Designing a Wired Life

Philip E. Agre
Department of Information Studies
University of California, Los Angeles
Los Angeles, California 90095-1520
USA

pagre@ucla.edu
http://polaris.gseis.ucla.edu/pagre/

Paper prepared for the WebNet 2000 Conference.

Please do not quote from this version, which differs slightly from the version that appeared in print.

Version of 16 December 2000.
4800 words.

 

Once upon a time, Western culture associated information technology with an old and powerful story about the future. The future, according to this story, lies in rationality, and the task of the engineer is to discover optimally rational social arrangements in a scientific manner and then impose these arrangements upon the world. This picture was already fashionable in the early 19th century through the works of Henri de Saint-Simon (Hayek 1952), and it remained vigorous until the end of the Cold War (Lilienfeld 1978). Information technology grew up in the midst of this project of social rationalization, and the main tradition of computer system design is still organized around a cycle with three phases: studying existing work practices, rationalizing them, and either automating them altogether or using technology to impose a rational order on them.

Even though the United States had associated technology and rationalization just as strongly as the Soviet Union, the Soviet Union's spectacular collapse was quickly followed by a collapse in cultural respect for rationalization. The very concept of design came into question, and information technology suddenly became associated with the opposite extreme picture of an optimal social order that arises from a totally undesigned anarchy [1]. Digital networks were accorded a tremendous power to smash hierarchies, overturn governments, short-circuit intermediaries, integrate the global economy, and bring about Adam Smith's idealized picture of the marketplace (e.g., Gilder 1992). The very idea of prediction becomes hard to understand in such a world. The whole point of bottom-up self-organization is that no single individual has all of the knowledge or intelligence that would be required to comprehend the whole, much less to design it -- or predict it. Yet predictions are plentiful, as they always are, and remarkably confident despite their supposed impossibility.

The truth, of course, is somewhere in the middle between these extremes. No matter what one thinks about the various methodologies of hard-core social rationalization, design and prediction are surely possible within relatively narrow and stable spaces, and on those rare occasions when existing social arrangements temporarily melt and seek a new form. But the insight of the anarchists is deeper than this. The great tradition of social rationalization applied technological metaphors to society: society was a machine, and designing society was supposed to be just the same thing as designing a machine. History has shown this idea to be false. What now replaces it is something more like evolution: local innovations in local niches, each selected for or against and propagating when conditions are right (Nelson and Winter 1982). This view has considerable basis in historical and economic studies of technology, whose development is much more incremental, with many more cross-cutting lines of influence and interaction, than the Godlike engineers of rationalization would admit (Basalla 1988, Rosenberg 1982). But the biological metaphor has its limits, and past a certain point we need a clear understanding in human, social terms of the niches within which design takes place.

This need for a new understanding of design in its social contexts has several aspects:

(1) A shift from an object focus to a process focus (Floyd 1987; Gronbaek, Grudin, Bodker, and Bannon 1993). The adoption and use of technology is a human process, and it is always embedded in a set of social arrangements. The design of technology, likewise, is embedded in the social world. Design is a human process, and it always brings together the contrasting interests and understandings of different individual and groups. In the sense that people are political animals, design is a political process, and has to be understood and reflected upon as a process. Otherwise it risks being torn apart by the social forces it is trying to ignore.

(2) Designing both the technical and social aspects. A newly designed machine may provide a compelling demo, but a machine that simply drops from the sky into an unsuspecting social system will probably not be used for anything. The object of design thus includes both technical and social aspects (Bowker, Star, Turner, and Gasser 1997; Jirotka and Goguen 1994; O'Day, Bobrow, and Shirley 1996). Thus, for example, the documentation and training that accompany a new technology should express the same unifying concepts as the technology itself. Because social arrangements cannot be designed in the same manner as machinery, however, it becomes necessary to understand the differences between the two kinds of design activities, as well as the ways that they can be integrated.

(3) Punching through the interface. Traditional information systems require a sort of mind-body split: information happens in the mind; physical work activities happen in the body. To use information technology, individuals must effectively deactivate their bodies by sitting still in a chair and fixing their attention on a glass screen. That screen becomes a metaphor for the boundary between the real world and the information world, the atoms and the bits (Mitchell 1995, Negroponte 1995). New technologies, however, make it possible to eliminate that artificial boundary (e.g., Bellotti and Bly 1996). Useful computational devices can communicate without wires, present information in any modality, fit in any physical space, be sewn into clothes, recognize one another from physical proximity, and so on (Fitzmaurice 1993, Gershenfeld 1999, Norman 1998, Pentland 1996).

(4) From engineering to arts. These three changes in design lead up to a revolution in design thinking that is worth naming. Engineering approaches to design have advanced considerably in recent years, for example with the maturation of concurrent design methods. But the real excitement in design right now is coming from the more artistically inclined design methods of architecture and industrial design. These methods emphasize an organic unity of conception that informs every aspect of the design; they involve iteration to explore a design space in both conceptual and concrete terms (Schon 1983); they focus on designing a physical object and not just an abstract network of data flows; and they attend both to the utilitarian and the symbolic aspects of the object being designed. Architecture has long been a visible site for the convergence of the arts and engineering. Product design companies such as IDEO (Nussbaum 2000), museums such as the Smithsonian's Cooper-Hewitt National Design Museum in New York (Albrecht, Lupton, and Holt 2000), and even mass retailers such as Target (Patton 1999) have been popularizing industrial design as well. And now these methods are migrating into the design of software (Krippendorff 1996, Winograd 1996). The contrast between the engineering and arts approaches to design might be illustrated using the concept of a trade-off: for the engineer, design is inevitably a matter of identifying the optimal compromise when faced with an array of trade-offs; but for the architect or industrial designer, a trade-off is a symptom of the need to rethink the design for one more iteration, until the process converges on a design that is fully resolved, and that "works" in an organic way.

(5) Structural analysis. For all their strengths, the arts-derived design methods have not placed great emphasis on systematic analysis of the environment in which a designed artifact will be used. This kind of analysis is especially important in designing the new information services because of the robust way in which these services interact with the social order around them. One can design and appreciate a coffee pot as pure form, and even a substantial building interacts only to a limited degree with other buildings around it. Information services, however, are all about interaction; their purpose is to represent, to be woven into, and indeed to intervene in, social arrangements of some complexity. In the past a computer system designer could be satisfied with getting basic algorithms to work, or with the interface between a single computer and a single user. But new systems are much more likely to cross organizational boundaries -- boundaries that can be unspeakably complex in cultural, economic, and legal terms (Friedman 1989). The way forward for the design of information services thus plainly requires more advanced forms of social analysis. An example would be the analysis of collective cognition -- that is, the diverse social mechanisms, using diverse technologies, by which real social communities think together (Agre 1998). Without analyzing the role of newsletters, opinion leaders, old-boy networks, rumors, historical memory, professional societies, and informal gatherings at bars in the collective life of a community, one has little hope of understanding how a new information service might be useful, or not.

(6) Services not systems. Even though the "systems approach" promises a kind of holism in design, in mainstream use a "system" is a machine, considered artificially out of its context of use. The main tradition of computer system design is focused on machines for the simple reason that its whole reason for being was automation: getting rid of people and replacing them with machines. But it is increasingly clear that the operative unit of analysis for design is not the machine itself but the service it is used to deliver. A service is not just a technical functionality. It includes a diverse set of social roles that come together to provide something useful to someone. These social rules might include buyers, sellers, system administrators, customer support people, authors, editors, advertisers, and graphic designers, as well as librarians, archivists, and other professional imposers of order on bodies of information. A service is also a business proposition, whether as a business plan for a private firm or as an organizational strategy for a nonprofit or public organization. And a service involves policies: what's allowed and not, how exceptional cases are dealt with, how liability is assigned, who gets served first, who gets access to what, and so on. A system can "work" wonderfully in a narrow sense, without necessarily being compatible with any practicable service design.

One way to conceptualize the new task of design is in terms of a "fit" between the designed service and the social arrangements around it (Kling and Elliott 1994). The social world is complicated, and its complexity can be found on several levels. These levels might be taxonomized coarsely as follows:

(1) Ergonomic. The vast literature on "human factors", originating with the context of industrial automation but growing into a vast scientific enterprise in the context of the Cold War, is concerned with the human being (the "user") as a biological organism. Contemporary computer users are generally oblivious to the scientific basis for detailed design features such as the small-scale workings of the mouse and menus in the Macintosh. Now that same body of knowledge is becoming infinitely more important in the design of devices that are integrated with the human body in more ambitious and intimate ways. The journal Presence, for example, largely consists of advanced studies in the ergonomics of "virtual reality" and other advanced interface technologies that require designers to take quite complex features of human visual perception and motor control into account.

(2) Individual. In the 1970s, computer science discovered the "user", who was a real human being in contrast to the machine component of the automation paradigm or the biological organism of ergonomics (Cooper and Bowers 1995, Friedman 1989). In one sense the "user interface" movement simply shifted the worldview and methodology of ergonomics from the biological level to the cognitive level (Card, Moran, and Newell 1983). But it also entailed a shift, as Bannon (1990) put it, "from human factors to human actors", and it was accompanied by a vigorous if often vague ethical commitment to the empowerment of the individual user as against the bureaucracies of computing. This user was still very much an individual, however, and the individualistic assumptions of that period can be seen today in the legacy of "personal computers" that have no coherent model of trust for purposes of sharing software or transacting business over the Internet.

(3) Group. The 1980s focus on team-based work in industry brought attention within computer science to work in groups. It seemed natural within the computer science orientation to invent a genre of software called "groupware" that captured and orchestrated well-understood or normatively motivated structures of group interaction (Greenberg 1991). For example, so-called design rationale capture software was intended to allow groups of designers to record their reasoning for the benefit of future designers who might be faced with fixing or extending their systems once the original designers are gone (Moran and Carroll 1996). One of the principal discoveries of the groupware movement is just how complex and subtle group processes are (e.g., Button 1993, Grudin 1994). Structures of collaborative action that seem crystalline and rule-like at first prove to be no such thing upon detailed investigation, and human beings have proven remarkably adept at making use of whatever communicative resources are available to them. Capturing group processes in digital form poses its own challenges, inasmuch as their intricate local organization tends to overwhelm the practical work of record-keeping.

(4) Institutional. An institution is a persistent structure of human relationships (e.g., Powell and DiMaggio 1991). Each institution defines a set of social roles that go a long way toward organizing people's lives together, and institutions such as medicine, banking, holiday rituals, the family, and the common law remain remarkably stable over decades and centuries. Institutions do change, of course, and the present day appears to be an period of institutional change to rival such great historical episodes of institution-building as that of 12th-century Europe (Berman 1983). Information technology does "cause" these institutional changes, in a sense anyway, but it is more accurate to say that the institutions are changing themselves according to their own internal logic, and that they are each taking hold of information technology and shaping it to their own needs.

These multi-leveled conceptions of "fit" pose a deep philosophical challenge for the designer: choosing between the essentially conservative approach of fitting one's design into existing circumstances and the essentially reformist or radical approach of designing services that will blow institutions apart, for example by enabling competitor institutions to rise and gain a foothold in the lives of their participants. As this way of explaining the choice makes clear, cultural constructions of technology have not entirely abandoned the world-shattering ambitions of their rationalist predecessors. In both the old dispensation and the new, one important thing remains out of fashion: detailed attention to the reality of people's lives. If you are going to remake the world, why spend time investigating the way it is? Markets move so fast anyway that there isn't time. But the world remains complicated on many levels, and information services that do not fit into their social environments are (other things being equal) less likely to be used. That, then, is the challenge for contemporary design: learning how to take the multileveled context of design and use fully into account while also taking a responsible stand, through the design process and its products, about the way the world should be.

At UCLA, we have been exploring ways to reinvent the design of information services for a world of pervasive computing. (I directed this project, which we organized as a course in our department's master's program in the spring quarter of 2000, together with Charlotte Lee. For more details on the project, see <http://dlis.gseis.ucla.edu/people/pagre/240/>.) This paper reports one set of conceptual design exercises provoked by the possibilities of spontaneous wireless networking technologies such as Bluetooth. These technologies allow devices to recognize one another automatically, so that they can exchange information about their capabilities and determine whether they might have useful business to transact. Devices that conduct such negotiations among themselves obviously raise questions about the social structure of the relationships among the devices' owners. They also challenge designers to develop a deeper understanding of place: devices can adapt their functioning to the contexts in which they find themselves, and those contexts can also adapt themselves to the devices whose presence they notice on a given moment.

Our own goal, however, was not simply to exercise the technical or interface potentials of the new technologies. We wanted to use them as points of departure for a deeper investigation of some of the emerging design themes that I listed above. The new technologies are so powerful and flexible that anybody can conceptualize a device and tell a just-so story about the ways that people might use it. We wanted to start the design process with a community, learn from that community's ways, and explore through the iterative design process how a new device might fit into the many dimensions of the community's ongoing life while making a compelling statement about what the community could be.

In designing for communities, we were aware of a tension. A strong design, in our view, expresses a clear unifying design concept and fits in diverse ways into the lives-together of the community that uses it. The design, in other words, should be influenced in numerous ways by the activities, metaphors, institutions, values, and identity of its user community. But equally strong economic motives require a degree of generality: a device that is useful only to stamp collectors, for example, may not achieve the economies of scale that it needs to be produced cheaply, or the network effects that it needs to draw in an ever-widening community of users (Messerschmitt 1999).

To explore this tension, therefore, we conducted an experiment on the subject of middleware for ubiquitous computing (Weiser 1993). Modern information and communications services are organized in layers, with complex functionalities being partitioned into clearly defined modules, some of which presuppose the existence of others. In computer networking, for example, the Internet protocols constitute "middleware" because they are built on top of a variety of underlying service layers for moving data packets from point A to point B, and because they provide a generic platform for the construction of a wide variety of applications. Once the Internet protocols have been implemented on top of every common type of network, those otherwise incompatible networks can be interconnected, and programmers can build applications without needing any knowledge at all about the workings of those networks. The point then generalizes to other service layers that might be built on top of the Internet protocols, providing generic functionalities such as cryptographic encoding or distributed databases that applications developers may wish to rely on.

Although middleware to support mobile applications is still an emerging area (Pappo 2000), our goal was not so much to develop such middleware ourselves as to explore the tensions that arise when applications designed in depth for several distinct communities must all be constructed on top of the same middleware layer. To this end, we posited a hypothetical middleware layer called the Suitcase Layer (SL). One can imagine SL being implemented on top of a distributed object database layer such as CORBA, as well as a resource discovery layer such as Jini. A suitcase, implemented using SL, is a virtual container that individuals can carry with them. "Inside" that container are documents, applications, configurations, and other packages of information that might be useful in operating devices that one encounters in work sites, public places, transportation facilities, and other technology-rich environments far from home. Individuals might carry in their pockets a very simple device, with no moving parts, no switches, no display, and no connectors other than battery recharge. This device might then use Bluetooth to "talk" to all of the other devices in its environment. When it finds a potentially useful device, it conducts an automated negotiation whose details are specified by particular applications. For example, one might design an self-configuring menu. Diners would carry dietary preferences in their suitcases, and the menu would query for these and use them to adapt its offerings to each diner's needs.

We set five teams of student designers to creating design concepts for SL-enabled applications for particular user communities. One team (Robin Dodge, Sidarth Khoshoo, Paul Miller, and Ping Wang) chose to design for dedicated movie-goers. They envision a "movie card", the size of a somewhat thick credit card, connected by a continous wireless link to the Internet and spontaneously to a variety of wireless devices that it might discover in its environment. It has a thumb-pad for purposes of authenticating users (e.g., for payments), a simple flat-screen video display (e.g., for movie previews), and a small number of touch-sensitive spots for taking actions by selecting items from simple menus. The card automatically detects when it is near a movie theater, and displays the available movies, their show times, which ones are sold out, and so on, and offers to show previews, collaborative filtering ratings, and so on. The user's suitcase will also contain debit information, so that individuals can pay for movies simply by holding their thumbs against the thumb-pad while walking through a movie-theater turnstile; the card and turnstile detect one another's presence and perform the transaction. Several individuals can enter the theater using a single card, just by handing it to the next person over the turnstile once they enter. Individuals' suitcases can also carry a library of previews, movie ratings, parental permissions, fan materials, and so on.

The movie card obviously raises the question of generality. The market will presumably settle on a limited number of information technology hardware platforms, ranging all the way from general-purpose computers with modern operating systems to simple embedded devices. Although the movie card is probably not a viable product on its own, one can imagine the range of uses that its hardware platform might find. Contexts similar to movie-going come to mind, such as amusement parks. Others will surely be found. The challenge is designing a platform that works for the whole range of uses. Contrasting uses such as the menu and the movie card raise the question of what kinds of proximity the devices might be able to detect and represent. But the hardware platform is the least of the design problems. Middleware is the hard part. What is the full range of applications that need to discover one another? With what vocabulary do users program their devices to screen potential overtures from other devices? Many devices, surely, will offer only annoying advertising or invasions of privacy. Personal identity will surely need to be protected for many purposes, while being revealed for others. Priorities may need to be established in cases where several applications could run at once. Every context in which an SL-enabled device might be used presents its own complex of difficulties, structured as it is by the practical logistics and the institutional meanings of the actions that the various devices and their owners are taking.

Although the movie card was designed with some sensitivity to the practical issues involved in going to the movies, another design exercise from our project explored in greater depth what it means to express the values of a community in the many aspects of a design. The "World Garden Network" (a conceptual design project by Heather Cleary, Rich Gazan, Heather Hessel, Nancy Hunt-Coffey, and Darin Plutchok) is a set of four types of wireless devices for community gardeners. A community garden is a cooperative association in which gardeners subdivide a plot of generally urban land, keeping their own individual gardens while sharing various resources and tasks. The four devices are as follows:

(a) The Firefly is a simple communications device that gardeners can use to solicit help from gardening experts. Within the community gardening movement, some individuals take a special interest in identifying "varmints" such as crop-eating insects, and these people are happy to be on call when potential varmints need to be identified. This is not only a necessary social function within the community; it is also a deeply symbolic function, expressing as it does an ethic of sharing and a spontaneous type of apprenticeship between advanced gardeners and relative beginners. The Firefly, then, is adapted to the rough and dirty environment of the garden. It includes a simple video camera, a microphone and speaker, a single switch, and a connector for the battery recharger. It is operated entirely by voice-recognized commands, one of which calls out to currently available varmint experts. Once an expert responds and establishes a voice link, the video camera can be used to show the varmint. A range of other community-related commands are available as well.

(b) Simple teleoperated devices permanently located in the garden itself, such as pointable video cameras and water sprinklers can be controlled wirelessly.

(c) My Garden Window is a more traditional wireless device. Resembling a medium-sized framed photograph that one might keep on the mantle, it is a portable video screen that is fed by one of the teleoperated Web-cams in the garden. (The image is superimposed with a thermometer reading and icons for some significant garden conditions.) Gardeners can use their My Garden Window to keep an eye on their crops. They can also press a small number of touch-sensitive points on the screen to activate the teleoperated sprinklers and perform other remote functions. My Garden Window is sufficiently robust to be moved about or carried in one's purse. It is also waterproof, to facilitate use as a virtual window by the kitchen sink.

(d) The Garden Weasel is a wireless device in the garden itself that bridges among the various devices. It listens to the Fireflies and teleoperated devices and connects them to the Internet, and it also communicates between the My Garden Windows and the teleoperated devices in the garden. Because a community garden has little effective security, the Garden Weasel must be camouflaged. Having no display or moving parts, or indeed any other need for physical connection with the outside world beyond a power cable, it can easily be disguised, and the designers envision rock cladding and garden lights that can either hide a Garden Weasels or serve as a decoy.

In each of these cases, the information services are woven into the customary places and activities of the people who use them. They are also woven into the life and values of the user community. They fit into the physical rhythms of the activity, and they fit with a world of existing artifacts. They take account of the relationships among the various parties (movie-goer and theater operator, beginning gardener and varmint expert). They take account of the meaningful proximities among devices, and when necessary they are invisible. This, perhaps, is what Weiser and Brown (1995) call "calm technology": the focus is on seeing movies and growing crops, and not on the physical and interactional demands of the computer. And the technology represents an intervention in the life of the community, not by doing any violence to the values and customs of the community, but precisely by expressing and amplifying them. The movie card eliminates many of the logistical and informational hassles of movie-going, while permitting the movie-goers' to deepen their immersion in movie culture. The World Garden Network expresses and amplifies the community gardening movement's values of sharing information and ongoing attachment to their plot and crops. The design practice, in short, does not try to reconfigure the forces and meanings of the communities. Quite the contrary, it moves with them. But it also takes a stand about values in the meanings and forces that it chooses to express and amplify.

These few anecdotes of conceptual design hardly constitute a revolution in the design of information services, and we want to be modest in our claims for them. Nonetheless, they do illustrate a number of the emerging trends in the development of digital design. Above all, they point toward a synthetic approach to the central tension of design at this point in history: how to reconcile the values of conservatism and rationalism in design, and the seemingly opposite pulls of the responsibility to context and the desire to change the world. The future hybrid sociotechnical practice of information service design will take these trends much further: more systematic analysis, deeper appreciation of the interaction between symbolic and practical aspects of design, best practices for the evolution of middleware, and a heightened appreciation of just what it means, both technically and morally, to design a wired life.

Footnotes

[1] The anarchist critique of rationalized design actually predates this broad cultural development; the Welsh industrial designer John Chris Jones, who had once published the definitive systematization of rationalized design methods (Jones 1970), turned during the 1980s toward a much more generalized approach to design as a way of life (Jones 1991). Like many others, he came to view rationalizing design methods as cognitive blinkers that prevent a fresh engagement with the materials and circumstances of design, and he pioneered methods for using randomness for force designers into engaging with the elements of reality that do not fit their preconceived models (Mitchell 1993). Lately he has turned this approach to digital media (Jones 2000).

References

Philip E. Agre, Designing genres for new media, in Steven G. Jones, ed, CyberSociety 2.0: Revisiting CMC and Community, Newbury Park, CA: Sage, 1998.

Donald Albrecht, Ellen Lupton, and Steven Skov Holt, Design Culture Now: National Design Triennial, New York: Princeton Architectural Press, 2000.

Liam J. Bannon, From human factors to human actors: The role of psychology and human-computer interaction studies in system design, in Joan Greenbaum and Morten Kyng, eds, Design at Work: Cooperative Design of Computer Systems, Erlbaum, 1990.

George Basalla, The Evolution of Technology, Cambridge: Cambridge University Press, 1988.

Victoria Bellotti and Sara Bly, Walking away from the desktop computer: Distributed collaboration and mobility in a product design team, in Mark S. Ackerman, ed, Proceedings of the ACM 1996 Conference on Computer Supported Cooperative Work, New York: ACM Press, 1996, pages 209-218.

Harold J. Berman, Law and Revolution: The Formation of the Western Legal Tradition, Cambridge: Harvard University Press, 1983.

Geoffrey C. Bowker, Susan Leigh Star, William Turner, and Les Gasser, eds, Social Science, Technical Systems and Cooperative Work: Beyond the Great Divide, Erlbaum, 1997.

Graham Button, ed, Technology in Working Order: Studies of Work, Interaction, and Technology, London: Routledge, 1993.

Stuart K. Card, Thomas P. Moran, and Allen Newell, The Psychology of Human-Computer Interaction, Hillsdale, NJ: Erlbaum, 1983.

Geoff Cooper and John Bowers, Representing the user: Notes on the disciplinary rhetoric of human-computer interaction, in Peter J. Thomas, ed, Social and Interactional Dimensions of Human-Computer Interfaces, Cambridge University Press, 1995.

George W. Fitzmaurice, Situated information spaces and spatially aware palmtop computers, Communications of the ACM 36(7), 1993, pages 38-49.

Christiane Floyd, Outline of a paradigm change in software engineering, in Gro Bjerknes, Pelle Ehn, and Morten Kyng, eds, Computers and Democracy: A Scandinavian Challenge, Aldershot, UK: Avebury, 1987.

Andrew L. Friedman, Computer Systems Development: History, Organization and Implementation, Chichester, UK: Wiley, 1989.

Neil A. Gershenfeld, When Things Start to Think, Holt, 1999.

George Gilder, Life after Television, New York: Norton, 1992.

Saul Greenberg, ed, Computer-Supported Cooperative Work and Groupware, London: Academic Press, 1991.

Kaj Gronbaek, Jonathan Grudin, Susanne Bodker, and Liam Bannon, Achieving cooperative system design: Shifting from a product to a process focus, in Douglas Schuler and Aki Namioka, eds, Participatory Design: Principles and Practices, Erlbaum, 1993.

Jonathan Grudin, Groupware and social dynamics: Eight challenges for developers, Communications of the ACM 37(1), 1994, pages 92-105.

Friedrich A. Hayek, The Counter-Revolution of Science: Studies on the Abuse of Reason, Glencoe, IL: Free Press, 1952.

Marina Jirotka and Joseph A. Goguen, eds, Requirements Engineering: Social and Technical Issues, San Diego: Academic Press, 1994.

J. Christopher Jones, Design Methods: Seeds of Human Futures, London: Wiley, 1970.

John Chris Jones, Designing Designing, London: Phaidon Press, 1991.

John Chris Jones, The Internet and Everyone, London: Ellipsis, 2000.

Rob Kling and Margaret S. Elliott, Digital library design for usability, in John L. Schnase, John J. Leggett, Richard K. Futura, and Ted Metcalfe, eds, Proceedings of Digital Libraries '94: The First Annual Conference on the Theory and Practice of Digital Libraries, College Station, Texas, June 1994, pages 146-155.

Klaus Krippendorff, ed, Design in the Age of Information, School of Design, North Carolina State University, 1996.

Robert Lilienfeld, The Rise of Systems Theory: An Ideological Analysis, New York: Wiley, 1978.

David G. Messerschmitt, Networked Applications: A Guide to the New Computing Infrastructure, Los Altos, CA: Morgan Kaufmann, 1999.

William J. Mitchell, City of Bits: Space, Place, and the Infobahn, Cambridge: MIT Press, 1995.

Thomas P. Moran and John M. Carroll, eds, Design Rationale: Concepts, Techniques, and Use, Mahwah, NJ: Erlbaum, 1996.

Nicholas Negroponte, Being Digital, New York: Knopf, 1995.

Richard R. Nelson and Sidney G. Winter, An Evolutionary Theory of Economic Change, Cambridge: Harvard University Press, 1982.

Donald A. Norman, The Invisible Computer: Why Good Products Can Fail, the Personal Computer Is So Complex, and Information Appliances Are the Solution, Cambridge: MIT Press, 1998.

Bruce Nussbaum, Welcome to 2010, Business Week, 6 March 2000, page 102.

Vicki L. O'Day, Daniel G. Bobrow, and Mark Shirley, The social-technical design circle, in Mark S. Ackerman, ed, Proceedings of the ACM 1996 Conference on Computer Supported Cooperative Work, New York: ACM Press, 1996, pages 160-169.

Nicky Pappo, Middleware bridges Internet, wireless, Telecommunications 34(5), 2000, page 47.

Phil Patton, For the mall rats, a new piper, New York Times, 14 January 1999, page F1.

Alex P. Pentland, Smart rooms, Scientific American 274(4), 1996, pages 68-76.

Walter W. Powell and Paul J. DiMaggio, eds, The New Institutionalism in Organizational Analysis, Chicago: University of Chicago Press, 1991.

Nathan Rosenberg, Technological interdependence in the American economy, in Inside the Black Box: Technology and Economics, Cambridge: Cambridge University Press, 1982.

Donald A. Schon, The Reflective Practitioner: How Professionals Think in Action, New York: Basic Books, 1983.

Mark Weiser, Some computer science issues in ubiquitous computing, Communications of the ACM 36(7), 1993, pages 74-84.

Mark Weiser and John Seely Brown, Designing calm technology, unpublished MS, Xerox PARC, December 1995. <http://www.ubiq.com/hypertext/weiser/calmtech/calmtech.htm>.

Terry Winograd, ed, Bringing Design to Software, Addison-Wesley, 1996.