Game Changer ADDING THE TIME DIMENSION


bradyforrest Brady Forrest tracks the intersection of the internet and all things geo. As the Chair for O’Reilly’s Where 2.0 Conference (conferences.oreilly.com/where) he has been following the growth of mapping and location-sharing for several years. Brady lives in Seattle when he is not on the road running Ignites (ignite.oreilly.com). You can find his regular tech thoughts on O’Reilly Radar (radar.oreilly.com).


arrows_2AS DI-ANN IS DRIVING AROUND THE BAY Area she uses Waze (http://waze.com) to find her way. From Waze she gets routing, traffic and road closure information. In exchange, Waze tracks her location and knows her ultimate destination. It knows that she is going slower than expected on Highway 101 and that she is probably in traffic.

By aggregating similar data from its users across the Bay Area, Waze is also learning new routes from Di-ann. If she takes a different route, Waze will record and remember it as a valid route. Di-ann is Community Geographer and Customer Advocate for Israeli-based Waze.

Waze is taking advantage of the new possibilities opened up by 24/7-connected location-aware mobile phones. Users are aware that they are being tracked to benefit Waze. As Waze rolls out in different cities and countries, it tells initial users to keep Waze on as they drive—even if they don’t need Waze’s help finding their destinations—because their driving patterns will train Waze, and users will ultimately benefit. To encourage this sharing of information, Waze lets users accumulate points for the miles they drive. The first person down a street gets extra points.

Large Companies Play Too

It’s not just well-funded startups that are taking this approach; companies of all sizes are learning to rely on user-contributed data. Tomtom, the large PND (personal navigation device) manufacturer, also uses user data from its large install base (though Tomtom is not using it in a realtime manner). Users upload the data after they travel, and Tomtom rolls out quarterly updates. The iPhone app users can take advantage of it, but they do not currently contribute.

Nokia purchased NAVTEQ a year ago. They have also announced that they will be using passively collected traffic data via Nokia handsets (http://bit.ly/gc2-1).

Even Google’s doing it. Google has announced that the company is now using its own mapping data. These data are generated by Google’s Street View cars, but are supplemented by mobile phone users (passively tracked while using Google Maps for Mobile) and online users (who actively submit changes). Earlier this year, Google announced that it was using user-derived traffic data.

Mobile Footprints
Sense Networks (http://sensenetworks.com) is another company that trades a service for valuable location data. Using data from mobile phones and cabs, Sense Networks is able to determine consumer activity. By comparing these to historical data, Sense Networks is able to spot abnormal blips that are valuable to hedge funds. Demographic and behavioral data reveal whether a person is a commuter, a tourist or a local who is out drinking.

Sense Networks is able to sell this consumer data to hedge funds whose managers are always looking for the next Alpha data set. What do consumers get in return? They get CitySense (www.citysense.com), a heat map of the city that lets them know how active bars and neighborhoods are, so they know where to head for a quiet drink versus a crowded club.

Not everyone relies on the consumer to upload data. London-based Path Intelligence (www.pathintelligence.com) is focused on hyper-local activity by deploying antennas at malls, amusement parks and conference centers—anywhere with crowds. These devices can detect the location of mobile phones, and malls are starting to use them to track foot traffic for use in setting rental rates and optimizing foot traffic patterns. In Figure 1, a mall manager/owner is able to correlate actual foot traffic with other relevant environmental shopping variables viewable in the dashboard. The location’s owner pays for the antennas, but Path Intelligence gets to keep the data as well. In the near future, the company will be selling data products that show trends across shopping centers.

FIGURE 1. Path Intelligence provides a graphical interface for viewing the data related to malls or amusement parks, mostly in the U.K. In this view you can see which areas are getting a lot of foot traffic and which could use some help. Courtesy of Path Intelligence.
FIGURE 1. Path Intelligence provides a graphical interface for viewing the data related to malls or amusement parks, mostly in the U.K. In this view you can see which areas are getting a lot of foot traffic and which could use some help. Courtesy of Path Intelligence.

Adding The Data

Making use of this time-based information will require new methods of viewing data (for example “when were people where?”).

Web-based cartographers find themselves using dynamic maps to cope with constantly changing information and are developing clever applications to ease the consumer’s research into, for example, finding the perfect place to live based on commuting time or crime patterns in the city, or on housing developments over time.

Here are three examples of representing time statistically, and of letting users manipulate time for analysis:

Commute time

MySociety, a nonprofit, open-source community based in the UK, runs most of the best-known democracy and transparency websites in the UK. It has created an application that will allow users to determine where they want to live in London based on commute time and housing cost. The application’s map has two sliders, illustrated in Figure 2, which allow users to run easy scenarios based on how much they want to spend and how much time they are willing to tolerate on the Tube.

FIGURE 2. This map shows all of the places where the median price of a home is higher than 230,000 pounds and has a commute to downtown London between 32 and 64 minutes. Courtesy of MySociety.
FIGURE 2. This map shows all of the places where the median price of a home is higher than 230,000 pounds and has a commute to downtown London between 32 and 64 minutes. Courtesy of MySociety.

The result is a recommendation of optimum neighborhoods that meet users’ needs.

Scene/Time of the Crime

Neighborhood crime watch has takenon new meaning with Stamen Design’s Oakland Crime Spotting application. Stamen Design has mapped crimes across San Francisco and Oakland, allowing anyone to monitor crime patterns in the city, from assaults to murders. The Stamen “Pie of Time” illustrated in Figure 3 allows users to see crime trends based on time of day.

FIGURE 3. San Francisco Crimespotting application maps the time of crimes for accuracy of crime statistics, and context for when different types of crimes take place. Courtesy of Stamen Design.
FIGURE 3. San Francisco Crimespotting application maps the time of crimes for accuracy of crime statistics, and context for when
different types of crimes take place. Courtesy of Stamen Design.

Development time

Trulia Hindsight (http://hindsight.trulia.com) is a housing development animation developed by Trulia, the online real estate search company, to show historical housing data for the 20th century. These historical data were traditionally viewed in table format and generally as an appendix to some other lengthy text about a city’s economic development history. Now Trulia Hindsight allows anyone to run an animation of these data that reveals the development progress; this animation will let users see when and where houses were built. You can imagine this same technique being used to play back a day’s or a week’s worth of sensor data.

The real world, and people in it, are becoming mapped and tracked. With millions of mobile sensors on the move, anyone will be able to collect needed data that was never available before. If a business depends on or could benefit from these types of data, it is now going to be possible to acquire it. The market is still being formed, and the companies mentioned above are only the beginning.