Government's location-specific apps need clear context

Real estate agents aren't the only ones trumpeting the importance of location, location, location. That nod to the value of a specific place is driving the development of a new technology called mobile augmented reality. It taps a smart phone’s built-in Global Positioning System device, compass, and other location- and motion-tracking features to display contextually relevant information in real time as the user moves about in the real world.

For example, with an augmented-reality app, you could stand in front of the Washington Monument on a humid, late-summer day and turn to your cell phone’s screen to see information about the structure’s construction and history, get current readings of air quality — or lack thereof — in the vicinity, then plot a path to the nearest Metro subway station for a quick ride to your favorite coffee chain outpost for an iced cappuccino.

“It’s a more powerful way for presenting information and helping people understand it rather than if they simply saw it on a graph or chart on a Web site back in their home,” said Steven Feiner, professor of computer science at Columbia University who specializes in user interfaces and augmented-reality apps.

It is still early in the game for the technology, but government will have an obvious role to play given its stewardship of large quantities of information with a geospatial dimension, such as data about crime, school performance, pollution, voting records, spending programs and much more.

Agency officials will need to give serious thought about how that data gets presented to users, particularly if certain characteristics of the data or shortcomings of the location-tracking technology provide misleading or outright false representations of the facts.

It is most likely that third parties will develop many of the mobile applications that use government data. Some early examples are a couple of augmented-reality applications developed by Sunlight Labs, a project of public advocacy group Sunlight Foundation.

The group’s Congress application, which runs on Android-based smart phones, allows users to determine who represents a state or district based on the phone’s physical location and get contact and other information about that elected official. Another application lets people use their Android and iPhone 3G smart phones to identify nearby businesses that have received contracts under the $787 billion economic stimulus law.

The data used in those applications is relatively straightforward. However, it will get much trickier when developers try to use a tiny cell phone screen to present real-time data that is complex or requires some explanation.

For example, what if that high smog reading a person sees on their smart-phone app in New York City is the worst smog day in five years? Was that reading taken in the middle of Central Park or next to a highway on-ramp with a backup of idling cars nearby? And what if that disconcerting crime data is a report from midsummer, when crime traditionally spikes?

“Consumers of these data need to understand when things are an outlier and not infer patterns if what they are really seeing is point-specific data,” said Jane Fountain, professor of political science and public policy and adjunct professor of computer science at the University of Massachusetts at Amherst.

The interest in augmented-reality applications is happening when there is a big push in Washington and many states to release all kinds of government data. So what is the responsibility, if any, of agencies to shape or dictate how government data is presented and used in such applications? Or is it up to those who develop the applications to provide the necessary explanation and context?

Fountain said she thinks responsibility starts with the agency that releases the data. “A very traditional and long-standing role for the civil service is to help the public understand what’s going on in a nonpolitical way,” she said.

However, it’s still too early to say exactly what that role might require of agencies when they supply data destined for the more restrictive interfaces of augmented reality applications.

Another problem lies with the limitations of smart phones, which weren’t designed to provide the level of location-tracking precision some applications might need. “The [current] technology can lead you to make mistakes about what you think is out there,” Feiner said. But he added that improvements are on the way.

And that’s good, because no one would like being at the right place with the wrong information.