But to transit wonks and officials in and outside the District, the new dataset falls short of a commitment to transparency. In advance of the release, open-data advocates worried it would continue a trend: that Uber is the sole custodian of its data, a valuable asset, and if cities want to understand its impact, they’re left to sort through piecemeal information the company releases — on its terms. A preview of the new platform didn’t ease their concerns.
“While limited data has been shared, most recently through Uber Movement, the data as currently provided does not allow for meaningful analysis and decision making on a street design level, even for the limited cities that currently have access to the platform,” the National Association of City Transportation Officials said this week, announcing a set of data-sharing principles for private firms and cities seeking its information.
To meet cities’ needs, NACTO guidelines say, Uber should provide anonymized information on multiple points: vehicle speed, volume, travel times, passenger pick-ups and drop-offs. The new dataset ticks off one of the categories, but Uber, which is fighting a regulatory battle with New York City over access to passenger drop-off times and locations, has shown little willingness to provide the others (New York City collects pickup data through its Taxi and Limousine Commission).
“We are not pretending that what we’ve put out today with travel times is the only dataset that a city or outside agency would want,” Andrew Salzberg, Uber’s head of transportation policy, said in an interview at the Washington release party Monday. “We think it’s one very compelling use case where our data — aggregated, anonymized, moved up geographically — can be really valuble for a purpose of traffic management.”
Movement, a soon-to-be-public website accessible to officials with the District Department of Transportation, contains six months of searchable data on travel times for trends and conclusions to incorporate into city planning, Uber said.
A presentation Monday night shed light on how it could be used. Uber demonstrated how the March 16 Metro shutdown impacted road travel times, as illustrated by its network. In most of the city, rides took 10 to 30 percent longer than usual, but in one corridor from Northeast to Southwest, the number exceeded 50 percent. In that band, Uber concluded the choke-points fell near major entryways into the District —Interstates 66 and 395 in Virginia, and Rhode Island and New York Avenue from Maryland, according to its post.
The maps also allows users to spot disparities in transit access in an interactive fashion: In one section of downtown, trips took 37.5 percent longer; in a section of Northeast, the figure was 92.1 percent. Uber’s report, which can be found here, is proof that ride-hailing can’t match the capacity and efficiency of mass transit, concluding that “when [Metro] was shut down, travel times increased dramatically.”
The company says it’s a way of quantifying conclusions that might otherwise be obvious.
“The acute challenge here in D.C. transportation … is Washington Metro,” Salzberg said. “And so, on the day when the whole system was shut down, what was the actual impact on traffic? And we all know it was bad, but can we use this system to measure exactly how bad, and where, and potentially use that to quantify the benefits that the transit system brings on a day-to-day basis?”
It’s similar to data put out by firms such as INRIX, a traffic-data firm that provides large batches of real-time congestion information to cities, planning agencies and private companies.
Still, transportation officials say Uber’s release falls short of a commitment to open data.
“It’s great that Uber is recognizing their impact on transportation congestion in cities and trying to provide information,” Scott Kubly, director of the Seattle Department of Transportation and NACTO vice president, said in a statement. “However, what they’re offering is not consistent with what cities require, nor is it in line with national best practices.”
NACTO argues that a company using public right-of-way has a public obligation to share its information. But Uber contends that some datasets, such as passenger-dropoff information, threaten customer privacy. So why would it be reluctant to comply with all the new guidelines?
Adie Tomer, a fellow at Brookings Institution Metropolitan Policy Program, who spoke as part of a panel at Monday’s Uber event, said there are numerous private firms dedicated to origin and destination data who sell that type of information at a premium. And there’s also the battle between Uber and Lyft.
“This is a little bit of a glimpse into that — it’s not the full thing,” Tomer said of Uber’s network. “To [Uber], that’s their market secrets, so I think they don’t want to give it away.”