Icons and symbols have been part of the user's experience of computing for decades, and many people tend to take them for granted as part of graphical user interfaces. But they weren't always there. The Apple Macintosh popularized "icons" as the slightly misnamed term for these visual signs, and by the mid-1980s they became part of graphical user interface paradigms and associated with the desktop metaphor. Windowing environments like the Macintosh, Windows, Open Look, Motif, NeXT, and other desktop and workstation platforms all adopted variations of the trash icon, folder icons, document icons, and specific application icons. Today, most work or play environments include (some might say are littered with) 50 to 100 icons. As I write these words, in fact, I counted about 75 visual signs currently on my screen. Is this good or bad? As user-interface designers, I think we care about the quality and use of icons and symbols as part of our responsibilities as user-interface developers. I think they will be increasingly important in the years to come. Let's take a closer look at their current status and probable future.
I've mentioned icons, symbols, and signs. What's the difference? It depends on which definitions you adopt. For decades, I've used a set of terms adapted from semiotics [6 (Chapter 7), 7,zrefx>,
- Icons: signs that are self-evident, "natural," or "realistic" for a particular group of interpreters, like a photograph of a person, a "realistic" painting, or a right-pointing arrow to indicate something should move or is to the right.
- Ideograms: symbols that stand for ideas or concepts, for example, the letter "i" standing for "information," "help desk," or "information available."
- Index: a special semiotics term for signs that are linked by cause-and-effect in space and time, like a photograph representing a scene, or a fingerprint on the coffee mug at the scene of the crime.
- Lexics: the attributes of how one produces signs, for example, with color or black-and-white raster-scan displays.
- Phonograms: symbols that stand for sounds, for example, the letter "s."
- Pictogram: an icon (or sometimes symbol) that has clear pictorial similarities with some object, like the person or men's room sign that (for some interpreters) appears to be a simplified drawing of a (specifically, male) human being.
- Pragmatics: the uses of signs by receivers, including their emotional, cognitive, and behavioral characteristics, for example, how memorable or appealing the signs are.
- Semantics: the reference of a sign to some object, structure, process, or concept; often referred to popularly as its "meaning," like a dictionary definition, or its denotations and connotations.
- Semiotics: the science of signs. One might call designing icons and symbols "applied visual semiotics."
- Signs: perceivable (or conceivable) objects that convey "meaning."
- Symbols: signs that are usually meaningful by convention and are often abstract, like the letters of this sentence or a national flag.
- Syntax: the arrangement of signs in space and time, especially their visual attributes such as color, size, and shape.
These terms may seem a bit daunting, but they allow us to talk about signs, icons, and symbols in a more straightforward, unambiguous way.
Icons and Symbols will be increasingly important in the years to come.
Looking backwards, let's recall that the Macintosh began with a corporate suite of approximately 250 icons. In the early 1980s, many software developers did not understand enough about the sign development process to realize that they would eventually have a significant visual-assets management challenge. During 1982 to 1985, when we first started designing icons for computer-aided design and manufacturing applications, it was not unusual for companies to brag that they had somewhere between 5,000 and 15,000 signs (they sometimes were not sure of the exact number and had no easy way to find out) in their suite of tools. Keep in mind that Basic Chinese has approximately 3,000 signs and that set of signs developed over millennia. Software developers were designing two to five times this number in a few years. It doesn't take a rocket scientist to figure out that many of these signs were poorly designed. To be more specific, they were illegible, unreadable, with poor usability (efficiency, effectiveness, and satisfaction), and poor internal system consistency (visual syntax, semantics, and pragmatics).
The status of icon and symbol design in the industry has improved, and some aspects have become more sophisticated.
In general, we calculated that it would take three to eight hours to design a single sign, depending upon its semiotic complexity. Often our clients thought it should take about half an hour, considering the price they were willing to pay for sign design. In fact, I recall that Digital Equipment Corporation did an in-house study of how much it cost to develop one sign in the late 1980s and found that reasonable costs, including all those involved in planning, analyzing, designing, evaluating, and implementing, was $2,000 per sign. If it really had taken a short time to design one sign, $2,000 would have been a very luxurious hourly rate indeed.
Gradually the status of icon and symbol design in the industry has improved, and some aspects, such as multimedia features and the extent of use of signs, have become more sophisticated. High-resolution color displays capable of animation and sound have made icons and symbols in some cases multimedia extravaganzas. Many specialists have emerged. Blattner , Bly , and Gaver [8, 9] were among the early CHI researchers in the 1980s who investigated and wrote about the use of sound, specifically "earcons," which are auditory signs associated with visual displays. In some cases, these sound characteristics have become routine experiences for mobile phone user interfaces and corporate branding (think of the elaborate ringing melodies to distinguish different callers and the "Intel Inside" four-note signature).
Today, icons and symbols are pervasive throughout most platforms (client-server, Web sites, and Web-based applications for workstations, desktops, mobile devices, appliances, and vehicle systems). For better or for worse, most computer users must master as novices an astonishing range of signs. Undoubtedly, analysts have studied how many signs average consumers must know, how many they understand well, and how many they can remember and use well in a given context. They have also studied how these measurements change by age, gender, language skills, education, culture, and emotional stress level, among other factors. Many factors are worth studying further, and I call attention to the humble icon or symbol as a subject deserving of further study, with a likely return on investment.
I have noticed some current trends and issues with current icon and symbol design. The pervasive use of icons and symbols, especially at small sizes, has tended to produce many icons with highly informative parts that are too small to be easily recognized or distinguished. A miniaturized document sign may be part of another, more complex sign, all of it displayed within approximately 16 x 16 pixels. Even with significant color depth, the resulting visual form is too demanding on our eyes and mind, especially for older viewers.
Another trend is large, overly complex signs that are overdone. I have in mind the Apple Macintosh X icons and symbols, which almost seem swollen or bloated in their visual design qualities, a far cry from the sleekness of Apple products in the past and minimalist styling of the Titanium series of portables. The Macintosh X "icon bar" does swell with animation as the cursor rolls over a particular sign. Together with other visual characteristics of user-interface elements (window detailing, scrollbars, buttons), the overall effect is one of obesity. Perhaps this is a holdover from the excesses of the DotCom era of the late 1990s. The lean and mean 2000s may bring us an entirely different style harking back to some of the styles of earlier decades.
Still another interesting trend is the emergence of widely used emoticons among Japanese users of mobile phones, who are quite happy to select from a complex palette to their text messages beyond the Kanji (Japanese) or Roman letterforms that typically are used. This trend takes the smiley face to the next level. We already see the use of creative text usage, like "U R Cute 2" to enable youthful paramours to communicate effectively. The widespread use of what I call "baby faces," that is, user interfaces with small-screen displays, makes the space-saving, denotation- and connotation-rich use of new kinds of signs like these emoticons naturally desirable and effective. Written language may "fast forward" to a form of the past, the rebus, which is a mixture of alphanumeric and other visual signs.
The likely value of advanced visual sign systems for professional users in health and medicine, finance and banking, travel and transportation, and education and training, among other markets, appears to be self-evident. Many professionals, such as electrical engineers, landscape architects, medical pathologists, and stock-market analysts, already use specialized sign systems. The challenge with widespread, ubiquitous, small-scale, portable, mobile displays to take advantage of well-designed, new sign systems is daunting, especially in conjunction with information visualization forms (tables, forms, charts, maps, and diagrams), but it is also enormously invigorating for both the developer and the user.
As we look to the future, it is worth taking a look at the past and at other cultures. Many intriguing experiments or actual exotic systems in use may prove extremely useful, practical, and appealing. In some ways, these may be forms of, well, if not "killer apps," then "killer sign systems."
Let's not forget the standards proposed for worldwide mass transit and mass communication systems. These have been codified and published widely (see, for example, [1, 16]). In addition to the current technical sign standards proposed by the International Standards Organization  and the various national organizations in the United States, Japan, Germany, and elsewhere, internationally designed standard signs are available from many manufacturers in various media .
One particular set of design inventions is that of universal visible language sign systems, that is, systems intended to supplement (or, zealots would say, replace) traditional systems of writing, which tend to be idiosyncratic amalgams of signs, except for a few more rational approaches that have occasionally occurred in history. An example is Korean Hangul, which was created in the 1700s by fiat of the emperor, who grew tired of his people using Chinese symbols to represent Korean speech sounds and commanded the development of a system that "even a fool could learn in a short period of time." He succeeded.
Throughout history, developers of language systems have been intrigued by universal writing systems, called pasigraphy. These systems, based on universal signs, were especially popular after the Renaissance. Books detailing the many attempts to design and implement these systems exist [2, 10]. More recent examples of proposed universal, visible-language sign-systems include the inventions of two innovators, Charles K. Bliss and Yukio Ota, who sought to develop something like a visual Esperanto (a previously proposed and partially implemented, universal spoken language). These visual-language systems provide systematic means of depicting nouns, verbs, adjectives, adverbs, numerical, spatial, and temporal components, more or less like natural spoken languages. They may, or may not have systems of pronunciation.
In the early 1940s Bliss, born in Austria and later an Australian resident , invented a system he called Semantography (now known as Blissymbolics), a "logical writing for an illogical world." He tried, but failed, to get the United Nations to recognize his system. Bliss symbols have been incorporated into interactive computer-based systems that are able to help mentally disabled patientsfor example, those with aphasiacommunicate better.
Another example is a system called LoCoS, designed in the 1960s by Yukio Ota, a Japanese graphic designer. LoCoS stands for Lovers Communication System , a charming reference to one of the hopes of both of these inventors: improving human communication among those who do not speak the same natural spoken language, thereby helping to reduce misunderstanding and promoting world peace.
Mr. Ota has demonstrated around the world over the past decades that his system can be learned in one day and has published books and many articles about the system, although most of the publications are in Japanese. Because of my interest in this system, our firm has arranged with Mr. Ota to launch a Web site devoted to his system. Readers interested in accessing this extranet may contact me for information.
I believe that novel and systematic approaches to the use of visual sign systems are needed to help make baby faces successful, especially for the emerging wrist-top devices. Fossil, a well-known manufacturer of wristwatches, has ported the Palm operating system to one, and IBM announced some time ago its porting of Linux to a wrist-top device. To be truly useful, these small devices will need to master the use of small areas of display and to combine that skill with the use of sound for input and output and of hard buttons on the body of the device.
Another major, new area for sign development is vehicle user interfaces. A revolution has begun in the worldwide platform for automobiles, trucks, and other vehicles as they acquire advanced telematics, become hubs for communication, are connected to the Web, mix informational with other forms of communication (persuasive and aesthetic/entertainment), and have flexibility of visual display unlike anything preceding it. New forms of visual signs are likely to play a role for status displays and information visualization in general, as we indicated in a recent article .
The signs are everywhere for the increasing importance of good visual sign development. In the future, all phases of planning, research, analysis, design, implementation, evaluation, documentation, and training, especially for information visualization and localization by language group and culture, will provide some exciting opportunities. Perhaps someone's name will go down in history as the eponymous developer of a breakthrough approach. Nowadays, you don't have to be an emperor to have this happen, you just need to be a winner in the commercial marketplace of ideas, products, and services.
In the meantime, through visible language systems, if we help improve the chances for world peace through better visual communicationwell, that's not too bad, either, on your résumé. Perhaps you may also get a Nobel Peace prize for good visual communication in user-interface design.
19. Peirce, C.S. Existential graphs. In Hartshorne, C. and P. Weiss, eds., Collected Papers of Charles Sanders Peirce, Vol. 4: The Simplest Mathematics, Book 2, Chapters 1-7, Harvard University Press, Cambridge, 1933, pp. 293-470.
President, Aaron Marcus and Associates, Inc. (AM+A)
©2003 ACM 1072-5220/03/0500 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2003 ACM, Inc.