By Donald A. Norman

The MIT Press

(C) 1998 Donald A. Norman
All rights reserved.
ISBN: 0-262-14065-9

CHAPTER ONE

Drop Everything You're Doing

"Drop everything you're doing," my CEO said to me. "I have a really important job for you to do."

    This was my rude introduction to the world of high technology. I had just left the hallowed halls of bickering academia to join the harried frenzy of industry. Everything was different, yet, to some extent, everything was the same. In academics there is much emphasis on cleverness and creativity. It doesn't matter how successful you were in the past. What counts is: What did you accomplish this year? What did you publish? What do colleagues and opponents think of your work? It is more important to be clever than to be correct, better to be profound than practical. In the pecking order of the university, those who are the most abstract and irrelevant are at the top, those most practical and useful are at the bottom.

    The world of high-technology business doesn't work the way you think it does. Technical, business, and social factors affect the way new technologies are deployed. Once ideas are let out of the laboratory, common sense disappears, especially in the rush to show that one company's products are superior to another's very similar ones.

    In industry, there is much emphasis on action, on shipping products. Profit margins and sales figures rule the day. It doesn't matter how successful you were in the past: How much have your market share, sales, and profits increased this year, this quarter, this month? Market success is everything, much more prized than the nature of the product. It is better to be first than best. Everything is always in a rush: to get to market, to beat your competitor, to get the product out in time for the holiday season, or the school season. In the research university there is much thought, little action. In industry, there is much action, little time for thought.

    The request of the Chief Executive Officer of my company, I would soon discover, was not unusual.

    "But you told me to drop everything last week," I replied, "I've barely, gotten started on that job."

    "That's OK," said my CEO, "this one is more important. Drop everything you're doing, and by the way, you have a 9 AM meeting tomorrow on the east coast. You can sleep on the plane. My assistant has already arranged for the tickets."

    The abrupt and secret nature of this particular task was not at all unusual. The high-technology industry rushes to do this, to do that. The race is to the swift and the clever, not to the best. The continual rush makes it impossible to step back and reflect, to take a broader view. To the product manager, thinking about the future means to think six months ahead, maybe twelve. To the researcher, thinking about the future means to think ten years ahead, maybe five. The gap between these styles is almost unbridgeable. Don't believe everything you read. In fact, don't believe anything. How much impact do science and research actually have on products? Less than you might think, less than you might hope.

    We live in interesting times. We are indeed in the midst of a major revolution, one brought about by the converging interests and technologies of communication, computation, and entertainment. Technological revolutions have several interesting properties. First, we tend to overestimate the immediate impact and underestimate the long-term impact. Second, we tend to place the emphasis on the technologies themselves, when it is really the social impact and cultural change that will be most dramatic. And, finally, we think revolutions are fast, with changes occurring in months or, at worst, a few years.

    In reality, technological revolutions are fast only from a historical point of view. Look at the revolution triggered by Gutenberg's development of the movable-type printing press. Its impact was rapid--within 100 years it was felt all across Europe. To the historian, 100 years is a short time; for someone living at the time, it is an eternity--more than most lifetimes. All new technologies take a long time to affect the lives of ordinary people. The telephone was invented in 1875, but it didn't have a major impact until the 1900s. The airplane was conceived in the late 1800s and flown successfully in 1903, but it didn't become commercially available for passengers for 30 years and, even then, it was a risky, foolhardy means of transportation. The fax machine was invented in the mid 1800s and even today has not yet made much of an impact in the home.

    Today we often hear that the pace of change has speeded up, that changes happen in "internet time," in months or weeks, not decades or years. False. The internet had its start in the 1960s as a government-sponsored research network for universities and company research laboratories. Thirty years later, at the end of the 1990s, it still is not present in the majority of homes in the United States, and its adoption in other countries is even lower. The digital computer is fifty years old, the personal computer more than twenty, yet fewer than half the homes in the United States have computers and the number is far lower in other countries. These rates of adoption might be faster than the airplane and the telephone, but not by as much as you might have been led to believe. Technological revolutions are rapid when measured by the time span of civilization, but slow when measured by the time span of an individual's life.

    Nonetheless, we now really are in a period of rapid change, of rapid convergence of technologies. One hundred years from now, our descendants will wonder about the primitive life we led, where we had to contend with diverse communication facilities: government mail systems for personal delivery of letters handwritten or typed on paper; private companies for more rapid delivery of time-sensitive material and small packages; a separate system--fax--for the delivery of pictures of material; electronic mail for typed material; yet another system--the telephone and its associated answering machine and voice mail technology--for voice messages; and no video mail. It doesn't take much thought to recognize that all these are similar systems and that society would be better served with one solution that merged the delivery of personal and business mail, voice messages, images--whether static or moving--and graphics. It also doesn't take much time for those trying to merge these technologies to throw up their hands in frustration at the myriad hurdles society places in their path.

Where Edison Went Wrong

Today's high-technology business is in a muddle. It has arrived at its current state by its heavy emphasis on technology, quite often technology for technology's sake. The modern computer is the culmination of this process and, as I explain more fully in this book, it has led to an overly complex, fundamentally difficult machine--but one that increasingly has come to dominate our lives. Today we cannot do business or conduct much of our daily activity without the use of modern information technology, both the computer infrastructure and the associated communication technology. But the computer does not really meet our needs. It suffers from the rush, the haste, and, for that matter, the arrogance of the technology industry. This is an industry that puts the device first, the customer second. The real needs of consumers are ignored.

    Today, at the start of the twenty-first century we find the industry dominated by personalities, by major magnates whose personal presence controls the development of technology. The same was true at the start of the twentieth century: Edison, Bell, Marconi, and Westinghouse all played major roles in the development of the early days of the information industry, and all were colorful, powerful, and very public figures. Let us begin with those early years, when the first information technologies were introduced to the marketplace: the telegraph, the telephone, the phonograph and, later, the radio. For my purposes, the story of Thomas Alva Edison is the most relevant; he played a major role in many of those early information industries, from communication (the telegraph and telephone), to entertainment (the phonograph and movie projector), to the underlying infrastructure (the incandescent light, electric power distribution, and electric dynamos and motors).

    Thomas Alva Edison was a great technologist--one of the best--but not a great businessman. Not only was he an inventor, but he started one of the first industrial research laboratories, hiring some excellent technicians and scientists. Moreover, he realized that no invention could succeed by itself; it needed an entire infrastructure to work. With this in mind, he put together all the other components of the total system. He had all that logic indicates should be necessary: He was first with the technology, the technology was usually superior to that of his competitors, and he understood that success required the construction of appropriate infrastructure. What he did not have was a solid sense for marketing. He often did not sufficiently understand his customers.

    Edison had many strengths, but he also had several major weaknesses. It is for those deficiencies that I am telling this story, for the deficiencies are common ones, still with us. This book is really about the high-technology industry, and especially the personal-computer industry, and how they should change. Edison's story is a great place to start. In many ways, Edison invented the high-technology industry. His work combined the information processing and communication industries. He played a major role in the development of the entertainment industry by inventing the phonograph and the motion picture camera and projector. And like the business tycoons of today, he had a cantankerous, colorful personality, suing and spying upon his competitors, courting the press, hyping his inventions often even before they were invented, and leading a lively, eccentric, and very public life. The feuds, suits, and rivalries among technologists at the turn of the nineteenth century far exceeds the similar excesses of today.

    Consider the phonograph. Edison was first, he had superior and more versatile technology, and he did a brilliant, logical analysis of the business. But the logical approach is the wrong way to go about understanding the needs of customers. You have to talk to them, watch them; this is the only way to understand their interests, their motives, their needs. Edison thought he knew better and, as a result, he built a technology-centered phonograph that failed to take into account his customer's needs. I'll return to this story later, but for now, note that at the end, his several phonograph companies were unsuccessful.

    Sound familiar? It's like today's personal computer business, and there are even more parallels. Do you think the computer is difficult and the phonograph simple? The phonograph was judged to be too complicated for office use; it took about two weeks to master, but only if you were willing to persevere. It took years to get the phonograph to the state it is in today, changed so thoroughly that all of the underlying technologies differ and even the term phonograph is no longer used. By this analogy, the computer industry still has a long way to go and, if we are fortunate, the end result will be quite unlike today. The problem is that whether phonograph or computer, the technology is the relatively easy part to change. The difficult parts are social, organizational and cultural.

    Today's technology imposes itself on us, making demands on our time and diminishing our control over our lives. Of all the technologies, perhaps the most disruptive for individuals is the personal computer. The computer is really an infrastructure, even though today we treat it as the end object. Infrastructures should be invisible: and that is exactly what this book recommends: A user-centered, human-centered humane technology where today's personal computer has disappeared into invisibility.

    This change will not be easy. It requires a disruptive technology. For manufacturers, it requires a new approach to design, a human-centered design. This means hiring new kinds of people, changing the design process, perhaps restructuring the company.

    Why not? Remember, there are far more people in the world who do not use computers than there are who do: That is the marketplace, that is where the opportunities lie. Companies shouldn't always talk to their customers: they should talk to those who are not (yet) customers. Among other things, this was Edison's main fault. He thought he knew better than his customers; he didn't provide them with what they wanted, he provided them with what he predetermined was best for them. This was a bad idea then, it's a bad idea now.

Why Being First and Best Isn't Good Enough

Success does not always go to those who are first to market, nor to those with the best product. Alas, the history of technology is filled with failures of those who were first. Quick, what was the name of the first manufacturer of automobiles in the United States? Duryea was the first. Most people today have never even heard the name.

    The history of the phonograph is illustrative. Edison invented the phonograph in 1877, and by 1878 the Edison Speaking Phonograph Company was marketing the first machines. It even made a profit for the first few years. At first, the technology was crude--the recordings were made on tinfoil, the machine was delicate. Edison and his competition engaged in a series of technological improvements, substituting wax for tinfoil, discs for cylinders, electric motors for hand- or spring-propelled drives, and so on. People weren't quite sure what the machine would be used for, and so, at first, it was used primarily for public demonstrations (with paid admission). Edison thought they could lead to a paperless office in which dictated letters could be recorded and the cylinders mailed to the recipients, without the need for transcription. He also tried putting a small phonograph into a doll and selling it as a talking toy. Owners of the early machines held parties where guests could record songs and games and enjoy listening to them being played back. Prerecorded speeches and songs became popular, and this eventually became the primary usage.

    The practical phonograph did not arrive until the late 1880s, by which time Chichester Bell and Charles Tainter had developed a competitive company, the American Graphophone Company, that recorded on wax-coated cardboard cylinders rather than tinfoil. In the early 1890s Emile Berliner became the first to produce prerecorded stamped disc records commercially with his machine, the gramophone.

    Edison's phonograph had a number of superior features to the competition. But having the best technology is no route to success either. Edison had the best technology. Sony's Beta technology for video cassette recording is widely considered to have been superior to the VHS format for video cassette recorders and tape championed by a conglomerate led by JVC and Matsushita, but Beta lost. The Macintosh operating system clearly had advantages over the DOS operating system for personal computers, but it lost, first to DOS, and then to Microsoft Windows, a system that took ten years to catch up to the Macintosh, but that now dominates the world of computers. Edison's recordings were superior, but he used technology that was incompatible with that of his competitors. His competitors provided a product that fit the customer's needs better than did Edison's phonograph, and as a result, the superiority didn't matter: the competition won.

    Being first helps, but it is not enough. Being best helps, but it is not enough. What was Edison's mistake? He used a technology-centered approach. His logical analysis of the technologies did not take into consideration the consumer's viewpoint.

    When Edison invented the phonograph, he studied both the cylinder and the disc, the form of the medium in use today in CDs and DVDs. He recognized the superiority of the cylinder as a recording medium. He knew that as the cylinder revolves, each part passes under the stylus at the same speed. With discs, the outside edge moves past the stylus more rapidly than the parts near the center, and so the sound at the center deteriorates. Because the stylus in Edison's phonograph was propelled across the record by a feed screw, record wear was decreased in comparison with the disc, where the groove itself propelled the needle. Edison's machine could also be used to make home recordings. The cylinder's semipermanent jeweled stylus was more convenient than the discs' steel needles, which had to be changed after every record side.

    But the discs offered many advantages over the cylinders. They were less fragile than the wax cylinders. Their hard shellac surface allowed greater playback volume, if a more raucous, scratchy sound, than the soft wax cylinders. They took up far less space, and were easier to store, package, and ship. They could accommodate longer playing time simply by increasing their diameter, and they had a second side that could provide more music without increasing storage space, for less money. Most important, they were far easier to mass produce.

    The real use of the phonograph record, discovered after much trial and error by a variety of manufacturers, was to provide prerecorded music. Emile Berliner moved quickly to exploit this and his company rapidly picked up a dominant market share. His gramophone became the Victrola, manufactured by the Victor Talking Machine Company, later RCA Victor. Berliner and his successors rapidly established recording studios across the world and engaged the world's most famous musicians in the effort.

    Edison's failure to recognize the real value of the phonograph is understandable: New technologies often end up being used very differently than anyone--especially their inventors--predict. The real error lay in letting his competitors get ahead of him and in failing to recognize the practical advantages of the disc over the cylinder in terms of ease of use, storage, and shipping, to say nothing of mass production. Instead, Edison scoffed at the raucous, scratchy sound of the disc machines compared to the superior sound of the cylinders.

    Eventually Edison did realize the importance of compatibility and convenience. The problem was that by the time he switched over to discs in 1913 he was no longer the market leader. Worse, he failed to understand the real desires of his customers; once again, he let logic triumph over marketing wisdom.

    Even after he began manufacturing discs, Edison continued to use the vertical recording method, called "hill and dale," in which the sound wave was represented by the vertical motion of the needle and a corresponding depth of the groove. The needle vibrated up and down as the record rotated. The competition used lateral recording, which meant the needle vibrated side-to-side as the record rotated. Once again, the differences are mostly technical, but the real point is that the world wanted a single system. Early phonographs could only play back one system, either vertical or lateral, so whichever system customers bought, they couldn't play back the recordings of the other. In chapter 61 discuss the problems of nonsubstitutable, infrastructure goods, when the infrastructure of one company differs from that of another. Basically, if you have the dominant infrastructure, you win. If you choose the wrong one, you lose, and you lose big. Because the Victor Talking Machine Company had the dominant infrastructure, Edison lost.

    Which method of recording was actually better? Edison, of course, thought hill and dale superior, and in numerous "tone tests" he sought to prove that an audience could not distinguish the sounds produced by his phonograph from those of a live singer. Once again, it didn't matter. The sound quality of Victor's lateral recordings was "good enough."

    Another serious mistake that Edison made was in the choice of recording artists. Edison decided that big-name, expensive artists were not much different from the lesser-known professionals. In this, he is probably correct. Take the ten best piano players, or opera singers, or orchestras in the world, and the difference between those ranked at the top and those at the bottom is not likely to be noticeable by the average listener. But the top two or three musicians are a lot better known, whereas few people can recite the name of the tenth best performing artist or group. Edison thought he could save considerable money at no sacrifice to quality by recording those lesser-known artists. He was right; he saved a lot of money. The problem was, the public wanted to hear the big names, not the unknown ones. It had been cleverly educated in this by Victor. As a Victor advertisement put it (see figure 1.1, page xiv):

If you had your choice of attending two concerts--the greatest artists in all the world appearing at one, some little-known artists at the other--which would you choose? You would quickly decide to hear the renowned artists who are famous for their superb interpretations. And this is exactly the reason why the Victrola is the instrument for your home. The world's greatest artists make records for the Victor exclusively.

    Edison pitted his taste and his technology-centered logical analysis on the belief that the differences among musicians were not important: He lost. He thought his customers only cared about the music; he didn't even list the performers' names on his disc records for several years. He failed to understand that people want to hear the big names. It doesn't matter if others are just as good. It doesn't even matter if they are better--it is the name that matters.

    This would have been less important had it not been for Edison's choice of an operating system that was incompatible with the machines most people owned, hill and dale recording rather than lateral. If Edison had used the same standard way of recording the sounds as his competition, then it wouldn't have mattered that the big names were on Victor records. People would have been able to buy Edison phonographs and play Victor records. But with a specialized, incompatible infrastructure, if customers wanted the name musicians, they had to buy both the records and the phonographs from Victor. To put it in today's terms: There were two competing, incompatible operating systems. Eventually some companies did make instruments that could play both kinds of records, but by then, it was too late.

    The Victor Talking Machine Company cemented its lead over Edison when it introduced the Victrola in 1907; this machine, with its amplifying horn concealed within the cabinet, became so popular that the word victrola became the generic term for any record player for the next five decades. Again, understanding the desires of the customer, who was growing tired of the intrusive, ever-larger horn, let a company maintain its dominance in the market.

    Note the moral of this story, for it will apply over and over again in the high-technology marketplace. Know your customer. Being first, being best, and even being right do not matter; what matters is what the customers think. Edison was first, and he did have the best sound quality; but because he failed to fulfill his customer's desires, he fell behind in sales. He lost out by pushing the cylinder even though the customers preferred the more convenient disc. Edison pushed the phonograph as a recording media, when customers were more interested in listening to prerecorded music. When Edison finally did switch to discs, he wouldn't use famous--and expensive--musicians, but instead hired excellent but lesser-known musicians. As a result, he was never able to capture market share from his competition, especially from the Victor Talking Machine Company. And finally, he used a different technology than that of his leading competitors: at first cylinders instead of discs, and then vertical instead of lateral recording. Edison had studied all the methods; he thought his choices were superior. Maybe they were.

    This is the important lesson about infrastructure technologies. It doesn't matter whether or not your technology is superior; it only matters that what is being offered is good enough for the purpose. Moreover, if you lead the marketplace in sales, it is permissible to use a nonstandard infrastructure. After all, if you have the majority of customers, then what you do becomes the standard. Your competitors have little choice but to follow. If you are not the leader, then having a non-standard infrastructure is a bad idea. Ultimately, it leads to extinction.

But It's a Horrible Product

"But it's a horrible product," I complained. "It's not the sort of thing our company is famous for. We are known for ease of use, but look at this thing. Look at the remote control--38 buttons! I thought we were famous for simple, easy-to-use products. Nobody will be able to use this device!"

    The time was almost midnight. It was the first few weeks of my new job, and already I was fighting with the vice president in charge of the new consumer products division. Wisely, however, he chose the site and time of the battle: his living room, from 10 PM to whenever we finished. And he provided the wine, a fine, rich northern Italian red. By now, we were in the second hour of discussions, on the second bottle of wine.

    "Yes, yes," he said, "everything you complain about is absolutely true. But it doesn't matter. You see, nobody is going to buy the product."

    "What!" I exclaimed, "If nobody is going to buy it, why are we in such a rush to produce it, such a rush that you won't let us take the time to fix the problems and do it right?"

    "Ah," he replied, "it's a business strategy. See, by getting this product to market now, we announce to the world that we are not just a computer company, that we make things for the living room, for the home entertainment center. It's really too expensive for most people, and besides, most people have no use for it. But we will have established a foothold in the marketplace. Market share is everything. Speed is essential. Whether or not it works doesn't really matter. By being first, we will dominate. Then, we have time to make lots of products, and to make them right. And to make sure we make them right, I will put you in charge, OK?"

    This was yet another introduction to the world of business. Being right wasn't always the point--being first was much more important. And what a disarming way to win me over; ply me with good wine, agree with every criticism, and put me in charge. I left a bit after midnight, complaining that I was tired and had an 8 AM meeting. "Actually, I have a 7 AM meeting myself," he said, "but I still have a few hours work to do. Midnight in California is a perfect time to make phone calls; it's 9 AM in Europe and 5 PM in Japan--perfect for both countries." That's life in the world of high technology. The pace is unrelenting. Action is prized. There is little time for thought.

    The product shipped. It was a failure. Sales were miserable. And no, I was never actually put in charge. I did try to fix the complexity with an alternative, superior design, but we had only one month to get the new design into production, and two weeks of that were taken away from us because the factory in Portugal was going on vacation. Time, or rather the lack of it, I was starting to learn, is one of the greatest barriers to quality.

    Would it have mattered if I had managed to fix the problems I knew about? Probably not. Was the executive correct? Probably. The product failed mainly because there was no real need for it. It didn't fall into any well-understood product category. It was targeted for the family room or living room, to sit on top of the television set, but consumers didn't yet have any experience with this kind of device, so they didn't know what they would use it for. It wasn't meant as a computer peripheral, but this is how it was reviewed and judged. As a result, reviewers didn't find it compelling; for a peripheral, it was too expensive and too slow. For these people, the attractive industrial design was irrelevant. Ease of use--or its absence--was equally irrelevant.

    This product really was ahead of its time. So the fact that it was hard to use didn't matter. Yes, at least one product review commented upon the lack of usability, but that isn't what killed the product. Was the vice president right to say that market share and being first mattered more than doing it right? Yes, he was. The problem, I now realize, is that this was a new product category, a disruptive technology, but it was being aimed at the mass market, for the everyday user to put in the living room on top of the TV set. As I point out in chapters 2 and 11, disruptive technologies require special handling. The early adopters will be technologically sophisticated. For them, industrial design and ease of use is irrelevant. They would be perfectly happy with a horrible product as long as it delivered some capability they felt they needed. This one didn't deliver, at least not yet, not in its first manifestation. It might have succeeded in the long run, but it only got one chance.

    What kind of a world is this, anyway, where horrible products don't matter? Welcome to the world of the technology enthusiasts, of the early adopters, of the fan clubs, and of the belief that technology comes first. But also welcome to the real world, where products do not exist in a vacuum, where for a product to be accepted by the public it has to fit into some recognizable niche, to provide value the customers understand. The right product can fail if introduced at the wrong time. The telephone took decades to be understood and accepted. The radio was first dismissed as a toy. The fax machine was invented more than a hundred years before it became an essential tool of industry. So, too, with the automobile. So, too, with almost every new technology.

    When a new product category is invented and manufactured, it does not fit naturally into people's lives. In order to survive, it has to be introduced gradually. It has to be able to demonstrate its unique benefits that outweigh the trials and tribulations of all early technologies. Product categories, as a result, have a special life cycle in which they at first offer novel capabilities at the price of complexity, limited functions, and high cost. Over a period measured in decades or longer, they transition into everyday objects that offer what have become essential capabilities with simplicity, appropriate functionality, and low cost. They start out being technology dominated, of interest only to those few adventurous souls who are willing to pay the costs in order to get the benefits. They end up as everyday consumer items of value to everyone.

    This is the story of the automobile, of the telephone, of the phonograph. It will be the story of the personal computer, although right now the computer is still in its early days, still complex and expensive. The computer has yet to make the transition to being an everyday object, with simplicity and low cost. Personal computers are not yet consumer items of value to everyone. To make the transition requires a very different view of the world than is held by the technology enthusiasts. It requires a consumer-centered view. Edison, for one, was resistant to such a view, and his companies failed.

    Once upon a time, when the computer industry was young, the miracle was that you could make a small, personal device that could actually accomplish something useful. Youngsters in garages could put together products that soon became thriving businesses. Technology was king; those who mastered it could do no wrong. Those who could innovate became the stars--rich, influential stars. The push was to develop better and better technology, fancier and fancier tricks. Hurrah for this invention, hurrah for that. The fire was fueled by capitalism: Invest in the correct small company and you could increase your investment many fold. Everyone fought for a piece of the action, and the action was hot new technology. Never mind that people couldn't use it. Never mind that the products kept getting more and more complex, more and more difficult to build, to maintain, to understand, and to use. Never mind. People kept buying--in part because they too were trapped by the technology mania, in part because they were forced to.

    The notion that the marketplace decides only applies if there are real choices to be made. In the early days of a technology, it is often a choice between purchasing unwieldy, expensive technology that is difficult to use and maintain and doing without. To the technology enthusiasts, the technology, whatever its failures, offers advantages that they simply can't live without. In the relentless pace of business, there is always the nagging fear that the new technology offers a superior competitive advantage, and if you didn't keep up, your competitor might get an edge.

Technological Change Is Simple; Social Cultural, and Organizational Change Is Hard

When will electronic mail replace much of the paper mail now hand-delivered by the international postal system? Not for a long time, because to make the change requires agreement with unions, satisfying concerns about worker's rights and national pride, and overcoming the sheer inertia of culture. Some countries legislate against anything that might interfere with the transport of mail. Electronic mail is not necessarily welcomed.

    Want to make it easier to send a fax to the home? You must cope with the existing international standards for telephones, originally intended to carry only a few signals: a dial tone, busy signal, ringing signal, and voice to the connecting parties. Today it would be useful if a message could electronically indicate whether it is voice, computer modern, or fax, or whether it might go to a personal machine, or directly to a voice or fax mailbox. This can't be done, even though the technology that would be required is relatively simple. The problem is that the existing infrastructure is not compatible with such signals.

    To the person using these technologies, the infrastructures are irrelevant. With a telephone answering machine, one leaves voice messages. On a fax, one sends a copy of the image of the page. Email seems like fax in that one types a message and then commands the system to send the document visible on the screen, but what gets sent from one machine to another is very different from a fax; a fax will look like a degraded replication of the original, while email sends the computer code for the individual letters, and the receiving end recreates the text with its own choice of font. Fax looks like the original, email does not, even if it has the same words.

    People trying to communicate wonder why they can't get voice messages over email. They can't understand why the letters so visible on the paper or on the screen from a fax can't be edited or operated upon in the same way as the similar appearing words from email.

    In reality, these technologies use very different, barely compatible infrastructures. This historical accident makes it difficult to reconcile the differences. Voice mail is an analog representation of sound waves, encoded and stored according to the conventions of the telephone switching networks. Fax is also an analog signal, a picture of the letters and words, encoded in such a way as to pass through the telephone networks even though they were designed for voice, not data. The only way data can be sent over conventional telephone lines is to change the digital signals into audio tones that can pass through the network, then transform them back into data at the other end; hence all those tones and strange sounds created by fax machines and modems.

    Email contains a digital representation of the letters, encoded in a binary encoding of the alphabet according to a scheme named by the acronym ASCII: American Standard Code for Information Interchange. ASCII is also a historical relic that doesn't work well in today's international society, where it cannot handle all the letters and diacritical marks of European languages, let alone the nonalphabetic characters of Asian languages. The first letter of the acronym ASCII indicates the problem: "A" for "American."

    The impact of cultural influences is apparent in the different ways in which fax and email are used in business. Email and fax serve almost identical functions, yet they have come to be used rather differently. Email is the medium of informality; fax is more formal. Email tends to be typed quickly, by the person doing the correspondence. A fax tends to be prepared with the normal office procedures, with secretaries or administrative assistants neatly polishing the words and with care taken in grammar, spelling, and format. The differences reflect their historical origins.

    Today's fax standards and machines resulted from a development process by large companies for business use. The technology of fax fits well within standard office routines, the same procedures used for letters, with responses written or dictated for a secretary to type. In both cases, a formal letter is produced. The only difference lies in the method of transmission and reception. Letters are put into envelopes and picked up and delivered by the mail service. Faxes are put into the office fax machine and the appropriate number dialed. The formality of a fax reflects the formality of the letter.

    Email has evolved from the world of computer science, primarily in universities and government and industrial research laboratories. It was developed for quick notes between researchers, sometimes dealing with business, but more often with personal matters, such as restaurant suggestions or arranging meetings and social gatherings. Informality, ungrammatical writing, and misspellings were common. A well-formatted, properly constructed, formal message is considered out of place in email. In addition, the restriction to the limited character set of ASCII and the lack of formatting tools severely limits the style, leading to the use of special conventions, such as *asterisks* or UPPER CASE to indicate an *EMPHASIZED* word, and special symbols, such as ;-) to mean that the preceding remark was meant in jest (the symbols ;-) looking somewhat like a winking, smiling face if viewed sideways).

    The difference between email and fax makes an important point: Once a technology becomes entrenched, it is very difficult to change. Thus, it would be valuable if email would allow formatting and the inclusion of pictures and diagrams right in the text, much as is possible in a book or letter. Similarly, it would be useful if faxes encoded the characters so that one could perform a computer search on the text, or copy and paste words and phrases from a fax into other documents. The problem is that these functions would require new standards and agreements among all the manufacturers and users of the relevant technologies to adopt equipment compatible with the standards. Such a task is almost impossible once technologies have become established. Too many organizations would have to change, too much expense in existing facilities would have to be expended. Hence, the social, cultural, and organizational aspects of a technology are more difficult to change than the technical ones.

    Today we are at a critical juncture in the deployment of information technology. The world is dominated by the personal computer and its attendant communication and network structures. The personal computer has evolved historically to become the standard tool for doing things, despite its many flaws, despite its complexity, despite the fact that it is ill-suited for many of the tasks that it performs. As this chapter has shown, what is best is not necessarily what succeeds: The nontechnical aspects of a technology can dominate.

    Once an infrastructure gets established, it is difficult to change. Even when it is clear that new methods would provide superior results, the old ways linger on, for they are so deeply embedded into the culture of a society, so deeply ingrained in the ways that people have learned to live, work, and play, that change can take place only very slowly, sometimes taking decades.

    There is a better way, a world of information appliances. The task, then, is to look carefully at the alternatives. Examine information appliances. Examine just why the personal computer is so complex. We cannot forget the lessons of history: The best products do not always succeed. Social, cultural, and organizational factors can predominate over technological advances. Nonetheless, the world of technology today is far too complex; there has to be a better way. And there is.

    Let us begin by examining the way market forces affect the life cycle of a product. At different points in its life cycle, a product appeals to very different market segments, who demand different attributes. As a result, the way that a product is developed, designed, and marketed has to change radically as a product moves from it early youth to maturity. The nature of these changes tell us what needs to be done to move from today's world of the personal computer to the power and simplicity of information appliances.

CONTENTS Preface............................................................vii 1 Drop Everything You're Doing.......................................1 2 Growing Up: Moving from Technology-Centered to Human-Centered Products.............................................23 3 The Move to Information Appliances................................51 4 What's Wrong with the PC?.........................................69 5 There Is No Magical Cure..........................................89 6 The Power of Infrastructure......................................113 7 Being Analog.....................................................135 8 Why Is Everything So Difficult to Use?...........................163 9 Human-Centered Development.......................................185 10 Want Human-Centered Development? Reorganize the Company.........203 11 Disruptive Technologies.........................................231 12 A World of Information Appliances...............................247 Appendix Examples of Information Appliances........................263 Notes..............................................................273 References.........................................................285 Index..............................................................291