Open Doors
  Who Is Who  

Amsterdam - 14, 15, 16 November 2002

Digital Ground: Fixity, Flow and Engagement with Context

(this article was published in Archis Magazine, in their special November Flow-Issue)

Malcolm McCullough

Flow needs contexts. A river, for example, needs riverbanks otherwise it spreads out in every direction until it becomes a brackish swamp. Similarly, cars need highways, capital needs markets and life’s energy needs bodies through which to course.
Flows influence one another. For example, we know that telecommunication generates transportation at least as often as it substitutes for it, starting with Alexander Graham Bell whose first words on his new telephone were ‘Watson, please come here’. Similarly, when you order a book from Amazon, the flow of data on the web has an effect outside the web, namely it causes a package to be put on an airplane. This in turn has geographic consequences: the warehouse where your order is filled is probably located near an airport.
Where regular crossovers between flows occur, places emerge. Amsterdam, for instance, is principally an entrepot. And much as goods came off the boats and onto carts here, so too did ideas. Like Amsterdam, most of the world’s great global–digital cities are also seaports. Some of these cities move more voice and data than most nations.

For a vivid emblem of how a city can act as a switch, the environmental historian William Cronon has drawn our attention to the invention of the grain elevator in late nineteenth-century Chicago. A grain elevator is a switch that directs an incoming flow – in this case of soybeans, wheat or corn – into a graded bin. This had remarkable consequences, beginning with a fundamental act of commodification: instead of buying three tons of, say, farmer Smith’s wheat from Independence, Iowa, you now bought three tons of Grade A wheat from the elevator company, without regard to its origins. Not only did this eliminate the impedance of so many micro-transactions, it also allowed the accumulation of reserve capacity, and, infamously, market speculation on that. And of course it allowed much, much more voluminous Flow. Furthermore this was inherently linked to the latest network technology of the age: railroads. Eastbound grain paid for an unprecedented concentration of railroad building whereby Chicago’s hinterlands soon fanned out for a thousand miles across the prairies. Railroads have notoriously high fixed costs, however: once they are there, something had better ride on them. So instead of the cars riding back west empty, they were priced to favour centralized volume trading in merchandise. It is no accident that Chicago’s merchandise mart was among the world’s most massive buildings, or that its tallest tower belongs to Sears, Amazon’s obvious precedent in highly centralized retailing. Indeed, as you can read for yourself in Cronon’s superb book, Nature’s Metropolis, the cultural geography of half the American interior was shaped by resource Flows through Chicago.

The pertinence of cultural geography to design in the space of flows is quite simple. When designing flows of information and interaction, it is necessary to recognize the increasing importance and value of context. But the time has come to understand the interactivity contexts beyond the scale of the task at hand. In particular, look for the crossovers between infrastructures, and expect to add design value right there, as with fixtures and fittings. Conversely, don’t always expect new flows to do away with older ones, or to prosper without them. And, by extension, don’t expect cyberspace to be a world apart.

For the sake of argument, let us call this attitude toward mediated fixity ‘digital ground’. In a series of terms we are here at Doors to explore, that expression conveniently describes how interactions, embodied in the periphery, through pervasive computing, do in effect take place. It represents one response to the design challenge of pervasive computing. Its explanation begins from some paradigms given to us by Silicon Valley, continues with some wisdom we think we have gathered in the interaction design community and leads towards new outlooks on the role of interactive technology.

From universal to local

Intrinsic to pervasive computing is a paradigm shift beyond cyberspace. The alleged passing of cyberspace may seem absurd to anyone who conflates it (and much else as well) with the technologies of the Internet, which is hardly going away. What is passing are expectations of one single coherent ‘there’ on the other side of the looking glass, which would supersede any previous architecture of the object of design. When a term spreads through a culture very quickly, it often represents a current way of seeing the unexplainable world. Then the metaphor wears off; the need to explain new technologies in terms of older realities generally tends to diminish. Thus did carriages cease to be horseless. ‘Ban cyberspace,’ ran a headline in The Economist back in 1997 – hardly just yesterday – ‘The word, not the thing itself, whatever it may be. And dump the rest of the Internet’s lame metaphors too’. When, at the turn of millennium, the Association for Computing Machinery, the world’s largest membership organization of information technology researchers, launched a general-readership publication named Ubiquity and named its plenary conference After Cyberspace, the paradigm shift had become more or less official.
After all, the cutting edge dulls in everyday life. Sometimes the technologies on which new expectations are based disappear into the fabric of everyday existence, as did the telephone. Likewise the Internet has begun to fade into banal, unlovely normalcy. Other technologies are rejected for errors of principle. Much as bloodletting turned out to be a misguided approach in medicine, so immersive virtual reality left out some important details – such as the fact that we orient spatially not just with the eyes, but also with the inner ear and with joint kinesthetics. Then too, right as they are at one time, some technologies are later surpassed by unforeseen alternatives: Chicago’s radiating freight trains eventually gave way to random-access interstate trucking. And so might, for one or more of these reasons, our present universalist, one-size-fits-all version of information technology design computing, give way to something that is more responsive to context.
Even in purely technical terms, pervasive computing points towards more local arrangements. There are several reasons for this. The foremost software question in pervasive computing is who is present, and what they are trying to do, here. In hardware, net addresses remain bound to the device addresses of connectors, which in turn are managed by physical location. In protocols, if mobile devices are to become less obtrusive they must recognize encountered situations. In information design, the geodata industries are exploding. In systems integrations, specialized site-embedded technologies cannot operate independently in ever-larger numbers, but must somehow recognize and corroborate one another. This requires another level of software abstraction: location models. Those models effectively represent unique local aspects of organizations and their work (or play). Lastly and most of all, these accumulations of technology still need to be housed, maintained and tuned by their owners – not everything will fit in your pocket. Like architecture, those accumulations may still come to represent owners to their constituents. They also ground and cue particular activities in particular places.


So what is Ground, and what has it got to do with Flow? The word ‘ground’ originally emerged from much earlier words for ‘bottom’. Its most usual meaning is some sort of foundation. In the graphic arts, for instance, the word applies to both the preparation and the visual role of a base on which figurative or relief objects appear. In academic endeavours, it describes an established theoretical base on which some current argument stands. In psychology, its connotation of firm support extends into subjective constructions. For example to be ‘grounded’, psychologically, is to have stability, resilience and repose. In phenomenology, a sense of groundedness is more distinct where the ground has been modified: a horizon is more powerful in the presence of a tower, for instance.
Let us pin down a few more terms, at least for the present argument. Usability, ever the first goal of interaction design, may be first-time conceptual accessibility, measurable mechanical efficiency in competent use, or ultimate transparency under practised mastery (as is a fine pair of skis to a skier who thinks not about them but the slope). Usability in context may depend on enterprise data Flow, on effective management of expectations and intentions by good interaction design, or on interoperability with other pieces of technology incidentally present.
Interaction is more slippery still. Badly abused, the word has come to describe just about any relationship between people or things, as if the shapes in a Picasso painting ‘interact’ with one another. More properly, the word implies a combination of reciprocity and deliberation in the exchange of messages – in other words, messages sent in reaction to messages just received, and messages received as a consequence of messages sent. Thus you don’t interact with a book: you just read it. But because of electronic communications, you do interact with other people who are not physically present, or who are taking part in the exchange at some other time. Thus through digital media, we interact indirectly.
Context generally means not an a priori setting, but the engagement of it, as well as the bias that setting gives to interactions that occur within it. In other words, it is not some preexisting space to be filled with objects and events, but an emergent perception of those aspects of surroundings which give meaning to what one is doing. Environment, meanwhile, is the perceived sum of all present contexts, for all actions one might consider taking. Contrary to the popular conflation of this term with ‘nature’, and with planetary ecological issues, in this usage ‘environment’ is local, and can be largely technological. (Meanwhile, beyond present scope but a fine topic for future research, ‘nature’ is getting difficult to define as the absence of technology.)
Two more terms: periphery describes interaction design’s goal for getting technology out of our face. Mark Weiser and John Seeley Brown defined it as ‘background that is outside focal attention but which can quickly be given that attention when necessary’. Finally, embodiment is a state of being in the world, best left to philosophers (such as Lakoff and Johnson: see Philosophy in the Flesh), or to architects, who have known, at least since Vitruvius wrote about it, that the body imposes a schema on space. Or as Paul Dourish has explained (in Where the Action Is), ‘embodiment is an emergent quality of interactions’.
Before all that homework seems like a digression, note that well-embodied interactions tend to involve Flow – in this case Flow of tacit knowledge. Here we arrive at Mihaly Csikszentmihalyi’s often-cited expression: Flow is the sense of engagement that emerges, between boredom and anxiety, when practised abilities are applied to challenges that are just about manageable. This notion of engaged tacit knowledge grounds much interaction design. We tend to be familiar with psychological notions of ‘activity theory’, ‘situated actions’ and ‘persistent structures’. We know how possibilities for action are perceived especially amid engaged activity (and we overuse the word ‘affordance’ to describe this). Increasingly, we understand how that perception depends on persistent structures, both mental and physical, that surround and give meaning to those activities. We recognize how such response to context is not necessarily deliberative. We find the phenomenology of engagement at the roots of interactivity.
So this is the heart of the matter: Flow needs fixity. Persistently embodied intentional settings, also known as architecture, provide a necessary context for Flow.

Architecture as background

Note that in its recent emphasis on active engagement, interaction design has departed from earlier generations of inquiry into environment and behaviour. Where that work once aimed to reduce design to a linear, predictable process, based on measurable models of conditioned response, present work recognizes the importance of expectations. This is why meditation teachers insist that a particular spot in the house be set aside for no other purpose. Information technologists thus come around to the psychology of intent, which surely sounds like subject matter for design. As psychologist Bonni Nardi has explained, intent makes people different from machines in any Flow. Intent gives an asymmetrical cast to the relation between people and things.

The more that principles of embodiment, activity and environmental perception underlie pervasive computing, the more it all seems like architecture. How the physical environment casts an organization’s activity and represents its aspirations now interests information technologists. This is because information technology, like buildings before it, has become social infrastructure. Conversely, architecture has acquired a digital layer, which involves the design of organizations, services and communications and not just, as Le Corbusier put it, ‘the correct and masterful play of forms in light’. As both hardware and software, both architecture and interaction design compose the fixity necessary for better Flow of the tacit knowledge that resides in communities. The word ‘architecture’, which derives from the Greek for master builder, has been appropriated to describe all manner of technological designs that are infrastructural, irreversible and which shape everyday activity in one way or another.

Instead of approaching architecture as attention-seeking form, as many architects do, for interactivity we must understand it as background. Quiet architecture surpasses most other technological productions at institutionalizing spatial arrangements to the extent that they shape cognition. A culture’s perennial spatial forms reflect and perpetuate a particular cognitive background. This is why one of the best criteria for appreciating architecture is whether it is memorable.

Perhaps one key to meeting the design challenge of pervasive computing is to reexamine the connections of body, memory and architecture. This suggests typological approaches to design. The older and more persistent the grounding structure, the more likely that it has shaped environmental predispositions. The more resilient the formal type has been amid changes in society and technology, the more power it accumulates as a basis for design. Type is not a mere functional category in this regard, but a generative abstraction, an implicit balance between convention and invention and a living expression of cultural memory. Space does not permit an explanation here, but type has become a key topic of architectural debate. For clues to why this is so, just look at Amsterdam. The often-mentioned beauty of this city does not arise from the contention of so many free-form inventions, but from the subtle inflection of a concise repertoire of workable architectural types. Now generalize from this: a culture’s perennial spatial forms reflect and perpetuate a particular cognitive background. Articulating the value of that background’s uniqueness has become a topic for interaction design.

For much as contextual design of information technologies can no longer stop at the scale of the task, neither can it stop at the scale of the building. Today the right scale for designing in the space of Flows is the region. At least for the more mobile among us, region has replaced the contiguous neighbourhood as the scale we most truly inhabit. Regions are the usual scale of ecosystems, economies and infrastructures. And now these necessarily become the subject matter of design. The architecture of the region has become particularly important in questions of smart growth, industrial ecology and competition for mobile knowledge workers, for example.

If this seems too daunting, too big to address and too far out of control, recall our point of departure in this essay. Places emerge at crossovers between infrastructures. Think of all the Flows that mingle at the regional airport, for example, and of the need for better interaction design in so many situations there. Or consider resource economies in water or power, which at last have become design subject matter in the profligate United States. For more artistic prospects, think of situated interventions in, say, the use of geocoded marking and polling for social navigation on a Friday night on the town. Other classes of contextual pervasive computing include building community, field documentation of technical operations, learning the regional biosphere, expanding recreational statistics, monitoring larger-scale environmental patterns, and so on.

These are all examples of getting into place. Somehow we know that places are more valuable than fiscal markets account them to be. Somehow we must have faith that our new century’s economics will mature enough to represent human, natural and cultural capital more accurately. Then we know design will be more essential than ever. We then wish for interaction design to serve the basic human need for getting into place.

In a world with so many means and so little sense of what to do with them, it seems only fair to raise such large notions of ends. Ultimately, this is how design becomes economically vital, and in the process changes economics. Homo faber, the maker of things, usually tends to add value through means, that is, technology. Yet psychological economics perennially reminds us that things have no value per se – only value to someone. Nothing is of more general value to someone than a roof overhead, something to eat and a cure for what ails them. Nothing, among all the societal strategies attempted by humankind, is more useful for meeting basic needs than modern technological consumerism. But when those material needs have been met, continued industrial production and consumption yields only diminishing returns. No amount of stuff to buy and own can fill the holes left by neglecting those other forms of capital. Similarly, when non-material entertainment needs have been met, continued fantasy production yields only diminishing returns. What is missing, amid so much artifice, is contact with the world itself. There are children in Texas who have never gone for a walk.

But now the world comes calling, with the biggest picture of all: an entire epoch shifts as a consequence of modernity’s encounter with environmental limits. Suddenly we need fewer technologies for overcoming our environmental constraints, and more that help us understand and maintain them. Despite all we love about mobility and Flow, the kinds of non-fiscal capital so essential to this shift are still best found in fixed settings and local service ecologies. Can design help us to recognize and develop the many kinds of capital of which places are the repositories? Can interaction design become more instrumental to natural capitalism?

Malcolm McCullough is Associate Professor at the Taubman College of Architecture and Planning and School of Art and Design. He was one of the speakers at Doors of Perception.

updated Monday 31 March 2003
Address: Wibauthuis, Wibautstraat 3 • 1091 GH, Amsterdam
The Netherlands • T +31 20 596 3220 • F +31 20 596 3202
Doors of Perception 2002. We are happy for this text to be copied and distributed, as long as you include this credit: "From Doors of Perception: www.doorsofperception.com".
Want to send us your comments? Email desk@doorsofperception.com