How cars became mobile spatial media: A geographical political economy of on-board navigation

Introduction

The claim to study cars as mobile spatial media has two core components. First, an exploration of how cars became mobile media. Through their ongoing computerization, combined with the successive incorporation of various technologies of information and communication, cars have gradually enhanced their mobility with capacities to reproduce audiovisual information, and connect into multimedia communication networks that include radio, TV, and Internet broadcasters; car manufacturers; other vehicles; and users of a wide range of devices (such as mobile phones). This transformation is so pervasive that most cars manufactured today are equipped with some combination of technologies such as radio, Bluetooth, in-built phone, digital screens, and Internet-enabled computers, among others. The result of this has been the construction of mobile media environments that provide car drivers and passengers with a vast array of possibilities of consumption, interaction, and experience.

Second, while many of the aforementioned technologies enrich the experience for drivers or passengers through the creation of an in-vehicle environment for communication and multimedia entertainment, there is another set of technologies that enhance the car’s most essential function: navigation through physical space. Since the early days of the automobile age at the turn of the 20th century, early developments in navigation technologies promised to assist drivers in finding their way (French, 2006). However, it has been over the course of the past three decades that, by means of innovative combinations of GPS receivers, mapping interfaces, and even dead reckoning, cars have incorporated increasingly sophisticated positioning and navigation functions to become full-fledged spatial media. The mobile and spatial developments converging to transform cars into mobile spatial media have separate, but interrelated paths. The former, responding to demands for on-board media and communication satisfied by technologies developed at the nexus of the entertainment, electronics, and telecommunications industries (radios, car stereos, Bluetooth connections, etc.). The latter, stemming from interlocking military, scientific, and automotive industry developments in navigation, and aided by advancements in the mapping and geospatial industries (navigation interfaces, GPS integration, parking assistance).

The two paths laid out have intersected in significant ways to shape on-board navigation technologies and practices, which I classify in three broad stages. Early innovations like the Jones Live-Map in 1910 (French, 2006), which displayed routes by means of rotating paper discs, and the scrolling dashboard map Iter Avto in 1930 (Rigg, 2018) characterized recurrent attempts to develop automotive navigation throughout the first half of the 20th century. However, in spite of intermittent advances throughout the following decades, navigation as a consumer technology did not achieve a critical mass for several decades. It is only in the 1980s that a first stage of on-board navigation can be said to start in earnest with more advanced (and commercially successful) developments such as Honda’s inertial navigation system Electro Gyro-Cator in 1981 (Arai, Nakamura, & Shirakawa, 2015) and the appearance of Etak in 1985 (French, 2006), the first digital auto navigation system. In this stage, automakers and device manufacturers began to seriously explore options to market car navigation at a massive scale.

A second stage, beginning in the mid 2000s, corresponds to the widespread use of stand-alone navigation devices made by manufacturers such as Garmin and TomTom, along with the rise of GPS-equipped smartphones, combined with mapping applications (such as Google Maps, Apple Maps, and Waze), as tools for navigation. The lower cost, greater accessibility, and proliferation of mapping applications have led to the massive adoption of smartphones as navigation devices. Consequently, these have to a large degree substituted the use of both factory-installed navigation systems and stand-alone navigation devices.

The third stage of on-board navigation has not begun in earnest, but some of its contours are presently coming into view. This corresponds to the enhancement and reimagining of inbuilt car navigation as a key component in the reimagining of cars themselves—chiefly through the development of next-generation mapping and navigation interfaces that power self-driving cars. The use of various combinations of LIDAR (light detection and ranging) and artificial intelligence-enabled HD and 3D maps is reconfiguring the mobile spatial media environment within (and around) cars, while simultaneously laying the foundations for an entirely new way for cars to navigate—and ultimately produce—space.

These three stages can be understood, from a political economic perspective, through a process of competition and reconfiguration between various industrial sectors, and the specific arrangements that become dominant during each period. In the first stage, it was the automobile industry that led the development of navigation, while the second stage was defined by the rise of third-party firms, which included stand-alone navigation devices, and hardware and software companies involved in mobile phone navigation. The contours of the nascent third stage can be read as a synthesis of the previous competitive dynamics, where both automakers and third-party firms (navigation device makers, mobile phone manufacturers, and software companies) are now converging, in new configurations, towards the reincorporation of navigation into car manufacturing for the development of autonomous driving.

On-board navigation as mobile spatial media

As a means of transportation transformed by incremental computerization and media integration over the past decades, automobiles have come to embody the convergence of mobility and media. Their long-standing symbiotic relationship with media has been examined with attention to their significant impact on a range of aspects of contemporary life, such as consumption practices (Freund & Martin, 2008); commercialization of the landscape (Cronin, 2008; Gudis, 2004); experiences of the physical and social environment (Appleyard, Lynch, & Myer, 1964; Thrift, 2004); the stratification, individualization, and commodification of civil society and public space (Sheller & Urry, 2000); the production and consumption of various forms of popular media and national culture (Burnham, 1961; Flink, 1975, 1988; Osteen, 2008; Wollen & Kerr, 2002); the articulation of new sociotechnical configurations (Dant, 2004; Urry, 2004); and their highly personalized interpretation as symbolic objects with particular social and cultural encounters (DeLyser & Greenstein, 2015). Such is the power of these encounters that, as Urry convincingly argued, cars are less a means to travel “from A to B,” and more something we inhabit in multiple ways, shaped by the paradox of their “simultaneously flexible and coercive form” (Urry, 2006, p. 29).

Yet, for how much scrutiny automobiles have received, only recently have they become objects of analysis through the lens of mobile media (Goggin, 2012; Goggin, Dwyer, & Martin, 2014; Goggin, Dwyer, Martin, & Hutchinson, 2013; Hildebrand & Sheller, 2018; Pink, Fors, & Glöss, 2018; Pink, Gomes, et al., 2018; Whaiduzzaman, Sookhak, Gani, & Buyya, 2014). In this emerging conversation, a salient theme is how the interconnections between cars and new media, such as mobile Internet or smartphones, decenter each of these objects to reconfigure new and messier networks, media flows, and practices (Goggin, 2012, p. 318). A key vector of these interconnections is automobile navigation—or on-board navigation—which integrates communication, computation, geolocation, and visualization technologies. On-board navigation both connects the car to these “messier” networks and actively coordinates the vehicle’s position and movement within them. For these reasons, here I argue that on-board navigation is a productive site to examine cars through the lens of mobile as well as spatial media.

What are the implications of studying cars as mobile spatial media? A useful point of departure is Farman’s assertion that mobile media (from phones to subway cards) are constantly creating hybrid digital and material landscapes that transform how we “conceive of embodied space” (Farman, 2012, p. 15). In this case, understanding on-board car navigation as mobile media helps elucidate how it has contributed to the ongoing digital, material, and experiential transformation of the cars: from the practices of drivers and passengers to the space within the automobile, as well as its relation to the spaces it traverses and occupies. Navigation, however, cannot be meaningfully understood without attention to its incorporation of locative data.

The locational dimension of data qualitatively changes the composition and function of devices and communication networks. As de Souza e Silva and Frith point out, one of the main features of locative mobile social media networks is that they “enable the mapping of online social networks onto physical space” (2010, p. 486), and that this in turn affects our understanding of public spaces, while influencing “how we connect to other people in these spaces” (2010, p. 487). One of the key implications of this is that location awareness alters the character of networks to emphasize paths (movement across physical space) over nodes (individual users; de Souza e Silva & Frith, 2010, p. 487). In other words, in these types of networks, “the totality of the immediate space matters, not just the final destination” (de Souza e Silva & Frith, 2010, p. 491).

Cars are participating in this spatial shift as they incorporate new modes of on-board navigation, becoming part of locative mobile media networks that redefine the connections between nodes and space itself. However, locating points in space is only one aspect of navigation—which is also shaped by how those spaces are (re)produced, represented, experienced, and commodified. In order to assess the full range of implications of on-board navigation, it is productive to understand this technology within the broader context of spatial media, since this encapsulates technological components, spatial content, and emergent sociospatial practices that mediate and condition daily life to produce new spatialities and mobilities (Kitchin, Lauriault, & Wilson, 2017, p. 5). This, in turn, allows us to connect the developments both leading to and stemming from on-board navigation with the larger social, political, and economic dynamics of digital (geospatial) capitalism (Alvarez León, 2016; Elwood & Leszczynski, 2012; Leszczynski, 2012; H. Smith, 2014). In the sections that follow, I leverage the perspectives of spatial media, locative social networks (de Souza e Silva & Frith, 2010), and mobile interface theory (Farman, 2012) to trace three key stages in how navigation has come to reconfigure spaces (of mobility, information collection, product development, consumption, etc.) both inside automobiles and throughout their larger social, physical, and informational contexts.

Building a navigation ecosystem: From TravTek to Guidestar

In the early 1990s, advances in computer processors, digital mapping, communication, and navigation led to the rollout of test projects to assess the viability of large-scale integration between advanced traffic management systems (ATMS) and advanced traveler information systems (ATIS; Booz Allen & Hamilton & Highway & Vehicle Technology Group, 1998). Projects like FAST-TRAC in Oakland County, Michigan, and Pathfinder, along the Santa Monica Freeway Corridor (I-10) in California, were emblematic of this ambitious effort (Auer, Feese, & Lockwood, 2016, p. 19), and built on the potential shown by earlier designs—such as the U.S. Bureau of Public Roads Electronic Route Guidance System (ERGS), which, “although technically sound . . . required expensive roadside infrastructure and . . . was terminated by congressional mandate in 1970” (French, 2006, p. 272). Among the new wave of projects of the early 1990s, TravTek stands out as both a demonstration and a “grand experiment” designed to “provide objective measures of the benefits [from] an integrated advanced driver information system and infrastructure” (Rillings & Lewis, 1991, p. 732). This would in turn determine the feasibility of this type of navigation ecosystem for nationwide implementation (Mateja, 1992).

TravTek was a joint public–private effort between General Motors (as project manager and systems engineer), American Automobile Association, AVIS (a car rental company), and Motorola on the private side, and the Federal Highway Administration, Florida Department of Transportation, and the City of Orlando, as public backers. Building on the pioneering work from projects such as “Ali-Scout in Germany, AMTICS in Japan, and Pathfinder in the United States,” TravTek was the “largest, most comprehensive advanced driver information system project” of its time (Rillings & Lewis, 1991, p. 729).

This experimental project was implemented from November 1991 to June 1994 in the city of Orlando, Florida (Booz Allen & Hamilton & Highway & Vehicle Technology Group, 1998, p. 8). The on-board navigation interface was installed and deployed in a 100-vehicle fleet of 1992 Oldsmobile Toronados, 75 of which were assigned to visitors to Orlando, and 25 to local residents (Rillings & Lewis, 1991, p. 729). These vehicles were equipped with a touch screen, which displayed a menu of options and locations as well as a dynamic electronic map, which followed the car navigation and provided directions. Far ahead of its time, TravTek featured voice directions; real-time updates reflecting accidents, weather, or traffic jams (with alternative route suggestions); and a 24-hour helpline that could be accessed via the car-mounted telephone. TravTek operated through a combination of magnetic compass, wheel-mounted sensors, FM radio, and satellite navigation; all connected to a database that integrated maps, locations, and events (“How In-Dash Navigation,” 2012; Mateja, 1992; Rillings & Lewis, 1991).

Since it was a test as well as a product, TravTek logged “interactions between the driver and the system as well as the information available to the driver [including] key presses, screens displayed, congestion information, routes recommended and routes followed” (Rillings & Lewis, 1991, p. 730). In the case of TravTek, this information and implementation cycle was carried out between the three segments within its public–private partnership, which were linked by radio and telephone. These segments are described in Table 1.

OPEN IN VIEWER

Table 1.

Segments in the TravTek public–private partnership.

Segment	Primary actors	Responsibility	Tasks
1	American Automobile Association (AAA)	TravTek Information and Service Center (TISC)	Supply maps and information databases for TMC and TravTek vehicles. Provide live supplemental information relating to travel, emergency services, and technical assistance.
2	Public-sector partners (Federal Highway Administration, Florida DoT, City of Orlando)	Traffic Management Center (TMC)	Collect, analyze, and transmit by-the-minute traffic, weather, and event information to TravTek cars.
3	General Motors	TravTek vehicles	Process and display navigation and routing data using two in-built computers and dashboard screen. Communicate directions to driver via computer-synthesized voice.

Note. Source: Assembled by author with information from Rillings and Lewis (1991).

The technological integration and the organizational coordination behind the TravTek pilot paid off. According to the U.S. Department of Transportation’s commissioned assessment of TravTek, the results from this program projected a range of benefits for vehicles using this navigation technology: 37% reduction in wrong turns, 32% reduction in vehicle stops, 5% reduction in travel distance, 11% reduction in fuel consumption, and 6% reduction in emissions. Additionally, participants reported using the system between 40% and 80% of their trips, and evaluated all functions, except for traffic information, as useful (Booz Allen & Hamilton & Highway & Vehicle Technology Group, 1998, pp. 10–12).

Yet, in spite of the potentially transformative nature of these benefits, TravTek was not necessarily scalable, since its operation required both top-of-the-line technology and precise coordination between partners both in the public and private sectors. In spite of this, while it did not go on to become a standard product, TravTek did encourage General Motors to move forward with the development of what would be the first preinstalled on-board navigation system in the U.S. market: the Guidestar. ¹ As a pilot project, General Motors had allowed TravTek’s design and installation in one specific car, the 1992 Oldsmobile Toronado. However, after this car was discontinued, GM persisted in pursuing car navigation by building on the advances achieved through TravTek. In an effort to maintain this leadership position, GM was first to market by partnering with Japanese components manufacturer Zexel and adapting that company’s Navmate navigation system for factory installation into the 1994 Oldsmobile Eighty Eight model (Borcherts, Oshizawa, & Fujii, 1995). The resulting system, now branded Guidestar would later be made available on Oldsmobile’s LSS and Bravada models (Automotive News, 1996).

The early leadership achieved by GM in the navigation market did not have the expected impact. Guidestar’s first-mover advantage in the market was undercut by the confluence of three main factors: (a) high price—US$2,000 was too high for the average consumer, while car rental company AVIS bought almost half of the total units ever sold; (b) limited rollout—installed in only three Oldsmobile models; and (c) narrow geographic range—it only provided coverage of 17 U.S. states and the District of Columbia (Autos of Interest, n.d.; Mateja, 1995). In addition to this, two innovative features of the TravTek pilot were lacking from Guidestar: real-time traffic updates and automatic driving instructions. As a result, this pioneering system only sold about 2,000 units in its first 2 years in the market (Automotive News, 1996).

While Guidestar did not achieve commercial success, it did raise consumer expectations and nudged the U.S. car navigation market forward in the late 1990s and 2000s. Instrumental to this was the Clinton administration’s order to end the scrambling of GPS signals on May 1st, 2000, which were previously reserved for military use in their full capability, leading to an immediate tenfold increase in the locational accuracy (Reuters, 2000). Consequently, the navigation market experienced substantial growth in sales of stand-alone GPS navigation devices, and the growth of brands like Garmin, Magellan, and TomTom in the 2000s.

As the navigation market grew, it was on the verge of transformation via rapid advances in mobile phone tracking. Spurred by the Wireless Communications and Public Safety Act of 1999 and the Enhanced 911 Ruling of the Federal Communications Commission (Wireless Radio Services; Compatibility with Enhanced 911 Emergency Calling Systems) the same year, mobile phone manufacturers began to incorporate GPS location functionality in mobile phones (van Diggelen, 2009; Yoshida, 1999). In the first decade of the 21st century, the expansion of GPS use and the development of smartphones would eventually converge and transform the automobile navigation market— and with it, the mobile spatial media environment of cars.

Navigating by smartphone

Along with the development of on-board navigation systems, the transformation of mobile phones into navigational tools is key to understand automobiles in their development as mobile spatial media. As the case of TravTek illustrated (and other field tests implemented to a certain degree), telephone communications have long been considered a useful way of providing drivers with assistance via link with a traffic center. This created a media ecosystem where the mapping and navigation were undertaken by a purpose-built device, while the telephone acted as a support feature in an integrative environment for driver assistance. Yet, in spite of much research and forethought into driving behavior, traffic management, and anticipatory problem solving (Krage, 1991; Taylor, 1991), the inclusion of the telephone into a fully fledged car navigation environment did not take hold. Although the small user pool from the TravTek pilot test had access to personalized attention through car phones, telephone assistance centers for car navigation proved unfeasible at large scale. Furthermore, very few cars in the 1990s or even 2000s were equipped with telephones, which dramatically curtailed the direct role of this communication technology in creating a sustained multimedia navigation ecosystem.

Throughout the first decade of the 21st century, the telephone began to shift from ancillary navigation technology—reserved for assistance and only available in rare circumstances—towards a more prominent role as navigation device in its own right that would come to substitute a growing share of on-board navigation interfaces. Instrumental to this shift was the launch of Google Maps in 2005, after Google made its second-ever acquisition by purchasing Sydney-based Where 2 Technologies, the company that started developing this mapping service (Hutcheon, 2015). With a dynamic interface and rapid loading time, Google Maps soon revolutionized the online mapping market, previously defined by more static products, such as MapQuest. This advantage was translated to the mobile market: Google Maps released its first full smartphone app by preloading it on the first commercial Android smartphone on September 2008 (Google, 2008). With this release, the rising trajectory of the smartphones as navigation tools began to intersect with the decline of the stand-alone navigation device (O’Brien, 2010).

In half a decade of existence, Google Maps became the most popular app for smartphones, with 54% of global smartphones reporting having used it in the past month, according to a survey conducted in 2013 (C. Smith, 2013). As Google Maps became the standard of mobile navigation apps, even Apple—one of Google’s main competitors—added it as the default mapping application for the iPhone iOS until the release of their own Apple Maps in 2012—which received a significant amount of backlash due to its many inaccuracies (Arthur, 2012). With this, Google swiftly came to corner the mobile mapping and navigation market.

Google Maps represents a component of the search engine company’s goal to “organize the world’s information” as it expands to include all manner of geospatial information. This ambition manifested in extensive spatial data collection projects through the Google Street View program (Vincent, 2007), along with the development of Google Earth and the comprehensive worldwide purchase and digitization of government agency maps (e.g., U.S. Geological Survey, U.S. Forest Service, U.S. National Parks Service) through their Base Map Partner Program that would make the basis for Google Maps (Fisher, 2013). Google has maintained its mapping leadership by incorporating new navigation features and buying out potential competitors. The search giant’s 2013 acquisition of Waze (an Israeli navigation application powered by real-time user data) for a price tag of over US$1B suggests the strategic importance of staying ahead in the “mapping wars” that are primed to escalate over the coming years, as high-definition and three-dimensional maps become key assets for the self-driving car revolution (Bergen, 2018; Fiegerman, 2017; Lunden, 2013).

This intensified competition for the mobile map market came hand in hand with sweeping user transition to smartphone use as the primary instrument for navigation. In light of the high cost of preinstalled car navigation systems, as well as purpose-built add-on devices, most drivers opted to find their way with smartphones. It did not help automakers that smartphone mapping applications provided a smoother interactive experience, a larger array of options, and user interfaces that were familiar for consumers used to mobile touch screens. The standardization of smartphone mapping applications for car navigation has forced automakers to reckon with the role and relevance of manufacturer-installed on-board navigation technologies, which they have spent the past two decades developing and marketing with varying degrees of success. Illustrative of the more radical reactions to this shift in the face of smartphone navigation was the 2017 decision by Chevrolet to remove the on-board navigation system from its Bolt model and substitute it with a media environment designed to enhance the smartphone’s functions through 4G Wi-Fi Internet connection and a 10-inch touch screen. Similarly, other automakers have opted for strategies of smartphone integration through the installation of Apple Play and Android Auto in the car media systems, allowing users to display and use smartphone apps on the dashboard screens (Colias, 2017).

While many cars continue to include on-board navigation systems, the compositional changes in the car navigation environment exemplified by the move to smartphone navigation point to a pendular motion in the development of cars as mobile spatial media. While for some time automakers tried to compete with smartphone navigation through “smarter” displays, it seems many are opting to capitulate and accommodate users’ preferences for mobile phones. However, this strategy comes with an expiration date, since a new battle over car navigation is already under way. In this coming stage, automakers who have invested in developing on-board systems are positioned to reclaim the mantle of navigation since this function is at the core of the next technological revolution in the automobile industry: autonomous driving. In this coming technological shift, the task of navigation and its associated technologies will be redefined in the context of autonomous driving through a reconfiguration of a mobile spatial car media environment—one that is simultaneously enabled by new technologies while drawing from long-standing ideas of integrated transportation, communication, and connectedness. The following section explores how the car’s mobile spatial media environment is changing through a process of intensifying interconnection led by new navigation technologies at the core of autonomous driving.

Interconnected media spaces

The mobile spatial media ecosystem that has developed around car navigation reflects the gradual computerization, transformation, and integration of cars, navigation devices, and mobile phones. The previous stages of car navigation have been defined by the competitive dynamics between the firms that build and commercialize these three types of technologies. The coming stage, on the other hand, is likely to be shaped less by competition over navigation as a discrete service, and more by how this technology is incorporated into the development of autonomous driving. The industrial landscape that resulted from the prior two stages of car navigation, along with the particular technological and competitive advantages distributed throughout this landscape, laid the groundwork for this technology’s incorporation into autonomous driving.

While presently automakers continue to grapple with the disadvantages of their own preinstalled navigation systems against smartphone applications, their position in overseeing the entire car manufacturing process may prove to be a source of competitiveness, since the “next-generation [navigation] systems are not just an important way for carmakers to interact with drivers. They are also a crucial step in the development of autonomous vehicles” (Taub, 2018). On the other hand, companies that have developed navigation technologies are either partnering with (or getting acquired by) automakers (like HERE, now owned by a German automaker consortium), or, if they have the necessary scale, building their own self-driving cars powered by their navigation software (like Google’s Waymo).

In terms of the media ecosystem created by car navigation technologies, when these are placed within the context of autonomous driving, they connect, in functionally interdependent ways, the mobile media spaces inside the vehicle with increasingly sophisticated, multimodal spatial representations of the world it navigates, and the networks and informational infrastructures that make both of these possible. This is because autonomous driving requires both an accurate and constantly updated model of the world—which is gathered through a collection of sources (LIDAR, cameras, GPS, high-definition maps, etc.)—and a way to effectively navigate through it (through computers that coordinate the vehicle’s movement in space). This break from previous stages in the development of car navigation is intensified by how the media ecosystem created by this technology is increasingly enmeshed in a much broader range of potentially far-reaching sociotechnical infrastructures and widespread downstream impacts that include—but are not limited to—digital information networks, intelligent infrastructure, changing patterns of mobility, personal data collection, surveillance systems, targeted advertising, cyber security, and emerging regulatory frameworks.

Previous stages of navigation have not only created the conditions for this integration to take place but, in doing so, have also provided the blueprints for interconnecting the media spaces inside the car and those outside of it. One such example is that of intelligent traffic systems, which were first envisioned in the late 1960s and 1970s (French, 2006), and heavily influenced comprehensive navigation/traffic management projects such as TravTek in the 1990s and beyond (Booz Allen & Hamilton & Highway & Vehicle Technology Group, 1998). However, these early experiments lacked the technological wherewithal to enact and sustain their systemic vision of navigation-through-interconnection at scale. By contrast, the technological and institutional factors to make this vision a reality are coming in concert for the next stage of navigation. Current capabilities for computation and communication already make it possible to connect and coordinate cars with other cars, roads, and (physical as well as informational) infrastructure. This positions automobiles as mobile spatial media nodes within wider information networks that concurrently reframe how travelers experience movement through space, while also redefining, in terms of locational networks, the spaces through which they move.

To get a sense of how the interconnected media spaces of autonomous driving might be assembled in this next stage, it is useful to examine a specific self-driving car navigation interface, such as HERE Technology’s HD Live Map. HERE Technologies has its origins in Chicago-based Navteq, a leader in automotive mapping and navigation technology for three decades. The company was given its present name in 2011 when it merged with Nokia, which later sold it to its present owners, a consortium of German automakers (including Audi, BMW, and Mercedes Benz). Today, the Amsterdam-based HERE is at the forefront of self-driving car navigation technologies. Anchoring these is the HD Live Map, advertised as a “cloud-based service,” and a “self-healing map” is “made for cars, by cars” (see https://go.engage.here.com/self-healing.html).

This map’s key feature is the fact that it is continuously maintained with shared sensor data from other vehicles, as well as through crowdsourced data from various networks. Such data are validated and added to the existing high-definition map layers, providing real-time detail on changing road and environment conditions. In combination with various on-board car sensors, which detect the car’s immediate surroundings, the HD Live Map is enriched with three “intelligent map layers,” which source data from “industrial capture vehicles, crowdsourced vehicle sensor data . . . satellite imagery, aerial imagery, government data, and mobile probes” (HERE Technologies, 2017, p. 6). These layers are: (a) Road Model, which offers “global coverage for vehicles to understand local insights beyond the range of its onboard sensors such as high-occupancy vehicle lanes or country-specific road classification”; (b) HD Lane Model, which “provides more precise lane-level detail, such as lane direction of travel, lane type, lane boundary, and lane marking types”; and (c) HD Localization Model, which “helps the vehicle localize itself in the world by using roadside objects like guard rails, walls, signs, and pole like objects” (HERE Technologies, 2017, p. 4).

HERE’s HD Live Map integrates a number of technologies to build a model of the world that is intricately detailed and mutates closely tracking real time. This depends on maintaining continuous connections between each vehicle and an extensive network that is simultaneously informational, infrastructural, and institutional. This means that, to a much greater degree than in previous stages, the mobile spatial media ecosystem inside the vehicle is becoming functionally integrated into the communication networks that provide the media and data streams necessary for such ecosystem to exist. In fact, the developmental trajectory of on-board navigation technologies can be read as both the creation of mobile spatial media spaces within cars and also as the incremental interconnection between those internal spaces and external spaces increasingly characterized by their digital information density. For example, a central element of a previous iteration of the automobile media space, the car radio, already required an external infrastructure of antennas, radio stations, and spectrum regulations in order to operate. However, these requirements have multiplied with the expansion of devices, and their accompanying infrastructural, political/regulatory, and technical interconnections, which converge to enable the new mobile spatial media spaces of cars. The reliance of such media spaces within cars upon external spaces of increased informational density can be illustrated through the concept of code/space, as discussed by Kitchin and Dodge:

Code/space occurs when software and the spatiality of everyday life become mutually constituted, that is, produced through one another. . . . [An] example is a supermarket checkout. All supermarkets and large stores rely on computerized cash registers to process purchases. If the computer or the information system behind it crashes, shoppers cannot purchase goods, and in a functional sense, the space effectively ceases to be a supermarket instead becoming a temporary warehouse until such time as the code becomes (re)activated. (Kitchin & Dodge, 2011, pp. 16–17)

In this case, when a self-driving car becomes part of the HERE’s “self-healing map,” its meaning and function are redefined by simultaneously acting as a valuable node in an information network, and by entirely depending on its connection to the network for its most essential navigation functions. This paradoxical construction of the networked mobile spatial media ecosystem has roots that reach back a few decades. Recall the TravTek pilot project, where all components (magnetic navigation, dashboard display, telephone help center, etc.) worked in coordination as part of an elaborate navigation ecosystem—which itself built on earlier projects such as the Electronic Route Guidance System (ERGS) from the 1960s, and even from communications systems prevalent in marine and aerial navigation. For their undeniable innovation, HERE’s HD Live Map and other autonomous navigation systems are replaying old questions of coordination and traffic management in ways that, through new technological mediations, produce new mobile spatial media environments and new ways of interacting with—and inhabiting—them.

This emerging mode of integration between the car and its environments suggests the qualitative transformation of the mobile spatial media spaces within cars and the increased digitization and interconnection of the spaces through which these vehicles navigate. While on the one hand this is a question of technological advancement, on the other, it is also the confluence of particular sets of policies, sociotechnical imaginaries, and logistical arrangements that come together in specific configurations—which themselves exhibit a great degree of contingency, since they could have turned out differently under slightly altered circumstances, and they are likely to have important variations across diverse geographies. The following section explores the political economic dimensions and ramifications of these integrated media environments for travelers and other actors shaping an increasingly interconnected, digitized, and automated economic landscape—as well as the spaces they traverse.

Discussion

While the automobile has undergone radical changes over the past century, its core proposition remains stable: outsourcing the means of mobility to a machine while retaining navigation and decision-making by human agents. Yet, this stable configuration is at the threshold of a fundamental transformation. The ubiquitous discussion surrounding the development of self-driving cars, and the massive amount of resources aimed at advancing this technology, point to a new scenario where humans permanently take on the role of passengers by outsourcing not only mobility, but also navigation and decision-making to an autonomous vehicle. Here I argue that the transition between these two scenarios is made possible by the development and reconfiguration of cars as mobile spatial media, and their progressive integration with media spaces outside of the vehicle, with on-board navigation at the center of this process.

Advances in navigation technologies actively turn the car into a mobile spatial media environment because they facilitate the creation of new possibilities for inhabiting and experiencing the vehicle. Simultaneously, by relying on continuous multimodal information exchanges to locate and move the car through space, new navigation technologies also qualitatively transform the structure and function of vehicles in relation to broader environments by deepening their integration into networks of information, infrastructure, governance, and capital. For instance, new spaces and possibilities for consumption are catalyzed through the integration of mobile and spatial media, targeting the potential increases in available time and attention from car passengers.

Early signs of this development are already present in the appearance of advertised content and digital commerce in on-board navigation maps and dashboard interfaces, which turn the car into a space for data mining, commercial transactions, and context-rich advertising (Lindstrom, 2011; Markovich, 2018; Vaas, 2018). This new environment is vividly described in the press release documenting the partnership between Telenav, a leading navigation company, and Sionic Mobile, a provider of cloud-based commercial tools:

Telenav’s in-car hybrid-navigation solution will seamlessly integrate commerce capabilities within search, routing, mapping, and ETA [estimated time of arrival] sharing. This not only lets drivers more easily make purchases from the car, but also allows automotive OEMs [original equipment manufacturers] to optimize their monetization via in-car commerce, while still offering their brand-unique user experiences to enhance customer loyalty. (Telenav, 2018)

The emergence of new car mobile spatial media environments defined by continuous interconnection and increased commodification raises urgent questions on vehicular safety, driver/passenger agency, and liability—to name a few. Furthermore, the continuous surveillance capabilities of these environments merit amplified discussions about the continuous redefinition of privacy in the context of rapidly evolving geospatial data, media, and technologies (Elwood & Leszczynski, 2011; Leszczynski, 2017). As has become clear through the trajectories of social networks, mobile phones, and the Internet economy, the digitization and commodification of resources (including personal information) is not only an issue of privacy, but one that is entangled with material and economic inequalities—since the tools for extraction, appropriation, and monetization of information are unequally distributed across groups and individuals, as are the potential impacts of such information and its use (Eubanks, 2018; Noble, 2018).

While the exact configuration of the interconnected mobile spatial media environments of autonomous driving is yet to be defined, it is already clear that they will play a key role in articulating new spaces for consumption, data extraction, surveillance, and asymmetric exercises of power. This means that such environments should not be solely understood as technical or market-driven processes, but also through their political, institutional, and geographic dimensions. Such integrative analyses would highlight not only the interrelated sources of their development (firms, regulation, contextually specific ideas about privacy, agency, liability, etc.), but also would create opportunities for debate, deliberation, and the construction of alternative configurations that may leverage these technologies in ways that do not augment existing social inequalities and minimize the creation of new ones.

Conclusion

The long history of navigation devices, and their interaction with other vehicular media technologies within cars, illustrates the changing development of cars as mobile spatial media ecosystems. Yet, advances in navigation throughout the past two decades have decentered the car and other technologies, such as mobile phones, and rearranged them in new “messier” (Goggin, 2012) networks where many of their previous core functions are redefined. While once the phone was an ancillary navigation technology, it has come to dominate the on-board navigation market. While cars pioneered navigation as a manufacture-installed on-board feature, now this has largely been substituted by smartphone use. Yet, the development of automated driving is disrupting these previous arrangements and reincorporating navigation as a core technology that interconnects the mobile spatial media spaces within the car with those outside of it. This, in turn, is leading to cars becoming embedded into broader networks of information, technological systems, and institutional structures that bring with them new questions of privacy, agency, liability, security, and regulatory frameworks.

The emergence of autonomous driving, then, can be productively examined through the nodal role played by on-board navigation technologies. On the one hand, self-driving cars cannot exist without navigation technologies capable of delivering on the promise of a significant degree of vehicular autonomy. On the other hand, the existence of self-driving cars has the potential to update and redefine the very meaning of navigation as we have known it. In addition to the transformations of cars and the transformed experiences of passengers, the decisions and outcomes over “who navigates” a car have upstream and downstream effects on automobile design, testing, marketing, and government regulation (among others). A key factor for how these transitions take shape is the specific configuration of the interconnected media environment that will enable these changes, and which will create—following Urry’s (2006) prescient argument—an entirely new way of “inhabiting” the vehicle. Given the central role of the car in contemporary economy and society, the consequences of this shift are likely to impact consumers, public spaces, and sectors of industry ranging from information to services and manufacturing. In this article I have argued that examining cars and their ongoing transformation through the lens of mobile spatial media constitutes a productive window into understanding and assessing these impacts—an urgent task in the context of the potentially revolutionary shift brought about by autonomous vehicles.