You’ve reached the end of the line.
And how they’re giving public transit data back to the American people
June 22, 2015
If you know anything about government, it’s that the government can be pretty outdated when it comes to tech.
And that’s what makes Code For America so important.
Officially, Code for America’s goal is to make government services “simple, effective, and easy to use” with the help of open source technology. Which is good news for our team, since we need open government data for our app to work.
Open data! It’s a word bandied about a lot these days. But you should recall that just six years ago, transit agencies were still sending cease and desist letters to developers who tried to publish their timetables online. Weren’t those the days, eh?
Now why is open data so important to us? Our pride and joy — Transit app — is the most popular urban transport app in North America. It helps millions of people access transit information in 99 cities — including Cleveland, where we’ve just integrated real-time data. But without transit agency data, we’d have nothing to show our users.
Largely thanks to the efforts of organizations like Code for America, open data has come a long way — from the radical pilot project of a few rogue innovators to generally-accepted practice for most agencies.
With Code for America, we’ve been able to leverage the power of open data to accomplish some pretty amazing feats — and ruffle some pretty significant feathers along the way. And together, we’re changing the way public transit riders get around their cities. Here’s how:
Chattanooga is a city in Tennessee. According to Wikipedia, it’s surrounded by ridges and mountains, has a population of around 170,000, and is known by its various nicknames: River City, Chatt, Nooga, Chattown, and “Gig City”. That last one? Chattanooga also has the distinction of having the fastest internet service in the United States.
Beyond that, we didn’t know much about Chattanooga. Until the nice folks at Code for America got our attention.
At Transit, we integrate public transport information city-by-city, agency-by-agency, all over the world. Bus, subway, carshare, bikeshare, Uber — we have it all. And since we’re a lean start-up, we get the crowd to help us choose which cities to add next.
Users up-vote their agency on our team’s wishlist, and the top ranked ones (with open transit data) are prioritized when adding new cities. To date we’ve added support for 99 cities and close to 300 transit agencies. Unfortunately for Chattanooga, their transit data wasn’t open — and the small size of their city made it unlikely that they’d be able to outvote highly demanded cities like Berlin or Sydney.
That didn’t faze the team at Code for America.
This here is a handwritten schedule that was given to Giselle Sperber, a Code for America fellow, by her hostess when she arrived in Chattanooga last January. It’s a list of departure times for the bus stop outside.
Quaint, right? Until Gisele went to double-check the transit times on Google Maps, and realized that transit directions weren’t supported.
Chattanooga might have the fastest internet in America, but its transit data wasn’t even available online!
There was no easy way for city residents to plan transit trips or quickly check bus schedules. So Giselle and the Code for America team decided to partner with Chattanooga’s transit agency, CARTA, to build digital transit schedules that followed the global transit data standard, GTFS. Then, they made it available to third-party developers on GitHub.
The heavy lifting was over, but despite having access to open schedule data for Chattanooga, Google Maps had yet to integrate it. So instead, Code for America fellows bypassed the Googleplex and organized a social media campaign on Facebook, Twitter, and Reddit to try and gather votes for the Transit city wishlist. How clever.
Shortly thereafter, little old Chattanooga had passed the biggest cities on our list. What Chattanooga lacked in size, they made up for with red-hot transit passion.
Three days later, and Chattanooga was ready to roll. We added schedules, trip planning, and even integrated the city’s bikesharing program for good measure. Moreover, we added real-time transit info, which is in a format that Google Maps (regretfully) doesn’t support.
Just like that, Chattanooga had its very own transit app. For zero dollars, and without an RFP. It was a textbook example of how Code for America and its partners could leverage technology to make public services more accessible, and with very little overhead.
But you don’t live in Chattanooga. Why should you care?
After our experience in Chattanooga, we kept an eye out for interesting transit projects that Code for America was working on. We stumbled upon a project led by Code for America Brigade Captain, Chris Whong, who figured out how to scrape Baltimore’s real-time transit data from the agency’s clunky website. His new website showed real-time bus locations on a big map.
A short primer: real-time transit data is super useful. When you know exactly when to expect your bus to come, you can more efficiently plan your day, and you have less anxiety and uncertainty when waiting at the stop. It’s especially important in Baltimore, where more than half of the city’s residents don’t have access to a car, and where waiting at an unheated bus stop in the winter can get pretty uncomfortable.
Baltimore’s real-time transit tracking system had cost the city 2.7 million dollars to develop, but its data was locked up in a proprietary format that the city said would cost “an additional $600,000” to standardize. Other than Chris’ city-wide map, the only way to get real-time bus positions was within the agency’s proprietary bus-tracker.
It was kind of unpopular with users.
And yet, even without real-time information, 20,000 Baltimoreans were still using Transit app every month to get around their city. Baltimore’s bus system is notoriously unpredictable, and we wanted to give our users more accurate information.
So when Chris went ahead and figured out how to scrape the city’s real-time data, we reached out to him and integrated Baltimore’s real-time data ourselves. One afternoon later, and we were done — Baltimore finally had a usable transit app with real-time information.
We documented our second coup with Code for America in this cheeky blog post.
The post went viral, blowing up on Baltimore’s Reddit page and amassing thousands of referrals from Baltimoreans on Facebook and Twitter. The post created a firestorm in the press, with The Baltimore Sun, Baltimore CityPaper, Technically Baltimore, Ars Technica, Government Technology, TransitWire, Maryland Daily Record and the Baltimore Brew all offering their own commentary on our blog post. Even Baltimore’s own transit agency released an official response, and hesitantly endorsed our work.
We were surprised at how one simple blog post was able to get so much media attention, especially since it described a political issue most people don’t know or care about: how to leverage data to make transit more accessible. Of course, it helps that we sometimes save cities hundreds of thousands of dollars in the process, but that’s just icing on the cake.
The only thing that we really spend time worrying about is how we can give riders the best user experience possible — and that’s only possible when we have data to work with.
In Chattanooga, Code for America helped us by building the schedules from scratch; in Baltimore, they made it possible to access real-time information when it was locked up in a proprietary format. In both cases, everybody won: we gained more users, transit riders were given a better way to get real-time information about their transit system, and transit agencies didn’t need to spend any taxpayer money.
And we weren’t done yet.
After our blog post about Baltimore, we were amazed at the positive response we started getting from open data advocates across the US.
Here’s an email we received from a Code for America participant from Cleveland not too long ago:
We looked into it, but Cleveland’s data was a little bit different than what we dealt with in Baltimore. Not to worry — we ran into the CTO of Cleveland’s transit agency at a public transit conference (where we were giving a talk on open data and mobile apps) and he gave us his contact information.
When we got back to Montreal, we shot him an email. “We won’t overload your servers!” we promised. One phone call later, and his team gave us their enthusiastic blessing to scrape the agency’s data. Now — wouldn’t you know it — Cleveland has joined the twenty-first century.
Upgrading from this:
At no cost to the agency.
The power of open data (and even not-so-open data) makes Transit app possible. Along with our friends at Code for America (and civic hackers around the world), our team is transforming the way that public transit riders navigate their cities.
Instead of burying critical transit information in expensive proprietary apps and unintuitive websites, our app gives riders a way to access the critical information they need, quickly and easily.
Cleveland is one of the places where open data is making a real difference in civic life. They have a vibrant community of open data advocates, including groups like Open Cleveland (Cleveland’s Code for America chapter) and Hack Cleveland. Working with partners like Cleveland Neighborhood Progress and Equality Ohio, they organize hackathons and collaborate on projects like improving police accountability and providing more transparency on the workings of city council.
And of course, better data availability is also improving public transit. While Cleveland’s real-time information isn’t fully open quite yet — remember, we’re just scraping their website— if our friends from Code for America hadn’t pointed is in the right direction, there still wouldn’t be an easy way to get transit info in Cleveland. We hope that the quality and openness of the data will improve in the coming weeks.
Having the right tools for urban mobility is especially important in Cleveland, which is quick becoming a poster child for good public transit. More than 25% of people in Cleveland don’t have a car, and Cleveland’s rail ridership is booming, with the highest recorded ridership in 27 years. The city has the HealthLine, one of the best bus rapid transit lines in America, and Cleveland just announced plans for a bikesharing service with 700+ bikes to roll out before the Republican National Convention next July (which we’ll have ready to go as soon as it launches).
At any rate, we’re excited to see how our Cleveland users respond to real-time integration, and look forward to working with our friends at Code for America to open up transit data for more and more American cities.
Even though we’re Canadian.
What happens when you ask riders if there’s something strange (or great) on their trip?
What happens when you get transit agencies from across Canada together in one room?
App expertise from Transit and operational know-how from Keolis delivers better rider outreach, innovative AV solutions, and more