BKINVCMP.RVW 980926 "The Invisible Computer", Donald A. Norman, 1998, 0-262-14065-9, U$25.00 %A Donald A. Norman www.jnd.org norman@nngroup.com don@jnd.org %C 55 Hayward Street, Cambridge, MA 02142-1399 %D 1998 %G 0-262-14065-9 %I MIT Press %O U$25.00 +1-800-356-0343 fax: +1-617-625-6660 www-mitpress.mit.edu %P 302 p. %T "The Invisible Computer" The subtitle for this book is just a little too long to include in the publication data. It runs "Why Good Products Can Fail, the Personal Computer Is So Complex, and Information Appliances Are the Solution." As to the first part, Norman tends to imply products that fail aren't all that good. As to the second, the book has a few suggestions, but nothing really solid. As to the third, well, yes, this volume does extol the virtues of the information appliance, but doesn't do well at defining what it is, or why it will solve all our problems. It is ironic that Norman is so insistent upon the fact that Edison was such a disaster as a businessman that the term "phonograph" is hardly ever used today. Ironic, because when chapter one finally does get around to explaining the mistakes made (and, when it finally does, it does a good job) the company used as a comparison is the Victor Talking Machine Company, and Norman notes that "victrola" became the generic term for a music playback machine for quite some time. True, there are no more Edison machines on the market (outside of antique stores) while the Victor successor RCA does have a place on the shelves at Joe's Stereo Barn, but while everyone reading this column knows what a phonograph is, it is only the erudite few who are congratulating themselves on remembering what a victrola is. The real message of chapter one seems to be the everything is confusing; business is confusing, marketing is confusing, technology is confusing, and PCs are confusing. This is not a solid thesis on which to build a book. Chapter two reviews the standard "life cycle" model of technology, as well as a "product-promotion-market" view of business. Nothing new is added to these old workhorse analogies and, in fact, the supporting anecdotes weaken as much as strengthen the correspondence to reality. The information appliance is introduced in chapter three. However, the author seems to have failed to note some important material that he includes. The definition of "appliance" indicates specificity: the appliance is designed to perform a demarcated task. On the following page the concept is illustrated by two such appliances, a digital camera, and a business calculator, both of which perform their assigned tasks more easily (from the user's perspective) than could a computer. However, the cutline states that appliances can share data with other devices. I defy anyone to get these two to share anything in a meaningful way. Norman ends the chapter by outlining "versatility" as a design criteria for an information appliance, and therefore enters the fuzzy realm of wanting to have it all. The text then turns to attacking the personal computer in chapter four. The PC's crime? it is too versatile, and is, therefore, complex to use. Chapter five examines new and coming technologies such as large scale programs, more advanced interfaces, speech recognition, agent technology, and network computers, and finds them all wanting. There is no magic bullet to fix the problems of technology. Except information appliances, of course. But Norman's argument starts with a very telling illustration. He insists that technology need not be complicated. There are some things that are complicated, such as the theory of general relativity, but one does not need to know about relativity to deal with physical objects. We will assume, of course, that Norman is referring to physical objects of medium mass and low velocity. Or, we will take Norman at his word and note that he does not know anything about relativity. The point being that Newtonian mechanics, like information appliances, may be fine for the bulk of what we want to do. Sometimes, however, you will need something else. And it is generally (relatively?) those times that make life interesting. Chapter six talks about infrastructure. And economics, metaphysics, cameras, and email. From these points it arises that the proper infrastructure must be designed the right way, must have the right features, must be flexible to adapt to new needs, and must be universal. As the last sentence of the chapter notes, "[i]f only." Chapter seven notes that people are not like machines. Having started by misquoting (yet again) poor patent Commissioner Duell, the fact that devices are sometimes poorly designed is noted in chapter eight. If you are designing for people, chapter nine states, you should pay attention to people. (Another irony: Norman lauds the "glass cockpit" for simplifying flight, without noting that the simplicity comes at a cost of increased complexity when one needs to recall which menu and screen brings up data from a dial that is no longer displayed by default.) Chapter ten tells us that the current business structure is all wrong. All great ideas initially dismissed, says chapter eleven, thus protecting the book from any and all criticism. Chapter twelve discusses all kinds of different technologies and I'm not quite sure what they have to do with appliances. An appendix finally delivers some ideas of what information appliances might actually be. Unfortunately, most of the examples rely on significant advances in technology (a $900, 250 gram device is rather ridiculous as a memory aid for shopping, although a $50, twenty gram appliance might have considerable sales) or are current contraptions that have no relation to the PCs that Norman attacks so viciously. Everyone is familiar with embedded computers, and as processing power gets cheaper and smaller, more and more objects will get added "intelligence." I can sympathize with many of Norman's points about information processing devices. This review was written using WordPerfect 4.2 because for my relatively simple purposes I do not need all the bells and whistles of WordPerfect 8, and the 4.2 version, from a time of simple monospaced fonts, is actually more suitable to the production of a piece that will primarily be read online than the desktop publishing oriented defaults of release 8. (I certainly do not need to argue with a program that thinks it knows what I want to do better than I know myself, such as Microsoft Word 97.) Information appliances will undoubtedly have a majority of uses for those tasks we currently assign to PCs. After all, a pre-printed fax cover sheet is "easier" to use for that purpose than a piece of paper. But it will take significant time, and more advice than we have here, to find out what those appliances will be. As a talking piece to promote the idea of the information appliance, the book may work. It is hard to say that it will succeed, since a lack of detail will not appeal to technical readers. On the other hand, Norman's persistent antipathy to techies would imply that he expects someone else to come up with the necessary machinery. Who that might be is left open as well. copyright Robert M. Slade, 1998 BKINVCMP.RVW 980926