Tom Zillner, Editor
Human Needs and the New Computing Technologies
By Ben Shneiderman. Cambridge, Mass.: MIT Pr., 2002. 288 p. $24.95 (ISBN 0-262-19476-7).
Ben Shneiderman, author of Design ing the User Interface (Addison-Wesley, 1997), has written a new book. Head of the Human-Computer Interaction Laboratory at the University of Maryland for nearly twenty years, Shneiderman is one of the gurus of human-computer interaction (HCI). Leonardo’s Laptop is also concerned with HCI, although it is a broader and more expansive treatment than found in his previous work.
Schneiderman proposes in Leonardo’s Laptop that the genius of Leonardo da Vinci can be applied to the burgeoning world of usability, not only as a touchstone for comparison but as an inspiration as well. For example, he states, “The creative genius of Leonardo . . . has inspired technologists, scientists, and artists for more than half a millennium. His Renaissance integration of engineering with human values could be the path to appealing artifacts and provocative dreams” (3). His invocation of the greatness of da Vinci throughout the book in relation to usability is only partly successful, but given this ambition of Shneiderman that may have been inevitable.
Perhaps most successful in this work are Shneiderman’s fantasies of what da Vinci might think of some of the up-and-coming tools of today and the future. For example, da Vinci carried tablets of several sizes to jot down notes and make drawings; therefore, he would probably be interested in PDAs, desktop computers, and wall-sized displays. Not an entirely new insight, but Shneiderman extends the possibilities further to discuss what a contemporary da Vinci of software might build: “3-D medical simulations with tactile feedback that let you crawl through the human body, a complete environmental model of the world to study global change, and a building-size FrescoMaker drawing package” (9). That last application sounds like the most fun, both for experienced and nascent artists.
The real meat of the book is foreshadowed by Shneiderman's discussion of new computing: “methods that can produce more usable, more reliable computer software and user interfaces that yield much improved user experiences” (26). New computing software supporting innovation will offer exemplars and templates as well as creativity-guiding software to assist users. Shneiderman stresses that these aids to doing and making are compatible with the Renaissance idea of homo faber, man the maker.
Shneiderman then outlines the transformations from old computing to the new. The first of these transformations is a shift shift in focus from hardware capabilities to computing that supports human relationships. In new computing, users no longer talk about their machines’ properties, but about things like the number of e-mails they send and receive and how many postings they’ve made to discussion groups. We are still on the cusp of this transformation, because many people (not just geeks) still discuss the size of their hard drives rather than the number of e-mails they get, and some people mention both.
The second and related transformation Shneiderman describes is a trend toward user-centered services and tools and away from machine-centric processes. There is less interest in computer programs that make automated diagnoses than tools that help doctors make better diagnoses. Although there is still talk about house-cleaning robots, there is much more talk about music downloads. This may be a transformation that we take for granted, but it is important nevertheless. The personalization of Web pages (such as Amazon.com’s welcome page tailored for individual users’ interests) has been one trend that has helped to encourage personalized and customizable tools in other spheres. It’s part of a critical mass that has generated this new move to usability.
The new usability often is incompatible with old ideas about what people want. For example, human-like intelligent agents have not done very well. The epitome of this failure is BOB, Microsoft’s effort to put a human face on Windows that would assist users and automate tasks. BOB cost $100 million, but was roundly rejected by nearly everyone. (Shneiderman notes that there is a bit of residual BOB in the much annoying presence of the happy paperclip helper in Microsoft Office.)
Shneiderman claims this is but one example of the move from artificial intelligence (AI) to user interfaces. People consistently reject AI that is intrusive, particularly AI that attempts to mimic human capabilities. The latter was much hyped in the ’70s through the early ’90s, but never achieved success. What users wanted wasn’t an imperfect machine-instantiated clone but an aid to their own capabilities and talents—thus the age of usability.
One of Shneiderman’s ideas, discussed early in the book, is an example of usability gone too far. He proposes the bar coding of plants and minerals in parks that visitors could scan for a description of the item on a modified PDA. While the idea seems outrageous because of its interference with an enjoyment of nature in the raw, it is perhaps appropriate for small collections. The author’s other ideas seem to be plausible and usually welcome, although many of them are not new.
Shneiderman next provides a lamentably short set of guidelines for user-centered design. This may be understandable in light of the fact that he has pretty much covered the field in his previous HCI work , but it is a disappointment nonetheless. For example, Schneiderman writes that Web sites should be designed so there are many links at the top-level pages, making it easier to get to detail pages. In other words, many shallow links are better than few deep ones. Of course, this is good advice, but not covered in depth.
Shneiderman also points out that interfaces should be consistent, predictable, and controllable. For example, interface consistency can be achieved by placing a clickable icon in the upper left of a Web page or maintaining a uniform color scheme throughout a Web site or program.
Predictable design means that user intuitions and expectations will be met throughout a program or Web site. Shneiderman provides the examples of a shopping basket on the Web or the Save, Print, Open, and Close functions in a software program as examples of predictability.
Controllable interfaces are a favorite technology of this reviewer, and Shneiderman provides good coverage of them. As he states it, “controllable interfaces give you the power to do what you want” (65). Controllable interfaces provide the user flexibility, for example, to insert a spreadsheet into e-mail or select a virtually-limitless undo facility? Best of all, controllable interfaces allow programs to be set up for novices or experts with easy toggling between those modes or offer more advanced capabilities only when requested. Microsoft Office and Windows have taken that general direction, with most-frequently-used choices appearing first in pulldowns, and other choices only coming into view when you hold onto the mouse button at the bottom of the pulldown. Again, it is disappointing that these ideas were covered in only a couple of pages.
Shneiderman then proceeds to analyze user activities. Chapter 5, “Understanding Human Activities,” contains some seminal ideas that can be fleshed out by anyone in the HCI, Web design, or software development fields. The author offers four categories of relationships: self, family and friends, colleagues and neighbors, and citizens and markets. Little discussion is required to understand these relationships of anyone from the smallest slice of the world up to the largest slice. Secondly, he categorizes four stages of activities: collecting, relating, creating, and donating. Individuals collect information, relate pieces of information to others, create works, and donate these works to one or more of the categories of relationships. Finally, Shneiderman introduces the Activities and Relationships Table (ART), a four-by-four category grid with relationship categories along one axis and stages of activities along the other. He then discusses ART using two examples, one a set of photo collection and management tools, and the other a set of tools for ubiquitous computing. These examples flesh out his categories and the ART concept.
Shneiderman then offers up a number of chapters providing his view of possible near-term changes in education, business, medicine and politics, all supplemented by ARTs. He has some solid ideas, but while they are imaginative, they are not revolutionary. For example, the author devotes several pages to the possibilities of Web sites, discussion groups, and online chat for exchanges of patient information. Shneiderman’s purpose here and elsewhere in these chapters must be to explore the somewhat quotidian examples of everyday electronic life to back up his claim that we are shifting to a user-centered world, and he does that very well.
As you may gather, this reviewer was not quite as happy with these chapters as with the rest of the book. Perhaps it’s because Shneiderman’s examples have been discussed elsewhere; people new to discussions of the current trends and near-term future might be a better audience for this material.
This reviewer was, however, quite happy with the last chapters of the book. Shneiderman explores the many facets of creativity in his chapter “Mega-Creativity.” He provides a set of principles for being tremendously creative. Of course, this ties in very well with da Vinci, a mega-creator if ever there was one. Shneiderman ends the book with a short chapter, “Grander Goals,” that offers a wider vision of the future than is found in his examples of contemporary usability.
Although this review is somewhat mixed, I believe Leonardo’s Laptop is worth reading. Even if readers skip the chapters on education, business, medicine, and politics, they will still come away with an interesting and worthwhile viewpoint on usability. And usability is something that can stand plenty of books and papers.
As a sidenote, there are wonderful reproductions of the works of da Vinci, both paintings and notebook pages, throughout the book. These and some one hundred other da Vinci images are available from Planet Art for around $140, all of them royalty free. If you make a lot of presentations, have a large Web site, or produce a newsletter, you might want to invest in the Planet Art CD.— Tom Zillner, Wisconsin Library Services, Madison.