The country houses of Victorian England were curious places. Often situated on large estates, away from the industrializing towns and cities, they were the ancestral homes of aristocratic families and the playgrounds of the nouveau riche. It was in these grand houses that many of the domestic technologies that became commonplace in the 20th century were first to be found—namely plumbing, central heating, electric lights, and telecommunication. In their early experimental forms, these technologies functioned to publicly signal the wealth, influence, and learning of their masters. Sites like Lord Armstrong's Cragside were showcases of the new technologies, for the wonderment of guests. Over time, these large, demanding organizations became reliant on and enabled by the technologies, with a staff of unseen servants operating below stairs. As archaeologists Marilyn Palmer and Ian West comment, "planners devised ways of keeping the servants out of sight in the course of their duties as far as possible, burying them in basement walkways or service tunnels" . Beyond architectural structures, a series of domestic communications systems evolved through which to issue remote instructions: pull cords mechanically coupled to a system of sprung bells, buttons wired to electric bells, speaking tubes, and, finally, household telephone exchanges. These systems enabled the command of the home while holding its operation at a distance—the servants' conditions were vastly inferior to those of their masters and these technologies occluded the visibility of their consequent struggles.
From a European or North American perspective, the design intention to hide the work done by complex systems is to be found not only in the Victorian country house but also through the mass electrification of suburban homes in the 20th century and an automated postwar push-button era, then again in the home computing of the 1980s, the ubiquitous computing agenda of the 1990s, and finally in today's Internet-connected domestic things. This article unpacks some of this historical story and argues that while simplification and automation are seemingly commonsense responses to complexity, an indifference to the implications of invisible labor can regularly create situations in which people and resources are unwittingly exploited.
→ Making complex systems seem simple or automatic can often mean hiding the labor and the consumption of resources on which they rely.
→ An indifference to complexity can create situations in which people and resources are unwittingly exploited.
→ The invisible computer demands a visible user, who in turn can be subject to hidden exploitations.
As the technologies of electricity, gas, water, and telephony moved into densely populated cities from the country homes where they had been pioneered, they became organized and industrialized, first at metropolitan and then national scales. With suburbia came the growth of private housing and a reliance on public utilities and appliances rather than private domestic services and servants. By the 1950s and 1960s, these suburban homes began to consume these utility services through an explosion of new, mass-produced domestic electric appliances. Historian Ruth Schwartz Cowan's critique of these labor-saving technologies concluded that instead of reducing housework, they made it more solitary and time-consuming as ever-higher standards of outcome were demanded. The feminist sociologist Arlene Kaplan Daniels's analysis of invisible work reveals this private work of the home and the ways in which historically women's domestic labor was considered unremarkable and thus devalued. The housework inside a private home is then not just invisible to the outside world but also to the men within, who dismiss it as simply the moral order of things and perhaps even automatic.
The utilitization of services also hides work in another way, as the complexity of their production becomes hidden within and beyond the walls. Consumers typically are oblivious to the circumstances and impact of the distant production of these standardized services that arrive through standardized plugs and sockets. As designer James Auger and his colleagues point out, "Electricity, as a form of energy, comes through sockets on the wall that deliver a seemingly endless supply. These ubiquitous and generic sockets determine the design of every electrical product, providing a neat end to the designer's role and responsibility" .
The push button or switch became a near-universal solution to command these new domestic electric appliances, with clear parallels to the operation of bells in the Victorian country house. Consider the practical ubiquity of the light switch, the television remote control, and the electric doorbell to this day. The notion of the push button as a simple empowering action is prevalent from the 1950s, in corporate domestic fantasies like the RCA Whirlpool Miracle Kitchen (1956) and the Westinghouse Total Electric Home (1960), then forebodingly in the Cold War notion of the nuclear button. The Merriam-Webster Dictionary defines push button as "using or dependent on complex and more or less self-operating mechanisms that are put in operation by a simple act comparable to pushing a button." Simplicity is created by a gesture that triggers work of an unknown complexity in an unknown location for a tangible result. Unlike the visible mechanics of the Victorian country houses' pull-cord systems, these buttons operate invisible electrical signals, infrared light, or radio waves, which contributes to their inscrutability. The button is the input to a black box, with a simple contract, that creates some output by processes we need not concern ourselves with: "A black box contains that which no longer needs to be reconsidered, those things whose contents have become a matter of indifference" .
The push button is another means of rendering labor invisible and establishing relationships of indifferent authority.
The push button is then another means of rendering labor invisible and establishing relationships of indifferent authority, with the same intention the masters had in the Victorian country houses. As media scholar Rachel Plotnick puts it:
To push a button represented a particular fantasy of what I have termed digital command, where (certain) hands could direct anyone or anything to submit to their will. No longer did "manual" refer to effort and strain; rather, the gentle or "mere" touch of a button promised that only fingertips need engage with bells, lights, vending machines, elevators, or cameras. This "reversal of forces" that centered on hand practices—where a small human force could put great electrical forces into motion—suggested that human beings had truly tamed nature by sublimating it to a push .
There is a tension, however, between the mastery of the button pusher and the indifference to the workings of the black box—the operation of an inscrutable button can have uncertain (and potentially invisible) outcomes. This was played out to terrifying effect in Richard Matheson's short story "Button, Button" (1970), later adapted as a Twilight Zone episode of the same title (1986): A mysterious caller presents a struggling couple with a button that if pressed will cause someone they don't know (and who will remain unknown to them) to die, but the couple will also receive $200,000, tax-free. Will they overlook the hidden automated consequences of pressing the button in favor of the more obvious and desired outcome? The simple consequential command of the Victorian master's bell pull had become complicated by the dizzying outcomes of a push-button electric appliance and remote operation, now to be found in almost every home.
Early home computers, like the Altair 8800 (1975), were also commanded by switches, but these directly manipulated the system's internal state, made visible by a series of LEDs. These interactions with the machine became richer through (esoteric) text command-line dialogues and direct graphical manipulation once they were mediated by screens, keyboards, and, eventually, pointing devices like mice. The simple contract of a mechanical button press was now insufficiently expressive to be useful. Instead, these interactive systems present some model of the system to the user, making its complexity and its work (somewhat) visible. It became the work of interaction designers to determine what is seen and what is not in their design of such on-screen system representations. These early home computers were on the whole relatively simple systems operating in isolation, largely uncomplicated by the outside world, and thus straightforward metaphors (like the desktop) describing their work generally sufficed. By the end of the 1980s and early 1990s, however, entanglements of people, their stuff, their homes, and computer networks were growing in complexity and these representational interaction paradigms seemed to promise only to overwhelm users with information and complexity.
In 1991, Mark Weiser of Xerox PARC described the company's ubiquitous computing (ubicomp) vision in a Scientific American article, stating that "the most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it" . Ubicomp, then, explicitly seeks to manipulate the visibility of complexity, and by extension the visibility of backstage work and resource usage. Weiser goes on: "The first wave of computing, from 1940 to about 1980, was dominated by many people serving one computer. The second wave, still peaking, has one person and one computer in uneasy symbiosis, staring at each other across the desktop without really inhabiting each other's worlds. The third wave, just beginning, has many computers serving each person everywhere in the world. I call this last wave 'ubiquitous computing' or 'ubicomp'" . Since then, ubicomp has come to be understood by the apparent ubiquity of computing resources (and network technologies). Yet, this attempted utilitization of computing, meeting a desire for everything, everywhere, all the time, denies scarcities and hides the realities of everyday struggle.
Ubicomp is also a fundamentally different interaction paradigm. Unlike the explicit commands issued by a push button or the interactively negotiated outcomes of a textual or graphical interface, ubicomp seeks to build a real-time contextual understanding of the user, their desires, and their physical environment to allow the machine to determine its best anticipatory action—according to some normative model of behavior. A function of ubiquitous computing, then, is ubiquitous surveillance. The more context that is known to the system, the more functionality the system can provide. The invisible computer demands a visible user under the gaze of the machine.
Modern homes are of course palimpsests of every previous age, not least in the interaction paradigms they employ for complex systems. Push-button command, representational dialogues, and forms of ubicomp are easily identified, but they now more routinely reach into the network and implicate further remote unseen work and resource consumption. Homes have become increasingly dependent on the work of people and machines from across the network; the Internet starts to look very much like a utility service, with all that implies.
The invisible computer demands a visible user under the gaze of the machine.
Much of this work has become so tightly specified and technologically mediated that it can be initiated by remote software functions calls across the Internet. Here, it becomes impossible to distinguish the work of humans and machines—these are Mary Gray and Siddharth Suri's ghost workers  in the system and, increasingly, in the home. Like Victorian servants working below stairs, they operate, as Peter Reinhardt puts it, below the API . The API (application programming interface) is the contract by which software describes to the outside world its possible functions, their required inputs, and the output they will produce—a black box description of the system. APIs are the currency of modern software development practice where cloud-based services are reused, but the contents of these black boxes remain critically unexamined and unseen by both developers and users.
So far it has been the indifferent user who has unwittingly initiated the exploitation of unseen people and resources, typically remote. However, in the context of Shoshana Zuboff's analysis of surveillance capitalism this is brought full circle, and it becomes clear that the simplifying narratives of invisible computing can be exploitative of the user as well. From Zuboff's perspective, the surveillant quality of ubicomp becomes quite problematic in the home. Weiser had cautioned from the start that "overzealous government officials and even marketing firms could make unpleasant use of the same information that makes invisible computers so convenient" . However, there was an implicit contract that only sufficient data is collected that enables some convenience for the user. Zuboff's The Age of Surveillance Capitalism suggests that today's domestic IoT products invert this equation—that a surplus of behavioral data is produced to enrich existing online big data collection for the purpose of market prediction and manipulation, while offering little contextual interaction in return. For instance, interactions with voice assistants are on the whole simple, explicit commands—"Hey, Alexa, start a timer" or "Hey, Google, turn off the lights"—yet they allow the collection of far more data than is necessary for the task at hand.
The Amazon Dash Button (2015–2019) is an interesting, if short-lived, example of the interweaving intentions of IoT, ghost work, surveillance capitalism, and indifference to black box complexity. The Dash Button was a single-button WiFi device that when pressed would instantaneously place an Amazon order for the specific mundane product with which it was associated. Multiple buttons could be positioned around the home to be available at the opportune moment—for instance, a toilet roll button in the bathroom or a laundry detergent button on the washing machine. The product was marketed with the slogan Place it. Press it. Get it. Daniel Rausch, vice president at Amazon, said the company explicitly aspired "to make shopping disappear." The automatic work consequent of a button press happens in and behind the cloud, specifically in the Amazon Web Services (AWS) cloud, via which networks of machines and people are invisibly orchestrated to deliver the desired product to the right home.
Despite a history of domestic push buttons, the Amazon Dash Button was received with some puzzlement. Launched on April 1, 2015, the BBC sought reassurance from Amazon that this was not an April Fools' joke; by the time the product was discontinued CNET described them as the "goofy forerunners of the connected home" . Somehow these buttons represented an alternative imaginary of the smart home, at least in the eyes of some media commentators. This might, in part, be through their association with products like toilet paper and laundry detergent—mundanities that are meant to be invisible in the aspirational smart home, not materialized publicly as colorful buttons.
While the Dash Button reportedly sold in the millions, Amazon discontinued it in 2019. Since then, the company has continued to expand its range of Echo devices. Could it be that the Dash Button produced an insufficient surplus of behavioral data for the company's purposes? The data generated by a button press is far more modest than the rich interaction with a voice assistant. Simply put, did Amazon end the Dash Button because it didn't afford enough surveillance?
This article has unpacked some of the interactional paradigms to be found in homes today from a historical perspective that goes back to Victorian country houses. It has argued that there is often a design intention to hide complexity or be indifferent to it, and that this creates seemingly automatic systems in which people and resources can be unwittingly exploited. In the context of feminist framings of invisible work and the climate crisis, the exploitation of people and resources hidden by black boxing becomes intensely jarring.
This challenges the conventional strategies for designing simple and easy-to-use technologies, practiced by many in the HCI community, and it makes a particular challenge to the agenda of ubiquitous computing in the context of surveillance capitalism. Ubicomp still has a peculiar academic veneration after 30 years, a situation that Paul Dourish and Genevieve Bell highlighted back in 2006 . According to Google Scholar metrics, Weiser's Scientific American article has been cited more than 19,000 times since 1991 and more than 700 times in 2021 and 2022 alone. This article, however, perhaps more pointedly suggests that modern software engineering practices have become too indifferent to the systems they construct.
How then might we design with less indifference to complexity? This is a question I am currently engaging with through the REAPPEAR project, where we seek to identify design patterns that begin to make visible the invisible computer. Through a technically engaged research through design inquiry, we are exposing something of the black boxes we encounter, especially those relating to network technologies. We are then drawing on the interaction paradigms discussed here in seeking to represent some of the complexity of these systems and challenge some of the established assumptions of design simplicity—toward more mindful forms of computing. This is not going to be easy, but it is urgent.
I am extremely grateful for the guidance of Bill Gaver and Andy Boucher in developing the thesis work on which this is based and their feedback on drafts of this article. My thesis was supported by the Arts and Humanities Research Council (AHRC) Design Star Centre for Doctoral Training. REAPPEAR is a collaboration with Nick Taylor, Jon Rogers and Jayne Wallace, funded by the EPSRC PETRAS project (EP/S035362/1).
3. Callon, M. and Latour, B. Unscrewing the big leviathan: How actors macro-structure reality and how sociologists help them to do so. In Advances in Social Theory and Methodology: Toward an Integration of Micro and Macro-Sociologies. Routledge, London, 1981, 277–303.
4. Plotnick, R. Power Button: A History of Pleasure, Panic, and the Politics of Pushing. The MIT Press, 2018; https://doi.org/10.7551/mitpress/10934.001.0001
5. Weiser, M. The computer for the 21st century. Scientific American (International Edition) 265, 3 (1991), 66–75; https://doi.org/10.1038/scientificamerican0991-94
8. Reinhardt, P. Replacing middle management with APIs. Feb. 3, 2015; https://rein.pk/replacing-middle-management-with-apis
9. Rubin, B.F. Amazon stops selling Dash buttons, goofy forerunners of the connected home. CNET. Feb. 28, 2019; https://www.cnet.com/news/amazon-stops-selling-dash-buttons-goofy-forerunners-of-connected-home/
10. Bell, G. and Dourish, P. Yesterday's tomorrows: Notes on ubiquitous computing's dominant vision. Personal and Ubiquitous Computing 11, 2 (2006), 133–143; https://doi.org/10.1007/s00779-006-0071-x
David Chatting is a designer and technologist. He recently defended his thesis, "A Network of One's Own: Struggles to Domesticate the Internet," at Goldsmiths, University of London. He is currently at Newcastle University's Open Lab, engaged on the REAPPEAR project. [email protected]
Copyright held by author. Publication rights licensed to ACM.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2023 ACM, Inc.