258 A HISTORY OF AMERICA IN 100 MAPS
AFTERWORD: THE ROAD AHEAD
If this volume seems eclectic, it is so by design. To showcase
the wide influence of maps in American history, I have selected
examples that run the gamut. Across five centuries, maps
drove statecraft and diplomacy, exploration and imperialism.
They shaped settlement patterns, political strategy, and moral
reform. They galvanized social movements and stimulated
patriotism, asserted territory and investigated disease. They
advertised destinations and products, explained scientific
theories, and recorded personal histories. Occasionally they
made people laugh.
Yet however diverse their motives—not to mention their
circulation and appearance—maps have almost always been
physical artifacts that could be touched, used, updated, and
filed away. That generalization no longer holds true. Since
the turn of the century, maps have been more likely to be
generated by software and to live online; some are never
printed at all, much less archived. Geographic information
systems (GIS) and other platforms have also democratized
mapmaking, and we live with those results on a daily
basis. The news and mass culture are littered with graphic
information, and even animated and interactive maps now
seem par for the course. We have fluid maps of Internet traffic
and Twitter feeds, wind patterns, demographic shifts, and
even consumer preferences.
This avalanche of digital maps can make it difficult to
separate the signal from the noise. But one innovation on the
horizon promises to change not just the way we live, but the
very nature of mapping: self-driving cars. The implications of
this technology are startling. In the short term, autonomous
vehicles will transform commuting and bring mobility to the
elderly, the disabled, and the young. Even more consequential
are the long-term structural consequences for an economy
that has been inextricably linked to the automobile for nearly
a century. Auto repair, trucking, and insurance will all be
fundamentally affected.
Equally radical are the implications for maps, since self-
driving cars cannot operate without them. More precisely,
autonomous vehicles require elaborate, detailed, three-
dimensional systems of measurement to shepherd them
through space. This gargantuan challenge promptedthe recent founding of DeepMap in Silicon Valley, where
engineers are now working to support autonomous vehicles
on a wide scale. The foundation of this platform is the
Global Positioning System (GPS), a technology now
commonly used on cellular phones and dashboard
navigation systems. DeepMap integrates this GPS base
data with dynamic information incorporated in real time
through vehicles outfitted with high-definition cameras and
a range of sensors. This equipment reads and monitors both
the fixed and the fluid environment on the road, enabling
the vehicle to assemble multiple data points and to form
a “memory” of its surroundings. By comprehensively
recording the immediate environment, the vehicle is
safely guided through space.
But that is just the beginning, for every piece of
information read by those instruments on that individual
vehicle is then shared with a larger platform. This makes
the car itself a mapmaking tool, one that constantly gathers
information to improve the larger system. In this way, we
are taken back to the era of discovery and reconnaissance:
just as field surveyors took measurements that were then
translated into cartographic form, self-driving cars will
generate situational data that constantly advance the general
geographical system of any given locality. In a rather ingenious
loop, the vehicles gather information, use that information to
move through space, and in so doing generate new data.
As map scholar David Rumsey observed, the result is nearly
an organic entity, a constant exchange of information in real
time that is designed to achieve maximum accuracy.
Of course what guides the vehicle safely and successfully is
not a visual “map,” but rather a digital system of instructions,
the data themselves. Which prompts the question: Are “maps”
for self-driving cars really maps at all? DeepMap’s chief
operating officer Wei Luo explains that even for autonomous
vehicles, there is still a need to visualize the data. Some of
these visualizations are generated to show passengers what
the car “sees” and “knows” at any given moment; others
are designed for engineers developing the technology; and
still others are needed for quality control. In other words,
the visual maps are designed for human consumption. Luo
rendered this visual picture of the data to convey how the
system “teaches” the rules of the road to the vehicle, with
green lines guiding it through an intersection.
In this regard, autonomous vehicles are not so different
from humans: they need to sense their surroundings inDeepMap, data visualization for
autonomous vehicles, 2018