used second, pattern first. There are three reasons, all having to do with hardware (human or machine):

  • Not everyone has a color printer. Say that you have designed a lovely chart with four or five colored lines for your stock valuation tracker. When you print it on a black and white laser printer, what do you get? A lovely chart with four or five undifferentiated black lines.

  • Personal digital assistants and web phones. Many web pages are being sent to personal digital assistants (PDAs) and web-enabled cell phones. Eventually the manufacturers will sell color displays, but they're not available right now.

  • Color blindness, which is better described as "color confusion" or "color weakness."

Color Confusions

Approximately 8 percent of all males and 0.5 percent of all females have a color blindness (Hackman 1992, 653). Color blindness or weakness has four basic varieties:

  • green blindness--individuals confuse greens, yellows, and reds (6.39 percent)

  • red blindness--individuals confuse various shades of red (2.04 percent)

  • blue blindness--individuals confuse blues (0.003 percent)

  • total color blindness, which affects no more than 0.005 percent of both sexes.

"Color blindness" is actually a misnomer, since color-blind individuals see all of the colors in the spectrum, not just black and white or shades of grey. Color-blind individuals simply confuse certain colors. For example, a person with a red confusion might label a pale-green item as tan or orange. A person with a green confusion might label dark blue as purple or yellow as bright red (Milhaven 1989, VC16-19).

If you have standard color vision, you can reproduce the effect, although not the actual confusion, by looking around through a pair of deeply saturated colored sunglasses, a colored filter or theater-light gel, or half of a pair of 3-D glasses (close one eye). If you're looking through a red filter, for example, everything red looks white, and everything blue or green looks black. The reason is that only red light can pass through the red filter. Since pure blue and pure green contain no red light, they don't get through the filter: "no light" equals "black."

Hint: If you offer 3-D glasses with your software, use polarized glasses or red and blue instead of red and green. Red and green glasses don't work for people with red/green confusions, who are the majority of those affected.

Go to top of page

Background Color & Pattern

With the stumbling blocks of black and white displays and printers and color-blind users, how do you pick the right colors?

The answer is that you don't. Don't pick the colors first--pick contrast and pattern first, then pick colors. The next few sections explain how.

Start with Black and White

The development platform guidelines recommend that you design the objects in your application in black and white, and then add color. This guarantees, they say, that the icons and windows will work just as well in black and white as in color.

Un-Design for Clarity

Today the competition is at the user interface.... Skillful visual design of computer screens--with care given to color, typography, layout, icons, graphics, and coherency--substantially contributes to quality and usability. Poor screen design can destroy underlying excellence in software and hardware. Graphic design details are not cosmetic matters or decorative touches. In fact, careful attention to visual craft is a distinguishing characteristic of nearly all excellent user interfaces now in the marketplace. Edward Tufte, in introduction to Visual Design of the User Interface, written for IBM (Tufte 1989, 1)

All of the development-platform guidelines recommend against overusing color. Edward Tufte, in a report he prepared for the IBM Design Program (Tufte 1989) and in his own books (Tufte 1983, 1990), points out that less is more.

"In the simplest case, when we draw two black lines on a white surface, a third visual effect results, an active white stripe between the two lines," he writes. "Nearly all the time, such surplus visual activity is disinformation, clutter, noise. This two-step logic--recognition of 1 + 1 = 3 effects and the consideration that such effects clutter information displays--provides a powerful tool for editing and refining user interface designs."

Two fat black lines create a fat white line in between them.

Figure 1. 1 + 1 = 3

Any first version of a window or an icon will contain a lot of noise. But by studiously eliminating extra patterns, using gray instead of black lines, or eliminating lines altogether, you can reduce the noise and bring information forward. For example, see the three iterations of the chart in Figure 2.

Chart with a bad headline (all caps, hard-to-read type), bars with gradients, and three different colors for the bars and a pink background.

 

Chart now has a title with mixed-case lettering but it's in italic. It also has a gray background, and black and white bars with three different types of hatching.

 

The final chart has a pleasant sans serif title and the bars are now white, (with black outlines), gray, and black. The background is white.

Figure 2. Three steps to uncluttering a window

Go to top of page

Pick the Background First

Some designers try very hard to come up with imaginative ideas in an attempt to design attractive furniture for offices. They visualize pitch-black office machines on a bright table or dark furniture neighboring bright walls. Such designers don't care about ergonomic principles or balanced surface luminances. Etienne Grandjean in Ergonomics in Computerized Offices (Grandjean 1987, 46).

There are two types of background:

  • the environment of the computer or workstation itself

  • the application background--the application window or the video screen.

If you know anything about your typical user's workplace, you can easily design your program's backgrounds and foregrounds to accommodate your users' visual situation. If not, you can still pick a range of contrasts that reduce users' discomfort while they're looking at your windows.

Environmental contrast

Etienne Grandjean is one of the inventors of ergonomics and human factors. From 1950 to 1983, he was director of the Institute for Hygiene and Work Physiology at the Swiss Federal Institute of Technology in Zurich. From 1961 to 1970, he was general secretary to the International Ergonomics Association.

In Ergonomics in Computerized Offices, Grandjean describes the ill effects of high illumination levels on computer users. Although glare, reflections, and deep shadows on the monitor are often extremely irritating, an unrecognized eye-strain culprit, he concludes, is too much contrast between the foreground (computer monitor) and the background (everything else--desktops, walls, curtains, windows, and so on).

Two computers with a cubicle wall between them. One has a dark screen and sits in shadow. The other has a bright screen and sits in light.

Figure 3. Why office-mates fight over the light switch

The human eye adapts so well to shifts in light and dark that we can see nearly as well in moonlight as in the brightest sunlight, even though the illumination level differs by more than 100,000 times. However, dark adaptation takes a relatively long time--25 minutes to reach 80 percent adaptation, an hour for full adaptation. Light adaptation is quicker than darkness adaptation--a reduction in light sensitivity by several powers of ten in a few tenths of a second. However, light adaptation involves the entire retina. Whenever a bright image falls on any part of the retina, reduced sensitivity to light spreads to all parts of the retina. Since this adaptation includes the fovea, visual acuity for reading or fine details drops (Grandjean 1987, 23-24).

Since office designers usually recommend lighting levels of over 1,000 lux, luminance readings in typical modern offices can cause severe acuity problems (leading to eyestrain, headaches, squinting, wrinkled brows, and so on). A thousand lux is a lot of light: The range of light in a typical living room is 50-200 lux. Bright sunlight is 50,000-100,000 lux and a well-lit street at night is 10-20 lux. Reflected light is about 60 percent of the original brightness, so overhead light of 1,000 lux reflected off an office wall could be as much as 600 lux (Millerson 1991, 11).

Research indicates that the ratio in brightness between the video screen and a document from which the user is typing should be no more than one to ten (1:10); the ratio between the video screen and the immediate background (walls, desktop, and so on) should be no more than 1:20; and the ratio between the video screen and the entire room should be no more than 1:40. The reality is much different. Researchers who surveyed 109 VDT workstations found a ratio of 1:10 to 1: 81 (average 1:21) between the screen and source documents, and a ratio of 1:87 to 1:1450 (average 1:300) between the screen and nearby office windows (Grandjean 1987, 43).

In short, to avoid dazzling and prevent eyestrain (your own or your users), try to keep all surfaces at the same brightness by matching the overall brightness or dimness of your screens to the office environment. If users typically use your software in brightly lit offices or if they type from bright source documents, use light or off-white backgrounds. If your software is used in dim or dark areas (air traffic control towers, for example), use dark backgrounds. If you can't find out what the background will be, you might want to include two backgrounds, one light and one dark, with your program.

Note: If you design for a dim background, keep in mind that some color-blind users, who have no trouble distinguishing red from green or red from orange in bright light, cannot distinguish between the two colors in dim light (Kunz 1987, 315).

Contrast and focus

As Figure 4 shows, the area on which you can focus is about one inch at 20 inches (1° angle of view). You are aware of text or images within a circle about a foot in diameter (1° to 40° ), but can perceive only movement outside that circle (41° to 70° ).

A man with three cones in front of his eyes indicating the visual field.

Figure 4. Visual field: cone A is an angle of about 70°, cone B is about 40°, and cone C is about 1°

The rules for contrast on the screen, or between the screen and its immediate surrounding, take into account the size of the visual field as well as the dazzle effect described in Environmental contrast (Grandjean 1987, 41):

  • Surfaces in the middle of the visual field (around C in Figure 4) should not have a brightness contrast of more than 1:3.

  • Contrasts between the central and marginal areas (between A and B in Figure 4) should not exceed 1:10.

  • The working area should be brighter in the middle and darker in the surrounding field.

  • Excessive contrasts are more troublesome at the sides than at the top of the visual field.

Color inside the eye of the beholder

Once you've picked the overall background for your program, you're ready to pick the color scheme for the windows, buttons, and icons in your software. However, colors have some odd characteristics, due to interactions between the physiology of the eye and the physics of light. In short, different wavelengths of color come into focus at different points in the eye.

Since yellow and green wavelengths come into focus at the retina, they require the least accommodation from the eye (this is the reason for so many yellow and green monochrome monitors a few years ago). Red wavelengths, on the other hand, come into focus a little behind the retina and therefore seem to "pop out" of the background. Since blue wavelengths come into focus in front of the retina, blues seem to fade into the background.

Three eyeballs. The top one shows red focusing behind the retina, yellow and green focusing on the retina, and blue focusing in front of the retina.

Figure 5. Wavelengths focused hither and yon

So, when you choose colors, remember that:

  • Your eyes cannot focus clearly on blue, which is why it is such a good background color and such a bad foreground color.

  • Nor can your eyes focus well on red, but red has the advantage (if you need it) of "moving forward" in the visual field.

  • Yellow and green are just as visible in the periphery as they are as in the center of the visual field.

  • Black and white are equally visible throughout the visual field (Horton 1991, 228).

Other interesting effects include (Horton 1991, 227-228):

  • For most colors, hue seems to change as luminance increases or decreases. However, saturated blue, green, and yellow remain constant throughout the range of luminance. Use them when constancy is important.

  • Staring at a large patch of a saturated color for a long time shifts color perception towards its complement. For example, when you look up after working on a bright red figure, everything will look greenish. Called the "McCullough effect."

  • In bright light, red seems brighter than blue. In dim light, however, blue appears lighter but colorless, while red appears nearly black. In low-light situations, avoid reds. Called the "Purkinje effect."

Variation: Backgrounds and foregrounds for presentations

Use dark backgrounds and light foregrounds (text, lines, and so on) for long-distance, low ambient-lighting situations like slide shows or projected computer presentations. If you're creating a video presentation (live action or cartoon), use colors with low saturation (Marcus 1992, 84). Red, especially, blooms and bleeds all over the pictures in which it appears, especially after you've copied the video tape once or twice.

Use light backgrounds and dark foregrounds for situations with high ambient light--for example, when you're using an overhead projector (Marcus 1992, 84).

Go to top of page

Foreground Color and Pattern

Color in the foreground parts of an application--charts, icons, toolbars, and so on--must contrast with the muted background of the application. However, bright colors bring up these issues:

Hint: When you start to create icons and the rest of the foreground objects, avoid temptation. Don't work directly on the computer with all its multitudes of colors and brightnesses. Instead, start by sketching your pictures on a paper napkin, a cafeteria placemat, the back of an envelope--anything that keeps you working on the shape of the picture rather than its colors. Put the picture into the computer only after you're satisfied it that stands on its own (by testing it on your colleagues, friends, relations, and clients, for example).

Pick meaningful colors

Although picking culturally correct colors for the overall interface is useful (cool gray for an accounting system, hot pink for a Post-Modern game), selecting meaningful colors for the signals inside your windows is even more important. There are four issues:

Chunking: People can remember the significance of only seven colors, plus or minus two. In other words, don't create line charts with nine differently colored lines (unless the entire pattern of lines is the significant picture--for example, to indicate a noise level or a confusion level, as in Figure 6).

A chart with chaotic lines, but one jumps out of the normal range.

Figure 6. Intentional noise and confusion--chaos is normal, so the one out-of-range point shows up dramatically

Actual or spurious relationships: Since people automatically assume that items colored the same are related (Galitz 1993, 429), you must color related things the same, unrelated things differently. Don't just add color, in other words. Aesthetics are not as important as sense.

But you can take advantage of this automatic-association facility: For example, if your application has many charts, each of which contain the same types of data, color-code the data types. In a loan-analysis program, say, you could use green for principal payments, blue for interest payments, red for defaults, and so on. As well as simplifying the situation for your clients, the programming staff won't have to reinvent the color wheel every time a new chart is added to the program.

Task domain expectations: Find out how color (or pattern) is used in the area for which you're designing the program. "The designer needs to speak to operators to determine what color codes are applied in the task domain. From automobile-driving experience, red is commonly considered to indicate stop or danger, yellow is a warning, and green is go. In investment circles, red is a financial loss and black is a gain. For chemical engineers, red is hot and blue is cold. For map makers, blue means water, green means forests, and yellow means deserts" (Schneiderman 1992, 327).

Hint: When you ask about color, ask about relative position as well. For example, color-blind individuals in the U.S. use the mnemonic "Stop on top, go below" for stoplights. If you design a dashboard with red, yellow, and green lights in the wrong order or organize them horizontally, at least eight percent of your audience will get the lights wrong. See Color Confusions for details.

Cross-cultural differences: Colors mean different things in different cultures. For example:

Green and orange

politically suggestive in Eire and Northern Ireland

Red

suggests death in many African cultures

Red, white, and blue

suggests colonialism in some countries

White

suggests death or mourning in some Oriental cultures (Apple Computer 1992a, 219)

Use pattern for significance, color for reinforcement

If you always use pattern with color, you avoid most problems (except the problem of visual clutter). For example, on a line chart with one significant line, and two or three other lines, you can use a bright color with a solid line for the most important data, then dotted and dashed lines for the less important data.

Pick contrasting colors

Visual acuity is worse for color than for brightness (Gregory 1987, 151). If you stripped away the hue and left only the gray scale, would you still be able to separate items visually? Picking gray-scale values in addition to hue may solve the problems caused by color confusions, black and white print-outs, and low light or low contrast settings on users' computers.

The rule is: To create enough contrast between type, lines, or other small items, and the background, make sure that the colors' gray-scale values differ by at least 20 to 30 percent (White 1990, 73).

How to tell if your selected colors have enough contrast

To check your colors, create a gray-scale ruler:

  1. Pick a program with a color or palette editor. Open the editor and either find or create a set of nine grays and one black separated by 10 percent differences in darkness. Use white for the background. The values for each gray are:

  2. Gray

    RGB Values

    HSV Values

    HEX Values
    100% (black) 0,0,0 0, 0%, 0% 000000

    90%

    26, 26, 26

    0, 0%, 10%

    1A1A1A

    80%

    51, 51, 51,

    0, 0%, 20%

    333333

    70%

    79, 79, 79

    0, 0%, 30%

    4F4F4F

    60%

    102, 102, 102

    0, 0%, 40%

    666666

    50%

    128, 128, 128

    0, 0%, 50%

    808080

    40%

    153, 153, 153

    0, 0%, 60%

    999999

    30%

    181, 181, 181

    0, 0%, 70%

    B5B5B5

    20%

    204, 204, 204

    0, 0%, 80%

    CCCCCC

    10%

    232, 232, 232

    0, 0%, 90%

    E8E8E8

    0% (white)

    255, 255, 255

    0, 0%, 100%

    FFFFFF

     

  3. Draw a set of gray boxes on a white background, one color of gray per box, ranging from 10 percent to 100 percent (black).

  4. Draw diamonds of all the colors you want to test.

  5. Drag each color sample over the gray scale, squinting as you drag it. When the color and a gray box seem to match, you've found its gray-scale value.

  6. Ten rectangles of gray, ranging from 10 percent to 100 percent black.

    Figure 7. Gray-scale ruler

  7. Save the colors that are either 20 or 30 percent apart (separated by two or three boxes) and discard the rest.

Note: Some colors, because of their brightness, maintain high contrast no matter where you put them on the gray scale. However, check the size. Small areas of yellow disappear against white. Red, if used for something small (dots) or thin (lines), shrinks away to nothing against a dark background.

Go to top of page

Avoiding Problems with Adjacent Colors

Most computer illustrations come in full color nowadays, so the question of what colors to use is either irrelevant ("It's a picture of a face! The colors are face colors!") or too complicated to talk about here ("Remove that green cast in her face before sending the file to the color separator").

However, some colors, when used in blocks, do odd things in the presence of other colors. For example, bright colors like red look bigger than dark colors like black (White 1990, 15-18). In Figure 8, notice that the inner red square looks larger than the inner black square:

rA red square inside a black square looks bigger than a black square inside a red square.

Figure 8. Which center square is bigger? Neither.

Problem: Comparing widely separated colors

Since color perception is sharp only near the fovea, color coding is effective only within 10 to 15 degrees of the central area of vision. Widely separated colors are hard to compare, in other words, unless you lean back or step back from the computer (Horton 1991, 226).

Problem: Hues change in proximity to one another

If you want a color to look like itself and stay that way, don't put it next to a complementary color (White 1990, 16-18).

A blue penguin inside a light blue box looks blue. The same penguin inside a yellow box looks darker.

Figure 9. Against a shade of itself vs. its complement

Two penguins, both the same color. The penguin on a black background looks light, On a light blue background, it looks darker.

Figure 10. Same color looks light, then dark

Problem: Complementary colors flicker

Putting blocks of saturated complementary colors next to one another causes eyestrain. Because the cones in your eyes cannot see both colors at the same time, your focus shifts back and forth rapidly without being able to settle on either color:

Two blue penguins on an orange background.

Figure 11. Flicker in complementary colors

Figure 11 shows you what happens with orange and blue. Red and blue-green, yellow and dark blue, and purple and chartreuse (yellow-green) are also complementary colors.

Problem: Contrasting colors create intense edges

The edge between two bright contrasting colors can be very intense and often distracting. To avoid this effect, you can either lighten or darken one of the colors or separate the two colored areas with a white or black line.

Figure 12. Too much color contrast in the first penguin, not so bad in the second

Problem: Color does not make type stand out

Do not fall in the trap of thinking that color is as strong as black because it looks brighter, more cheerful, more vibrant, and so more fun to look at. It is not. You have to compensate for its weakness, to make color as visible as black. There just has to be more of it, so you have to use fatter lines, bolder type, or larger type to overcome the problem. Jan White in Color for the Electronic Age (White 1990, 24).

When you switch from black text or lines to light- or bright-colored text or lines (red, orange, gold), double the width of the lines and use either bold or a larger type size. One to two points larger should be enough for 8- to 12-point type, two to four points larger for 14- to 24-point type. However, check visibility by squinting at the text. Too-light type will recede or even disappear.

The reason for color's poor showing is physiological. Colored letters and numbers can only be read when they are quite close to the eye's focal point, although color itself can be seen far from the focal point. "This indicates that color is a useful aid for visual search but actual reading takes place in a restricted visual reading field. If a reader is familiar with the significance of colors, then colors will help to locate the required information quickly, but the recognition of a word or symbol itself depends on the legibility of characters and not on their color" (Grandjean 1987, 30-31).

Go to top of page

References

Apple Computer, Inc., Guide to Macintosh Software Localization, Addison-Wesley Publishing Co., Reading, MA, 1992a.

Apple Computer, Inc., Macintosh Human Interface Guidelines, Addison-Wesley Publishing Co., Reading, MA, 1992b.

William W. Banks, Jr., Jon Weimer, Effective Computer Display Design, Prentice-Hall, Englewood Cliffs, NJ, 1992.

Robert K. Barnhart, Hammond Barnhart Dictionary of Science, Hammond, Maplewood, NJ, 1986.

Hideaki Chijiiwa, Color Harmony: A Guide to Creative Color Combinations, Rockport Publishers, Rockport, MA, 1991.

Susan L. Fowler, "Banking on a New Interface," I.D., September/October 1993, 70-72.

Wilbert O. Galitz, User-Interface Screen Design, QED Publishing Group, Boston, MA, 1993.

Etienne Grandjean, Ergonomics in Computerized Offices, Taylor & Francis, New York, 1987.

Richard L. Gregory, ed., Oxford Companion to the Mind, Oxford University Press, New York, 1987.

LCdr. Richard J. Hackman, Capt. Garry L. Holtzman, Lt. Penny E. Walter, "Color Vision Testing for the U.S. Naval Academy," Military Medicine, Vol. 157, Dec. 1992.

William Horton, Illustrating Computer Documentation, John Wiley & Sons, New York, 1991.

Shiz Kobara, Visual Design with OSF/Motif, Hewlett-Packard/Addison-Wesley Publishing Co., Reading, MA, 1991.

Les Krantz, What the Odds Are, HarperPerennial, NY, 1992.

Jeffrey R.M. Kunz, Asher J. Finkel, The American Medical Association Family Medical Guide, Random House, New York, 1987.

Aaron Marcus, Graphic Design for Electronic Documents and User Interfaces, ACM Press/Addison-Wesley Publishing Co., Reading, MA, 1992.

Kathleen R. Milhaven, "Visual Communication and Color Blindness," Proceedings, 36th International Technical Communications Conference, 1989.

Gerard Millerson, Lighting for Video, 3rd ed., Focal Press (imprint of Butterworth-Heinemann, Ltd.), Oxford, U.K., 1991.

Adrian Nye, Tim O'Reilly, X Toolkit Intrinsics, O'Reilly & Associates, Inc., Sebastopol, CA 1990.

Winn L. Rosch, The Winn Rosch Hardware Bible, Brady, New York, 1989.

Philip W. Sawyer, ed., "45 Questions to Test Your Ad IQ," Starch Tested Copy, Vol. 4, No. 10, November 1992, pp. 1-4

Philip W. Sawyer, ed., "Quiz Answers," Starch Tested Copy, Vol. 4, No. 10, December 1992, p. 4.

Peter Slatin, "Darkness Made Visible," I.D., September/October 1993, 81-82.

Edward R. Tufte, Envisioning Information, Graphics Press, Cheshire, CT, 1990.

Edward R. Tufte, Visual Design of the User Interface, IBM Corporation, Armonk, NY, 1989.

Edward R. Tufte, The Visual Display of Quantitative Information, Graphics Press, Cheshire, CT, 1983.

Jan V. White, Color for the Electronic Age, Watson-Guptill Publications, New York, 1990.

Go to top of page

Resources

Color and Light

Leslie Stroebel, Photographic Filters: A Programmed Instruction Handbook, Morgan & Morgan, Inc., 145 Palisade St., Dobbs Ferry, NY 10522, 1974. This book teaches you, in the best pedagogical fashion, exactly how filters work with colors. It has an extensive bibliography, a glossary, and plastic filters in a pouch in the back (used for some of the exercises).

Robb Smith, Amphoto Guide to Filters, American Photographic Book Publishing Co., Inc., Garden City, New York, NY 11530, 1979. An excellent reference for types of filters and the effects you can get.

Color, Pattern, and Design

Hideaki Chijiiwa, Color Harmony: A Guide to Creative Color Combinations, Rockport Publishers, Rockport, MA, 1991. Distributed through North Light Books, 1507 Dana Ave., Cincinnati, OH 45209. Pages and pages of color swatches, in combination, plus a lucid description of color theory. This book is also available in the U.K., Phillippines, Thailand, Canada, Singapore, and Turkey.

Edward R. Tufte, Visual Explanations : Images and Quantities, Evidence and Narrative, Graphics Press, Cheshire, CT, 1997, Envisioning Information, Graphics Press, Cheshire, CT, 1990, and The Visual Display of Quantitative Information, Graphics Press, Cheshire, CT, 1983. How the pros do design. Once you've mastered the basics, go here.

Jan V. White, Color for the Electronic Age, Watson-Guptill Publications, New York, 1990. Not about interface design at all, but clear and practical about color in general. Also includes an appendix that compares color specification systems (Munsell, Pantone, Natural, and CIE Notation).

Color Standards

ANSI offers these color coding standards:

  • Color Coding of Discrete Semiconductor Devices, ANSI/EIA 236 Revision C.
  • Colors for Identification and Coding (includes 1988 supplement, 359-A-1) ANSI/EIA 359-A.
  • Safety Color Code, ANSI Z535.1-1991

For more information or to order these publications, contact American National Standards Institute, Attn: Customer Service, ANSI, 11 West 42nd St., New York, NY 10036; voice 212/642-4900; fax 212/302-1286.

Outside the U.S.:

American Technical Publishers, Ltd., 27/29 Knowl Piece, Wilbury Way, Hertfordshire, SG4 0SX, England

Japanese Standards Association, 1-24, Akasaka, Minato-ku, Tokyo 107, Japan

Standards Council of Canada, 45 O'Connor Street, Suite 1200, Ottawa K1P 6N7, Ontario, Canada

Visual Impairment and Adaptive Technology

For an excellent site on color confusions, see Color Vision, Color Deficiency by Diane Wilson.

American Foundation for the Blind runs the National Technology Program, a resource for visually impaired people and their families, rehabilitation professionals, educators, researchers, manufacturers, and employers.

  • The National Technology Program conducts objective evaluations of products and equipment used by visually impaired persons. These evaluations are modeled after "Consumer Reports," are published in the Journal of Visual Impairment & Blindness (JVIB), and are available from AFB's Information Center.

  • The National Technology Program also provides information on assistive technology used by blind and visually impaired people.

  • The Careers and Technology Information Bank (CTIB) features data from over 1,900 blind or visually impaired people who use adaptive equipment in a variety of jobs, including many non-traditional fields. Individuals listed in the CTIB may serve as resource people for consumers and professionals in the field.

As well as offering rehabilitative services to partially and legally blind individuals, the Center for the Partially Sighted prescribes and offers training in visual aids. The Center helped Citibank reprogram its original touch-screen automatic teller machine so that blind and partially-sighted customers could use it (this system was subsequently revamped)..

Go to top of page

Definitions

Parts of the eye

Parts of the eye

Figure 1. Diagram of the eye

A human eye can be divided into two parts: a pupil, cornea, and lens for focusing light, and a retina for gathering light. The retina itself has these parts:

Rods: Visual cells embedded in the retina that are sensitive to light and dark, not to color. There are about 130 million per eye.

Cones: Visual cells that are sensitive to color, not to light and dark. There are about 7 million per eye.

Fovea centralis: An area covering 1° of arc at the back of the eye. The fovea has the highest density of cones (about 10,000 per square millimeter) and the most direct connections to the optic nerve--each foveal cone has its own nerve fiber. (Rods and cones in the rest of the eye are connected in groups to nerves.) The high density and direct connections give the fovea the highest resolving power of any part of the retina. Since vision is most acute here, you instinctively move your eyes until the image you want to look at falls on the fovea.

Blind spot, which is the interface between the retina and the optic nerve. There are no rods or cones on the nerve, so no light is gathered at that point. However, you are never aware of the blind spot because the brain closes the visual field across the blank area automatically. Indeed, you can't find your own blind spot without specialized testing equipment (see Figure 2).

Find your blindspot

Figure 2. Find your blind spot: Cover one eye and stare straight ahead at a blank wall, while holding a spoon at arm's length. Move the spoon slowly back and forth, an eighth of an inch at a time, until the bowl of the spoon disappears.

Achromatic color:

Black, white, or grey--colors without saturation or hue.

Chroma:

A synonym for "saturation."

Complementary colors:

On the standard color wheel, complementary colors lie directly opposite one another. They are called complementary because, between them, they contain all the colors of the spectrum, not because they get along well. The standard complementary pairs are: red and blue-green, orange and blue, yellow and blue-violet, chartreuse (yellow-green) and violet, green and red-violet.

Contrast:

The greater the contrast, the better the visibility. Black on white has the strongest contrast. However, contrasting colors (colors with three hues between them on the color wheel) can cause optical illusions along the edges where they meet. Common pairs of contrasting colors are red and green, red and blue, orange and blue-green, yellow and blue, and violet and green.

Dithering:

Also called "texture mapping." A type of optical illusion. If you put pixels of two or more colors next to one another, the human eye automatically combines them into a third color. If you look closely at color pictures in magazines, you will see that only four colors of dots in various combinations make up the entire full-color picture. The four colors are cyan (light blue), yellow, magenta (pinkish red), and black. Dithering is also used to simulate intermediate colors on a restricted palette or gray scale when you only have black and white pixels to work with.

Gray scale:

A system in which all of the hues are replaced with various shades or brightnesses of gray. Not the same as monochrome.

HSV:

Hue, saturation, value (in some programs, HSL--hue, saturation, lightness--or HSB--hue, saturation, brightness). A system available on some palette editors as an alternative to the RGB color-definition system. Matches the widely used Munsell method of color notation.

Hue:

What is normally called "color." Hues are designated by such names as red, green, yellow, blue, and so on. Hue is a function of wavelength.

Luminance:

The degree of lightness or darkness in colors created by mixing lights.

Monochrome:

Black and white, period. No grays except those created by dithering. However, also used to refer to monitors with one color (usually amber, green, or orange) on a black background (or vice versa).

RGB:

Red, green, blue. Three wavelengths of light--red, green, and blue--create all of the hues visible to primates such as ourselves. Computer monitors use light, not pigment, to create colors. By adjusting the amounts of red, green, and blue light, you can create any of the dozens to millions of colors available on your or your clients' monitors.

When you paint with light, red, green, and blue together make white. When you paint with pigments, however, red, green, and blue make black (or actually, a dark muddy brown that is familiar to most of us from grade school). There are other differences as well. Red light and green light make yellow light, whereas red pigment and green pigment make brown pigment.

Each of the three scales (R, G, and B) has 256 points (usually shown as 0 to 255). Winn Rosch explains why there are 256: "The digital-to-analog converter chip used by the VGA system does more than just convert digital signals to analog. It's actually three DACs in one--one for each color. In addition, it contains the color look-up table for the color mapping process which assigns one of the 262,144 colors [2 to the 18th power] possible under the VGA system to each of the 256 values that can be stored in memory in the VGA 320x200 color-graphics mode. The look-up table values are stored in 256 registers inside the DAC chip itself" (Rosch 1989, 318). In other words, whenever you change a red, green, or blue scale, you are changing a value in one of three 256-cell tables.

Saturation:

Also "purity" or "chroma." The intensity or vividness of a color. Red is more saturated than pink, navy blue is more saturated than sky blue. The more saturated a hue is, the more visible it is at a distance. The less saturated it is, the more difficult it is to see.

Spectrum:

The band of visible colors produced when sunlight is passed through a prism--red, orange, yellow, green, blue, indigo, violet. (Just remember Mr. Roy G Biv.)

Value:

Also "lightness" or "brightness." The amount of white or black mixed into the hue. Some hues are inherently lighter or darker than others--yellow, for instance, is very light while violet is very dark. The word "shade" usually describes a darkened hue, produced by removing light. The word "tint" describes a light hue, produced by adding light.

Go to top of page

FAST Consulting
FAST Consulting Home | Site Map | Accessibility | Web UIs | Desktop UIs | Services | Websites | Colleagues | About Us | Lesson Plans | Amusements
FAST Consulting
© 2008 FAST Consulting All FAST ConsultingRights Reserved Privacy Statement