At the present time the conditions of town life have changed. We can no longer utilize most of the rejected food matter for feeding, except in small towns. We now discard more rubbish, our fuel varies more as to its kind and the quantity used, we can no longer use the older and crude methods of collection and delivery, we have materially increased the distances to which refuse must be delivered and we are abandoning the disposal by dumping except with ashes. The problem of refuse disposal has, therefore, become more complex than formerly, and this complexity may not yet have reached its limit.
In order that correct solutions for the best methods of disposal may be found, both from the standpoint of sanitation and economy, it is necessary to inquire into details far more than formerly, so as to have more definite facts and figures with which to solve the problem. … In short, we must have more special data and statistics before we can indicate the best methods for the disposal of a particular town’s refuse.
At the outset we should know what the refuse consists of, and ascertain and discriminate between its various parts, which may be enumerated as being garbage, dead animals, night soil, manure, street sweepings, ashes and rubbish.
—Rudolf Hering (1912, 909)
Waste has been described as a nuisance, a threat, a source of injustice, and a symptom of excess. Considering the overwhelming materiality of garbage and the urgency of environmental concerns, describing it as “information” might seem inappropriate. But studying waste from an informational perspective does not imply that it is immaterial, or that problems of pollution and overconsumption are imaginary. In some sense, waste is physically embodied information. Much of what we know about the past, we know from things thrown away, the discarded objects that record human activity and the passage of time. Historical dumpsites are of interest to archaeologists and anthropologists, just as contemporary landfills harbor a wealth of information about everyday consumption and behavior (Rathje and Murphy 2001).
Waste can be information in the literal sense. Until a few years ago, data generated by the data infrastructures of telephone providers, online platforms, or traffic control systems were seen as transient, unwieldy, and not worth storing—informational waste. Under the banner of Big Data, these leftovers are now praised as “the new oil” in Forbes magazine. This “digital exhaust” is mined by machine-learning algorithms that try to recognize patterns, classify items, and predict behaviors. The data sets are often collected for unrelated purposes, arrive in incompatible formats, and are stored without a strict data structure—making the data centers that host them a bit like landfills themselves. After all, actual landfills are increasingly mined for methane, metals, and other valuable commodities (Jones et al. 2013).
Viewing waste as information stems from the realization that waste is, above all, a designation. Waste is whatever is labeled as waste, and nothing exists that cannot become waste at some point. In his last book, the posthumously published Wasting Away, urban designer Kevin Lynch defines “waste” as a lost opportunity, “a resource not in use, but potentially useful: wasted time, a wasted life, an empty building or field, an idle machine” (Lynch 1991). The European Union defines “waste” as “any substance or object which the holder discards or intends or is required to discard” (European Commission 2008).
“Waste” is defined through context, relative to a value system. The anthropologist Mary Douglas observed, “Where there is dirt there is system.” As she argues, dirt is a byproduct of ordering and classification, and dirt is what violates this symbolic order (Douglas 1966, 41). In information systems, “dirt” often refers to erroneous data entries or classification errors. Studying such inconsistencies can reveal how data were produced (Loukissas 2017).
In the more prosaic language of economics, Richard Porter defines “waste” plainly as “anything that is no longer privately valued by its owner for use or sale” (Porter 2002, 2). Perhaps not valued by its prior owner, waste can still be valuable for the waste management industry. Like waste, value is a contextual and informational concept, as demonstrated by objects that become valuable antiquities after being discarded as rubbish (Thompson 1979).
Often, the process of becoming waste implies a loss of information, a development that entails an ontological shift. Spanish philosopher Jose Luis Pardo notes, “The process by which something is turned into trash can be described as a process of disqualification” (2006). Discarded objects lose their value by losing their characteristic properties and becoming part of an undifferentiated mass. The awe-inspiring aggregation of matter discovered in a hoarder’s apartment or unloaded from a thirty-ton garbage truck renders the individual components invisible.
But producing waste also generates new information by sorting and separating trash from non-trash (Strasser 1999, 5). At the same time, sorting can reverse the process of becoming waste, restoring hidden information and value. This is the value proposition of a material recovery facility (MRF, pronounced “Murf”), in which co-mingled materials collected from curbside recycling are separated by composition and grade. Discarded plastic, paper, and glass have different commodity values based primarily on their levels of purity, which reflect how well the materials have been cleaned and separated using processes that often require significant resources of water and energy.
One might associate waste with noise rather than information, with entropy rather than order. Once inside the waste stream, however, waste is no longer “matter out of place,” but subject to the regimes of diverse classifications. Not surprisingly, the early study of waste management by Rudolf Hering and Samuel Greeley began with a comprehensive categorization of municipal refuse (Hering and Greeley 1921). The European Union defines different kinds of materials in a regularly updated “List of Waste,” and the U.S. Environmental Protection Agency (EPA) maintains similar lists for different kinds of hazardous wastes. The inclusion of a substance into such lists has far-reaching consequences—defining waste is a highly contested process.
Waste is a globally circulated commodity, the material basis of long-term service contracts, a calculated risk that can “bite back” long after its disposal. Waste systems cannot be separated from systems of production, and notions of value cannot be seen in isolation: “without disuse there is no use, and without waste there is no value” (Gille 2012, 28). Whatever is found in one system bears the imprint of the other, and systems of waste can be instrumental for studying systems of production. Understanding waste as part of a hybrid relationship between culture and nature, as sociologist Gay Hawkins argues, offers a productive alternative to a dualistic opposition between human and nature in which waste can play only a destructive role (Hawkins 2005).
Solid waste management is the quintessential municipal service, the lowest common denominator of local government in the United States. Often, it just means collection and dumping at a local transfer station. But it is the gateway into a larger system connected to many different issues of public health, production, economic development, and urbanization.
The waste system is a global operation, yet it resists mapping. Its global dimension, structure, and flows can only be estimated. The Handbook of Solid Waste Management complains that local governments, policy makers, and waste management professionals lack accurate information about waste systems, a deficit it attributes to insufficient monitoring, the absence of shared definitions of waste in laws and trade agreements, and flawed coordination between administrative systems (Kreith and Tchobanoglous 2002). International estimates are often made without empirical basis or declared methods (Hoornweg and Bhada-Tata 2012). In short, there is no shared information model for waste.
For data enthusiasts, a lack of information alone might justify urgency. But many complex systems, such as the networks of informal recycling and waste picking described in part II of this book, are remarkably robust without any explicit monitoring or information infrastructures. It is therefore important to consider where, outside of pedestrian concerns about efficiency, insufficient data generates problems in the context of waste management. Disputes over waste policies and environmental justice are fought with data, statistical models, and maps describing situations that are difficult to observe. Here, a lack of information can lead to a lack of accountability, which can obscure pollution sources and cloak questionable practices. A lack of monitoring and enforcement can also undermine well-meaning policies, such as when incentives for recycling open loopholes that allow exporting electronic scrap as functional machines for “reuse” or abandoning hazardous materials held in “temporary storage.” Finally, data about waste composition and provenance are instrumental for recapturing value from the waste stream. A circular economy in which discarded materials become new products requires information to differentiate the confusing mess that is waste.
Like commodities, all waste is traded (Porter 2002). Waste is a local issue and a global industry at the same time. Many materials such as mixed plastics are exported from the United States to recycling facilities in China—backhauled by the same vessels that delivered goods to the United States. In their reach and complexity, waste systems resemble global supply chains of production and distribution, with the significant difference that information tends to diminish during waste disposal. While manufacturers, retailers, and consumers can track goods and resources through supply and delivery chains, no such options exist for waste. Cities keep track of recycling rates, but the exported quantities and destinations go unreported (MacBride 2012, 181).
An engineering joke goes, “Practice is when everything works, but nobody knows why,” meaning that complex systems often are not fully understood even by their managers and engineers. I have spoken to solid waste managers in local governments who did not know what happens to the city’s trash beyond the first transfer station, which is typically where information reported by haulers ends. As waste and recyclable materials move through the waste system, it is not uncommon for them to change owners multiple times with little or no information exchanged or preserved in the process. It has been estimated that up to 80 percent of electronic waste ends up in developing countries despite various export bans (Lepawsky and Mcnabb 2010).
One might argue that global supply chains also suffer from information deficits. Consumers rarely know where product components have been sourced. Even manufacturers struggle with their end-to-end systems. Sourcemap, a Boston-based startup, helps companies investigate and visualize their supply chains. But this merely underlines the fact that real-time data about the movement of goods is crucial to the success of these operations, a requirement that has resulted in the standard use of bar codes and radio-frequency identification (RFID) chips to tag parts and packages.
For waste systems, the incentives are structured differently. Knowledge of what happens downstream is a question of public interest rather than a source of economic advantage. Gathering and exchanging information about the waste system is more difficult than in commercial supply-chain management or in other infrastructural systems such as power and water. Yet good governance of the waste system requires reliable information to evaluate whether recycling programs have their intended benefits, whether questionable disposal practices have been “greenwashed” by labeling them as recycling, or whether environmental crimes have been committed.
As phones, dog collars, car keys, fitness wristbands, and a multitude of everyday objects become able to locate themselves and, to much concern, report their whereabouts to unknown recipients in a vast global network, a research question beckons. Can these gadgets partaking in the “Internet of Things” be leveraged to learn what happens inside the waste system? Emerging social practices concerned with tracking waste, documenting pollution, and producing environmental data beg another set of questions: Which roles do such civic technology initiatives and practices of appropriation and hacking play in the governance of public services? How are questions of accountability negotiated, and which roles do data representations and the design of mediating technologies play in this context?
When focusing on the larger goal of understanding the governance of waste systems, it is tempting to take technical methods, data formats, and protocols for granted. However, just as the different actors have their own ways of reading and representing waste systems, these technical details determine what we see when we try to read waste systems through the lens of technology. It is therefore helpful to examine how these technologies encode the material reality into symbolic languages and how they make the waste system legible through human and machine-readable languages, classifications, and social practices.
Until now, I used the term “information” interchangeably in a physical, abstract, symbolic, and material sense. Some clarifications are in order. Anthropologist Gregory Bateson defines a unit of “information” as “any difference which makes a difference in some later event” (Bateson 1972, 386). Although opinions might diverge about how “making a difference” should be interpreted, Bateson’s definition is both broad and to the point, defining information by what it does, not by what it represents. It does not imply an omniscient, objective observer. It works from a subjective perspective in which it does not really matter whether the immediate observer is a human or an image sensor.
In the perspective of natural sciences, information is often treated as equivalent to a physical property, expressed in the “it from bit” hypothesis stating that all physical phenomena can ultimately be reduced to questions of information theory (Wheeler 1990, 311). Such a realist conceptualization of information is less popular in the social sciences and humanities, which mostly deal with information constructed by humans and subject to semantic ambiguities. Bateson’s expression “makes a difference” subtly implies this constructed nature.
Information philosopher Luciano Floridi defines “information” as meaningful data, whereas a datum refers to a single difference, a lack of uniformity in a given context (2011, 85). This could be a difference in electric current, a microscopic bump in the groove of a vinyl record, or the difference between the symbols “a” and “ä.” While this definition of “datum” sounds very similar to the earlier definition of “information,” it implies that a datum is not necessarily meaningful, it does not always “make a difference.” A valid and well-formed data set can be the result of a random process—meaningless.
How the processes and flows of waste systems are captured as data is a central concern in this book. I will use “data” to describe a set of systematic observations that have been symbolically encoded and stored in material form. This working definition implies that data are necessarily constructed through several steps. In order to collect a data set describing a waste stream, a method of observation has to be devised (such as sorting a truckload of garbage), a symbolic system has to be formulated for its representation (for example, a taxonomy of materials), and a method has to be chosen for encoding the observations into symbols. Finally, the observation has to be stored in a physical form. At each step of this process, decisions have to be made and often later questioned, renegotiated, and revised.
Because of the need to interpret data, visual theorist Johanna Drucker has argued that the Latin word datum,1 meaning “the given,” should be replaced by the active form captum, which is Latin for “the taken.” She argues that data do not exist before they are parametrized, but are “constructed as an interpretation of the phenomenal world, not inherent in it” (Drucker 2011). Sharing Drucker’s concern that the term “data” downplays all assumptions, decisions, and actions involved in their construction, geographers Rob Kitchin and Martin Dodge have rigorously used the term “capta” throughout a whole book, introducing terms such as “captabase” in the process. They define “capta” with a more realist flavor as “those units of data that have been selected and harvested from the sum of all potential data. … with respect to a person, data is everything that it is possible to know about that person, capta is what is selectively captured through measurement” (Kitchin and Dodge 2011, 261).
Although, for clarity’s sake, I do not go as far as to adopt the “capta” terminology, when I describe “waste” as “information,” I do not construe information as abstract, but as materially embodied. An item in the waste stream bears material traces of many social, cultural, technical, and political processes that can be scrutinized, whereas the concept of waste remains vague and ambiguous. Treating waste as information means following the heterogeneous network of connections in which a piece of garbage is embedded.
For something usually viewed as a problem, waste continues to fascinate. While there is a lack of information about waste systems, there is no shortage of perspectives and opinions about them. The two issues might be related. As Ernest Hemingway declared, the things that are left out are the most important parts of a narrative: “the dignity of movement of an iceberg is due to only one-eighth of it being above water” (1932, 192). Everyone has a view of the waste system, but as with the iceberg, the viewpoints are based on partial knowledge, and often our imagination is defined by what we do not see.
Iconic incidents in waste folklore have left a strong impression in our collective memory. In 1987 the trash barge Mobro 4000 traveled to Central America and back in an unsuccessful attempt to offload 3,000 tons of New York City’s garbage. Another popular myth claims melodramatically yet erroneously that the Fresh Kills Landfill on New York’s Staten Island, closed in 2001, was the only human-made structure visible from space.2 Detailed histories chronicle the environmental justice struggles from Love Canal to “cancer alley” along the Mississippi River. Many have heard reports about villages in Asia and Africa where electronics containing contaminants like lead are dismantled to reclaim elements like gold, since one ton of electronic waste contains 40 to 800 times the amount of this precious metal that is typically extracted from one ton of gold ore (Bleiwas and Kelly 2001).
While the apocalyptic imageries around Fresh Kills may exaggerate, other popular images underestimate the realities of waste management. For decades, the electronics industry was able to maintain the image of a clean industry despite the toxic legacies of semiconductor production (Gabrys 2013). This lack of information about the infrastructural processes involved in waste generation gives rise to imaginative but inaccurate theories, similar to the way anthropologist Willett Kempton demonstrated how homeowners’ understandings of ubiquitous heat thermostats are inconsistent with technical realities (Kempton 1986).
How would the experiences of waste systems change if the public had more knowledge about their actual processes and geographies? What would we demand from municipalities? How would we express and support our doubts? How would the relationships and the interactions among citizens, governments, and other actors change? In short, how would this knowledge affect the governance of these systems?
Contemporary waste systems have been shaped by many different actors who use their own representational tools. Infrastructure governance and controversies are enacted through charts and tables, site maps and hydrological models, news photographs, and protests. Historian Martin Melosi describes how epidemiologists, engineers, citizens, and activists have approached waste as an issue of public health, an engineering problem, an aesthetic nuisance, or a manifestation of social injustice (Melosi 2004). As the range of these problems attests, the role of sanitation for modern urban planning can hardly be overstated. The overcrowded industrial cities of the nineteenth century were frequently struck by epidemics such as cholera and typhus. Ironically, a key driver of public health reform was a scientific misconception. The “miasma theory” that epidemics spread through contaminated air affecting rich and poor alike helped to establish consensus that these crises could be addressed only at the government level (Tarr 1996, 209).
As Melosi explains, the first to shape the waste systems were public health reformers such as Edwin Chadwick, whose descriptions of the unsanitary conditions among the urban poor set the basis for municipal sewer and garbage collection systems. Civil engineers saw public health reform as a technical and logistic challenge for building citywide sanitation infrastructures. With the City Beautiful Movement, affluent citizens gained influence over urban planning, perceiving waste primarily as an aesthetic problem that negatively affects moral sentiments and property values. Important for society but noxious for neighbors, waste facilities frequently became locally unwanted land uses (LULUs) that generated siting disputes and resulted in waste following the proverbial path of least resistance to marginalized communities. Toxic pollution in these neighborhoods gave rise to the environmental justice movement, which shed light on the politics of waste and contrasted with the goals and concerns of affluent environmental conservationists (Pellow 2004; Bullard 2000).
Even within the waste management community, one finds vastly different perspectives and agendas for composting, recycling, landfilling, waste-to-energy, or zero-waste. Not only do these approaches perceive waste as different kinds of problems, they use different toolsets for conceptualizing, observing, and representing the system. Epidemiologists, for instance, are concerned with spatial distributions of disease and medical pathways while engineers look at material flows and system performance. All representations share a common purpose in modeling the system as a coherent whole that can be investigated, manipulated, and optimized (Peattie 1987, 6). Each model defines inputs and outputs such as cases of disease and pollutants, or material flows and system capacities. Each input can be tweaked to achieve different outcomes that serve as a basis for decisions. By allowing different outcomes, each prescribes a specific perspective that offers a partial view of the system.
Among all of these diverging perspectives, can there be a shared representation? Opening the Global Positioning System (GPS) as a public resource for civilian use has given rise to a large industry of location-based services that has played a substantial role in facilitating the global economy.
Within certain limits, GPS offers a shared language for investigating waste systems expressed in geographical coordinates, shared data formats, and tools for collaborative data collection, analysis, and visualization. Abstract and reductive, these generic representations offer a mere partial perspective that points to issues, stories, and experiences not included in the data. Nevertheless, the data formats and technical protocols are highly mobile, allowing transitions between different scales, contexts, and domains of knowledge.
Within the larger context of such civic media technologies (Gordon and Mihailidis 2016), this book focuses on geolocalization for reading waste systems through their spatial structures, temporal processes, participant connections, and system governance. Facilitated by smartphones and online platforms, civic technologies fall within a participatory culture that engages citizens in issues of public interest. Proponents of civic technologies envision citizens gathering information, tracking spatial processes, and visualizing complex systems in their roles as watchdogs, community organizers, DIY hackers, resource stewards, and expert amateurs (Ratto, Boler, and Deibert 2014; Kuznetsov and Paulos 2010).
Participatory sensing has been used to monitor everything from urban noise to radioactive pollution (Bonner 2012; Maisonneuve et al. 2009). Community-based efforts to collect geographical information have aided humanitarian responses to crises (Meier and Leaning 2009). Civic technologies include initiatives to offer government data in machine-readable formats for public access, allowing for public scrutiny of governmental processes and facilitating new products and services that utilize these data (Lathrop and Ruma 2010). By improving access to local government, civic technologies are assumed to make cities more responsive and accountable, engaging citizens in an ongoing conversation and collaboration with public servants and the public.
These positive effects, however, are often proclaimed rather than studied. Critics of civic technologies caution of inequalities in access and representation, threats to privacy and public anonymity, quality issues of data collected by self-selected volunteers, undue simplifications of complex social issues, or hidden agendas masked by narratives of participation (Jensen and Winthereik 2013; Boyd and Crawford 2012; Morozov 2014; MacKinnon 2012; Toyama 2015; Cooke and Kothari 2001).
Civic technologies often frame urban issues as a problem of participation and information exchange. They are grounded in a belief that the city can be improved by enabling all actors to talk to each other, and by making these interactions fast and effortless. They introduce new representations of the city using the languages of data, interactive interfaces, and dynamic maps that show fine-grained processes in real time. The role of design in mediating this interaction between systems and individuals is generally underappreciated. Interfaces play a central role in facilitating, shaping, and constraining interactions. A critical examination of civic technologies requires a close look at how interfaces and visual representations influence how information is collected, how meaning is constructed, and how action is taken.
While reading waste systems has not been a concern in urban studies, the notion that the city can be read like a text has a long history in urban planning and architecture. It has been applied mostly to city morphology—the shape of plazas, street fronts, and coastlines, as well as the topology of the road network. This book expands the concept of urban legibility to the realm of infrastructure and its governance. We read urban systems not only through our senses and experience, but also through public data repositories, visualizations, real-time data from sensor networks, and the traces left by other users and their actions. It is easy to forget that the experience of a system is shaped by the design of all these things, even if the experience of infrastructure is never fully determined or exhausted in design (Edwards et al. 2007, 28).
Throughout this book, I explore the connection between design and governance. I argue that design is in many ways a form of governance in how it shapes and regulates behavior, interactions, and conversations. Conversely, the processes of governance have many similarities to design processes, often occurring as a series of incremental revisions and negotiations that political scientist Charles Lindblom called “muddling through” (Lindblom 1959).
My central notion is that infrastructure governance is enacted through the representations of the infrastructural system, and these representations result from the efforts of different stakeholders to make the system legible from their own perspectives and interests.
Many design choices shape the legibility of the waste system, starting with its physical interfaces. New York City’s waste bins come in many different shapes and colors with no single visual language to designate different material streams. In the streets, one can find separate bins for beverage containers, newspapers, and residual waste. Bins in the subway, however, accept all waste materials. Such inconsistencies send mixed messages even if governments want to encourage participation in recycling programs.
Design decisions also determine how discarded materials are categorized as waste or recyclables. These decisions shape collection routes and maintenance protocols that are explicated in service contracts and performance indicators. They determine how data are collected, which data sets are accessible to the public, how they are represented, and how one can gain access to them. Infrastructure legibility is not exclusively a concern of service providers, however. Citizens, advocates, and activists also attempt to make systems legible and to use system representations for their own purposes.
Practices of making systems legible may include textual, visual, and performative means. A substantial part of such practices is connected to questions of accountability. Public restrooms often include a written journal of cleaning times, thus making the maintenance process legible and holding the cleaning company accountable (Zinnbauer 2012). A constantly overflowing waste bin indicates the need for a response that either reduces waste generation or increases collection intervals.
Such disturbances can be an effective form of civic protest. When residents of a marginalized neighborhood in the Mexican city of Oaxaca blocked access to the adjacent problem-ridden landfill, they caused trash to pile up in the streets (Moore 2008). The performative display of their protest and its consequences underscores that public policy has an underappreciated aesthetic dimension. The waste system may be defined by organizational and regulatory structures, but it is their material and tangible consequences that determine how the system is perceived.
When we hear the term “design,” we might first think of the process of giving shape to a physical object, such as the iconic form of Eero Saarinen’s TWA Flight Center at the John F. Kennedy International Airport in New York. The architect of an airport has to organize diverse functions in space and make them legible to travelers and employees. Physical design regulates how people check in, pass through security, and are monitored throughout the process. In this narrow sense, design is about organizing functions into a consistent, ideally delightful shape that “speaks” to its potential users. In other words, the design of an airport regulates user behavior, either subtly or, in the case of holding cells and other hidden security facilities, more forcefully.
Complex structures such as airports, however, often require additional measures to make sure that travelers do not get lost. Information design is concerned with organizing information, making it available where needed in a visual language that is both accessible and understandable. In the view of information architect Richard Saul Wurman, successful design facilitates understanding (Wurman 2000, 94).
Travelers increasingly check in using their smartphones, which have become part of the socio-technical system that regulates the physical space of the airport as much as its architectural features do (Dodge and Kitchin 2004). The coordination of the travelers’ actions in physical and informational space is the domain of interaction design, which is concerned with how users engage with each other through a system. Websites and apps for booking and checking into flights require access to a technical layer of protocols, such as the proprietary AviNet Data Network Service or the federal Advanced Passenger Information System, which again are subject to design decisions and have to be coordinated between private and public stakeholders.
These protocols are not autonomous; they require the human action of traffic controllers or call-center operators. How customer representatives interact with travelers is a facet of service design, a discipline concerned with how services involving complex systems of people, interfaces, and technologies are organized and experienced. This practice includes specialized language and categorizations, such as the terminology the aviation industry uses to describe irate customers (Bowker and Star 1999, 37). As the conversation designers Hugh Dubberly and Paul Pangaro argue: “An organization is its language. Narrowing language increases efficiency. Narrowing language also increases ignorance. To regenerate, an organization creates a new language” (Shamiyeh 2014, 363).
Physical shape, the organization of information, people’s behaviors, standards and protocols, languages and categorizations. All of these aspects are important for how we experience a trip. If one aspect is neglected, the whole system might fail. Applied to complex socio-technical systems, the definition of “design” is necessarily broad and inclusive. Due to their scope, socio-technical systems can no longer have a single designer and the concept of a single type of “user” is no longer applicable. Design concerns a manifold communication process between numerous actors, which cybernetician Gordon Pask described as a conversation (Pask 1976).
The diverse number of actors raises questions of power, equity, and authority. Participatory design and co-design describe approaches that include users in the design process. Participatory design emerged from the political context of Scandinavian trade unions that demanded a voice in the computerization of workplaces and factories (Kensing and Blomberg 1998). The purpose of participatory design practices is not limited to practical outcomes, however. It also works as a research method for learning about a specific environment through reflection on the participatory design process.
Applied to waste systems, a comprehensive design approach connects systems of production with systems of disposal. This is the intention of extended producer responsibility (EPR) policies, which incentivize manufacturers to consider recyclability in the design of a product by making them pay for its recovery. As media scholar Jennifer Gabrys describes it in the context of electronic waste, design should consider the whole “career plan” of an object from production, use and reuse, to final dismantling (Gabrys 2013, 152). The technical aspects are probably easiest to address in such a scenario. The human aspects, including occupational health concerns and questions of dignity, are more complex and potentially raise more controversies.
The overarching role of design in socio-technical systems does not imply that everything always happens according to plan. Infrastructures are also spaces of improvisation, temporary solutions, ad-hoc repairs. In this context, the designers are no longer outside the system, and their involvement is not over once a system assumes operation. In fact, many systems take shape as they are used, and design issues frequently arise from their use (Norman and Stappers 2015).
Urban planning and development literature concerned with infrastructure often still ignores the close entanglement of people and artifacts in urban services and conceptualizes the city exclusively as a product of abstract social, political, and economic forces. There are, however, several schools of thought within sociology and the studies of science and technology that embrace the world of material artifacts and their active role in urban systems. These literatures include actor-oriented and interface-oriented approaches, which avoid abstractions of social forces and instead investigate individual interactions at the boundaries between groups or systems (Long 1989; Lewis 1993). Langdon Winner’s work investigates how technology embodies power relationships and influences social arrangements (Winner 1980). Social construction of technology (SCOT) literature investigates the development of technologies and how they are shaped by social interests (Bijker et al. 1987). Actor Network Theory (ANT) literature avoids any categorical distinction between people and artifacts in terms of their capacity to affect things in the world (Callon and Latour 1981).
To illustrate the interchangeability between human and nonhuman actors, Latour compares three means of traffic control: a sign with a speed limit, a policeman at the curb, or a speed bump in the middle of the road. In his view, all three are equivalent means toward the same goal, regardless of whether they involve humans or objects to achieve this goal (Latour 1994).
In traditional planning theory, this might seem to be a provocative position, but it deeply resonates with the diverse practices of design, which are all concerned with how material artifacts shape their environment, how they affect behavior and communication, and how they are affected and changed in their use. In short, with the differences that material artifacts make in the world.
All design practices share a concern for what should be hidden and what should be exposed. A prominent principle of a functionalist understanding of design is to make the system as unobtrusive as possible (Buchanan 1985). Computer scientist Mark Weiser asserted that computers had to become invisible to become ubiquitous. Comparing computation to the cultural technique of writing, he observes: “The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it” (Weiser 1991). Or as Bowker and Star note, “Good, usable systems disappear almost by definition. The easier they are to use, the harder they are to see” (1999, 33).
One strategy to “make a system disappear” is to hide its technical complexities and make its surface as seamless and consistent as possible. By minimizing the number of inputs and moving parts, a system can become more reliable and accessible. Such a paradigm of seamless design has drawbacks, however. Without the option for intervention in a simple interface, the user may not feel in control, especially if something does not go as expected. A seamless system can be difficult to adapt, upgrade, repair, or even diagnose in the case of a breakdown.
Instead of aiming for a homogeneous and monolithic system, seamful design emphasizes the seams (Chalmers and Galani 2004)—similar to ceramics repaired in the Japanese Kintsugi technique, which emphasizes the seams between the broken pieces. By deliberately exposing certain technical aspects, a system can become more adaptable and extensible, providing visible clues about its internal state. Provocatively, this could mean designing a system that is easily “hacked” through appropriation, study, and improvement by the user, accepting the possibility that parts of the system might break in the process (Galloway et al. 2004).
Given the prominent concern with invisibility, it seems contradictory that an equally important goal of design is to make objects and systems “talk” by exposing their functions to the user. In reference to this paradox, industrial designer Dieter Rams explains that design becomes unobtrusive by being informative about its functions rather than by evoking emotional response (Rams 1984). In Heidegger’s terminology, designers who subscribe to the functionalist paradigm exemplified by Rams and others are concerned with producing artifacts that are “ready-to-hand,” that exist in relation to other things, techniques, and actions, rather than “present-at-hand,” a solitary state that becomes manifest when a tool becomes unusable (Heidegger 1927, 73).
How can this desirable state be achieved? Since the 1980s, the cognitive aspects of design have been inspired by J. J. Gibson’s ecological approach to visual perception. Gibson’s affordances describe the visible properties of the physical environment in relation to what they offer an animal, such as shelter in the shape of a cave (J. J. Gibson 1979). In Donald Norman’s reading, a user constructs a system image of an object and its functions by interpreting its affordances. More than simple signs based on convention, affordances are possibilities for action. A metal bucket, for example, affords turning it over and stepping on it, making noise with it, or filling it with water. To learn how to use a system may involve clicking buttons to see what happens. Designers can therefore communicate by creating affordances that signify a system’s state and its possible actions (Norman 2002, 188).
This is not to say that symbolic information is not important. Other than with simple acts like using a bucket, things rarely speak for themselves. “Intuitive” interfaces are more accurately called “clear, if previously understood,” since they depend on assumptions of shared knowledge and experience. Because the heterogeneous infrastructures considered in this book encompass organizational and informational aspects as well as physical components, design in these contexts involves making a system legible by considering possibilities for action, questions of symbolic encoding, the organization of information, and associated human practices.
Based on the concept of infrastructure legibility, this book investigates existing and emerging methods for reading waste systems and examines how they influence the governance of these systems. The three parts of this book look at the formal structure, informal practices, and the interactions between individuals and service providers. They address a common problem of how to gain information about what happens “inside” waste systems given the scarcity of data and a lack of incentives to collect and share them. The case studies represent three ways to address this question:
In the context of urban data initiatives, the opacity of the waste system serves as a counter-example to the widespread assumption that the world is drowning in data. The three cases are therefore not arbitrarily chosen. They act as exemplars for the three most important directions in current urban data research initiatives, which include the Internet of Things approaches that use networked sensors, context-sensitive data strategies tailored to a particular community and situation, and urban analytics methods that mine existing data repositories for information about the state of urban systems. Each of the three methods represents a different way of reading urban infrastructures, and each comes with its problems and limitations, addressed in the following three parts.
I undertook these studies as part of a team of researchers from the MIT Senseable City Lab directed by the architect Carlo Ratti. All three studies make use of location-based technologies such as GPS to map the geographies of waste transportation, the organizational arrangements of service providers, and the interactions between citizens and local governments. Each study looks at waste infrastructure through a different lens, including the external perspective following items through a waste system, the internal perspective of workers within a waste system, and a boundary perspective of citizens crossing into the system to engage with service providers.
Part I investigates how waste infrastructures are represented and made legible through existing monitoring systems, and how these representations have shaped the public discourse around waste systems. It compares two foundational conceptions of urban legibility proposed by Kevin Lynch (1960, 1984) and James C. Scott (1999). Although remarkably different, these conceptions provide complementary perspectives for understanding infrastructure legibility, capturing physical-sensory as well as abstract-informational dimensions in the design and the politics of urban systems. Based on this foundation, I develop a framework for investigating how infrastructure systems such as waste collection and recycling are made legible.
The Trash Track case study that my colleagues from the Senseable City Lab at MIT and I carried out in Seattle examined the geographies and topologies of companies and facilities involved in the waste system (Boustani et al. 2011). With the help of active location sensors, we mapped how items discarded by a typical household moved through the waste and recycling system, gathering data that is difficult to obtain using traditional methods. The approach prototyped in this study allows us to estimate the extent to which transporting waste diminishes the benefits of recycling, to support the investigation of environmental crimes, and to offer ways of monitoring a waste system from “the outside.”
Part II focuses on the challenges that developing economies face when moving from informal to formal systems for waste management. The tools of legibility in the process of formalization involve legal, technical, and social arrangements, many of which are driven by a need for accountability in assessing service contracts, policy options, environmental pollution, and public communications.
The Forager study described in part II looked at the spatial strategies, social organization, and tacit knowledge of waste pickers working for Brazilian recycling cooperatives in São Paulo and Recife. It investigated the informal processes these workers used to organize their collection routes and processes, exchange data, and interact with governments, companies, and the broader public. These cooperatives operate mostly on tacit knowledge, and the study examined ways to improve their data collection methods to benefit their dealings with corporations and municipalities without petrifying their dynamic, flexible organizations.
The result of this inquiry, however, is not necessarily an optimistic story of empowerment. Instead it highlights the limitations and dangers of “formalization by measurement” in which location-based technologies expose fragile, informal systems to more powerful actors.
Part III looks at how infrastructure monitoring and participatory infrastructure governance are enacted through civic technologies such as citizen feedback systems. Urban infrastructures in many cities of the United States have become more visible and more participatory, for better or worse. Due to limited budgets, cities struggle to maintain urban infrastructure, and due to privatization of service provision, individuals are exposed to a more fragmented and unequal infrastructural landscape (Steve Graham and Marvin 2001). Various solutions are advertised by technologists. We are told that live data from sensor networks can make infrastructure more efficient, that performance metrics can make governments more effective, and that digitally mediated participation can make cities more civic and just.
To examine the role of participatory sensing in infrastructure governance, my third case study analyzes the design and use of mobile systems that allow citizens of Boston to report infrastructure failures such as potholes, broken streetlights, or garbage spills. It looks at how design of the various interfaces between citizens and government shapes their interaction, as well as how a service provider’s self-representation affects its legibility. This part raises questions about the role of mediating technologies in governance models that are based on interactions between governments and constituents. It makes a case for scrutinizing the political nature of interfaces and argues that standards and common protocols are critical components of democratic discourse.
The conclusion synthesizes the forms of legibility described in the book, extending the notion of infrastructure legibility beyond waste systems to address concerns about accountability, anonymity, and the role of the user in distributed systems. The conclusion makes a case for accountability-oriented principles that acknowledge the political responsibilities of design. Because interactions between individuals and system providers are shaped by interfaces, the designer often unknowingly assumes the role of regulator and moderator. I finish the book with a proposition for design principles that calls attention to hidden issues of governance and that situates the design and implementation process of interfaces, protocols, and platforms as a form of democratic discourse.