10- getting there: the science of driving directions
In the fifteenth century, Henry the Navigator, a Portuguese prince, presided over a court in Sagres that became a center for cartographers, instrument-makers, and explorers, whose expeditions he sponsored. Seafarers returning to Sagres from the west coast of Africa reported their discoveries, and new maps were produced, extending the reaches of the known world. These maps became very valuable, owing to their utility in trade, war, and religious expansion, and were jealously guarded as state secrets.
Today's equivalent is a company called Navteq. It is the leading provider of geographic data to the Internet mapping sites and the personal-navigation industry—the boiler room of the where-vou-are-and-what-to-do business. Its biggest competitor has been a Dutch company called Tele Atlas. Most of the websites, car manufacturers, and gadget-makers get the bulk of their raw material from these two companies. The clients differ mainly in how they choose to present the data. This allows civilians to have preferences.
Despite the digitization of maps and the satellites circling the earth, the cartographic revolution still relies heavily on fresh observations made by people. Navteq, like Prince Henry, produces updates periodically (usually four times a year) for its corporate clients. Its explorers are its geographic analysts. These people go onto the roads to make sure everything that the satellite data says about those roads is true—to check the old routes and record the new ones. The practice is called ground-truthing. The analysts drive around and take note of what they call "attributes," which are anything of significance to a traveler seeking his way. A road segment can have one hundred sixty attributes, everything from a speed limit to a drawbridge, an on-ramp, or a prohibition against U-tums. New signs, new roads, new exits, new rules: if such alterations go uncollected by Navteq, the traveler, relying on a device or a map produced by one of Navteq's clients, might well get lost or confused. A driver making a simple left turn can encounter a blizzard of attributes: one-way, speed limit, crosswalk, traffic light, street sign, turn restriction, two-way, hydrant.
Navteq has more than six hundred field researchers and offices in many countries. In 2006, there were nine field researchers in the New York metropolitan area. One morning in fall, I went out with a pair of them, Chris Arcari and Shovie Singh. "We're going to be working over by LaGuardia Airport," Arcari said. "One of the items we need to check out is some street names. They've put up new signs. Then we'll proceed to an area that we have targeted." Arcari, who was brought up on Long Island, was the senior member of the team, and he tended to speak in the formal, polite, and indirect manner of a police officer testifying in court. He'd been with Navteq for ten years. Singh, who grew up in Queens, New York, was a new hire. He'd got hooked on geography after taking some classes in the subject in college.
They were, you might say, free-driving—no navigation device or map—because they are not only locals but also professionals in the New York-area discipline of getting from here to there. They spend two to three days a week just driving around. Manhattan's grid may be the easiest road network to master in the developed worid (if we overlook some areas), yet the routes leading to and from it can be tricky. The highways are a mad thatch of interstates. parkways, boulevards, and spurs, plus river crossings galore, each with its own virtues and inconveniences. There are many ways to get from point A to point B in New York, and, because of all the variations, anyone can be a route-selection expert, or at least an enthusiast. Family gatherings inevitably feature relatives eating cocktail nuts and arguing over the merits of various exits and shortcuts.
Eventually, we pulled into a gas station near the airport. Singh and Arcari assemMed their equipment. They mounted a GPS antenna, shaped like a giant mushroom, on the roof of the car. The antenna was connected to a laptop, upon which a map would show our progress—a GPS track. Singh took the wheel. Arcari sat in back with the laptop, ready to note any changes.
The first thing the men noticed was a "No Left Turn" sign out of the gas station. "That doesn't go in the database," Arcari said. "That's unofficial, since it pertains to a private enterprise."
An analyst has some leeway in proposing research missions in his territory. "The situation at LaGuardia was something I had noticed myself and thought should be revisited," Arcari explained. In his free time, he'd been driving past the airport and, nudged by curiosity, if not conscience, had made a little detour. He discovered that the Port Authority of New York and New Jersey, which runs the airport, had put up a few new road signs.
"We'll circle around the perimeter and then check the terminals," Arcari said. "As we're driving, I'm checking our information against what exists in reality." Left on Runway Drive ("drop a name check"), merge onto LaGuardia Road (another name check), left onto Delta Arrivals Road. The sign for it was new. "A valid unnamed feature," Arcari said, turning the laptop so that I could follow along as he recorded it onscreen. "I point an arrow to where the feature occurred."
Seeing the road through the eyes of a ground-truther made it seem a thicket of signaee —commands and designations vying for attention, like a nightmare you might have after a day of studying for a driving exam. Once you start looking for attributes, you spot them everywhere.
"Why don't we loop around again?" Arcari said. "I want to be sure we collected everything correctly."
The familiar frustration of going around and around on an airport road was compensated for by the fact that no one was lost or late. After the extra orbit, we drove into a neighborhood next to the airport. Arcari approached the neighborhood by driving around the outside of the "project area," and then going up and down the streets within it. He observed that, driving around like this, you become acutely aware of how many people are not at work. Arcari said that one of the issues that has come up in New York in recent years is the naming of streets and squares for the victims of the attacks on September 11, 2001. We came upon one of them, James Marcel Cartier Way, and Arcari was pleased to see that the name was in the database. A kind of contentment took hold, as other anomalies encountered along the way—an unlikely median strip, a "Do Not Enter" sign— turned out to be accounted for.
Over lunch at a local diner, we discussed various attribute incidents. "One item that was an issue: on the Brooklyn-Queens Expressway, they started renumbering the exits. They did some but didn't do others, so for a while there were two Exit 41s."
After lunch, Arcari and Singh were due back at the central office, in Syosset, to download their findings. They offered to drive me back into Manhattan, but we agreed that it would make more sense for me to take the subway. None of us knew where to find it, though. Subway stations are not attributes; Navteq honors the automobile, a trend started by the makers of road maps of a century ago, whose mandate was to promote auto travel and, with it, the purchase of gasoline, cars, and tires. We pulled into a gas station, and I ran inside to ask for directions.
