Books & People

back Books & People Overview

Books, People

Archives

Book Reviews

 

Book Review: Digital Ground

Book | Author | Review

By Gerd Waloszek, SAP AG, SAP User Experience – February 10, 2005

In this review, I take a personal look at the book Digital Ground by Malcolm McCullough.

 

Book

Cover of Digital Ground     

Malcolm McCullough
Digital Ground: Architecture, Pervasive Computing, and Environmental Knowing
MIT Press, 2004
ISBN: 0262134357

Design: General

 

Author

Photo of Malcolm McCullough Malcolm McCullough explores digital media for the built environment. Beginning with computer aided-design in architecture, in which he was a pioneer in the 1980s, McCullough has consistently brought a human-centered approach to emerging practices in design. His latest book, Digital Ground, connects the fields of architecture and interaction design, and offers a theory of place for pervasive computing.
(From the biography on his Website)

 

Review

Malcolm McCullough introduces his book Digital Ground as "an architect's response to the design challenge posed by pervasive computing." This new type of computing has incited him to change his view on how the environment and technology fit together. The author sees pervasive computing as representing a paradigm shift from building virtual worlds – as promoted by cyberspace proponents – toward embedding information technology into the physical world. McCullough discusses the challenges (for example, location models representing context, protocols) for this new technology as well as the chances that it offers for changing the outlook of a technology-centered future into a human-centered one. McCullough particularly questions which experiences from the discipline of architecture we can build on to establish a design discipline that focuses on interaction design and takes our embodiment and engagement with the environment into account.

Let's start with a short overview of the book:

  • Part I, "Expectations," asks what we expect technology to be used for, what we expect using it to be like, which is the "user experience" aspect and how we expect technology to enhance our everyday lives.
  • Part II, "Technologies," describes the hardware and software components of pervasive computing (or embedded) technology and how it is applied. Here, McCullough offers a typology of "situated interactions" (also called situational types or genres) around classified life situations.
  • Part III, "Practices," explores the value of context-centered design.

Finally, in the "Epilogue," McCullough discusses what the idea of "going native" – the wish for a simpler life – means in an age of technological saturation. He concludes that rejecting technology is not a viable option. Instead, "response to place – and context, I would like to add – becomes the most practical strategy of all."

The following sections will pick up some ideas from the book.

The Role of Embodiment and Context for Interaction Design

When speaking of embedding technology into everyday life, McCullough uses the terms "situated technology" and also "embodiment." The latter, he characterizes as "not just a state of being but an emergent quality of interactions." It is our embodied predispositions, which – together with our intents and the contexts in which we act – shape our behavior; it is not just symbolic mental models and processes, as cognitive science taught us (including me) for decades. McCullough agrees with Lakoff and Johnson who stated: "The mind is inherently embodied. Thought is mostly unconscious. Abstract models are largely metaphorical. These are the three major findings of cognitive science."

As a consequence, McCullough calls us to recognize that deliberative mental models are complemented by direct engagement and peripheral awareness. This has direct consequences for user interface design: Creating usability requires taking contextual learning into account. According to McCullough, "contextual learning begins at embodiment, remains largely personal, and is life-long." For this type of learning, "engaged interaction is at least as important as detached perception. The possibilities of the world correspond to the capacities of the body." Thus, our engagement with the environment, or interactivity is so dependent on context that we can no longer afford to ignore it. For McCullough, context is more than the setting or space itself: It "subsumes the engagement with it, as well as the bias that setting gives to the interactions that occur within it." In this sense, contexts

  • "shape perceptual selectivity, provide background cues, and enable the application of tacit knowledge"
  • "are full of props and cues, which serve as learning resources and memory devices for evolving patterns of usage"

McCullough concludes: "This actively engaged way of learning about the world challenges the assumption that technology is a purely symbolic literacy, independent of ground." I would add from my own experience that the way our environment acts as a memory device and initiator of actions is very perspicuous.

By including the physical and social environment, interface design has, according to McCullough, become interaction design, because design now goes beyond mere user interfaces. Interaction design has also come into alliance with architecture: Both disciplines address how contexts shape actions. Nevertheless, McCullough does see the design challenge of pervasive computing more directly as a question of interaction design than of architecture.

To McCullough, the use of the term interaction design instead of interface design is a cultural advance in the design field: "Interaction designers claim to know at least partly what is wrong with information technology, and that overemphasis on technical features and interface mechanics has been a part of the problem. By turning attention to how technology accumulates locally to become an ambient and social medium, interaction design brings this work more closely into alignment with the concerns of architecture."

In summary, McCullough's essential claim is that interaction design must serve our basic human need for "getting into place," and "embodiment provides a continuing basis for human-centered design."

Example: Locus of Attention

As an example of how this view can change our design perspective, particularly with respect to embedded systems, I would like to quote some of McCullough's remarks on attention:

  • "Information technology design has occupied itself with tools for deliberate reasoning – a process that occurs in the foreground of human attention. This attention remains finite, while the number and complexity of tools continues to increase. In what has become a problem to almost all design disciplines, the foreground is full. Periphery, on the other hand, is background that is outside focal attention but which can quickly be given that attention when necessary."
  • "Graphical user interfaces have long been built on the principles of shifting focus – picking up a tool, opening and closing windows, etc. – but they still leave us staring at a cluttered screen. Portable and embedded systems take the information processing out into the physical realm, where the capacity for periphery is deeper and the act of bringing things to the center is more intuitive – this is mainly a matter of embodiment in context. Our embodied predispositions have been underfed while our foreground deliberate attention has been oversaturated."
  • "Norman's call for the invisible computer reflects the need for less attention-consuming technology."
  • And on the role of architecture: "The more enduring the environment, the more it shapes our expectations without saturating our attention."

Example: Patterns, Typology

Lately, patterns have been discussed in the user interface design community with increasing interest – and with recourse to architecture, particularly to Christopher Alexander's typology. Alexander's "Pattern Language" of about 250 architectural elements was "mainly built to the affordances of built spaces for living but also takes care of the 'situations' that are encountered there."

With respect to user interface design, however, patterns are mostly dealt with from a technical point of view: The component-based approach has been promoted as a means for simplifying and automating the implementation of user interfaces – context is not typically considered.

One might regard SAP's approach to user interface patterns as a first step in this direction, which relates patterns to typical user tasks. Pervasive computing promises to take this approach one step further because pervasive systems are embedded in physical environments, which we interact with. This move poses a design challenge, and McCullough asks: "Do we answer it with a one-design-fits-all approach, the chaos of infinite combinatorial possibilities, which design seldom benefits from, or can we find a manageable range of recognizable situations?" As McCullough is aware of the importance of habitual contexts, he proposes a rudimentary typology of thirty life situations (situational types) that are arranged in groups, such as "at work," "at home," "on the town," and "on the road" – probably inspired by Christopher Alexander's pattern language for architecture.

According to McCullough, these "situational types suggest one provisional basis for designing the emerging digital layer of space." He is confident that "typological design should advance the appropriateness of location-aware technology more quickly than universal standards or one-time circumstantial configurations" and adds: "The usability of well-made traditional places now appears as a rich basis for design of context-aware technology."

Example: Citations on User Interface Design

For the more "practically-oriented" readers, I have collected some "nuggets" from the book that deal with user interface design.

Technology-Centric Design

  • "The prevailing computer-human interaction (CHI) model of interface design has been partly responsible for the current state of the desktop computer. The breakthrough on which the field emerged was the admission of psychological principles. The resulting graphical user interface has been the focus of the field of computer-human interaction for nearly 20 years. This interface is a virtual control panel whose design has remained quite technology-centered."
  • "Software designers have pursued the accumulation rather than integration of technical features. Interface designers have emphasized first-time usability at the expense of more satisfying long-term practices."
  • "... current interfaces illustrate how many computer scientists are biased toward efficiency with technological resources rather than human attention; or to put it bluntly, toward convenience for computers before convenience for people."

Usability

  • "Usability metrics remain more an inspection and less an aid to conception, however. Thus their use perpetuates the notion of design as a corrective process. Also the metrics have little to say about off-screen context and how that shapes intent."
  • "Following from the principles of scientific management, whose applications to design were best demonstrated in Taylor's time-and-motion studies, industrial engineers sought efficiency in the repetitive mechanical actions of using a technology. When today's software developers seek efficiency in the motor skills of moving a mouse about a screen, toward a goal of idiot-proof, button-based interfaces, they perpetuate this legacy."
  • "Although some people identify with using a system that is easy the first time, or in occasional use, others identify more with a system that rewards lifelong learning even if it is difficult at first. ... The identity of design artifice is likely to involve an assertion about its usability in context. This rightly discounts immediate novelty and first-time "idiot-proofing" in favor of usefulness and long-term assimilation. As the mainly mechanical business of counting clicks grows into a sociological study of usability in contexts, it should involve the psychological study of how people invest themselves in mastery of technology."

UI Last or UI First?

  • "If until rather recently most businesses held little regard for design, that is because they saw it as something applied after the fact. When it merely dealt with packaging (including "front end" interfaces), design seemed superficial. When it was thought of as applied decoration, which may still be the most widespread connotation of the word, design implied cost rather than income. ... When design is applied to productions that have long been analytically conceived, then the self-fulfilling unimpressive results can be used to demonstrate the superficiality of design."
  • "As Jef Raskin has summarized: Once the product's task is known, design the interface first; then implement to the interface design. ... As far as the customer is concerned, the interface is the product."

(Highlighting by the author of the review)

Conclusion

Looking at the book with the "tunnel vision" of a user interface design practitioner who is interested in practical lessons, we can easily see that it does not teach "practical user interface design," such as Alan Cooper's or Karen Holtzblatt's books do – but this was, of course, not to be expected. (The work of Karen Holtzblatt & Hugh Beyer on contextual inquiry is appreciated by McCullough, though, and he quotes Alan Cooper with his call for a new class of professional interaction designers.) This book is about (interaction) design as a discipline and its future role in the presence of pervasive computing. Nevertheless, I found quite a few "nuggets" on user interface design in it that are useful for every UI practitioner to know and keep in mind.

Can I recommend the book? Yes, I can clearly recommend it for those people in the UI design field who want to look beyond the demands of their daily work, reflect new technologies, and ask how they can be designed and embedded in our lives and the environment. These readers will surely find a lot of interesting thoughts to ponder. And they should be willing to "dip" into some philosophical background.

By the way, while other reviews characterized the book as "easy-to-read," I must admit that, as a nonnative speaker, I found the book fairly hard to read. I also doubt that many user interface practitioners will have the perseverance to read the book from start to end. Regrettably, the book contains a comparatively high number of misspellings, which should be corrected in the next edition.

 

top top