Building the Self-Serve City
How do you know what needs fixing to make your city better? Let’s say you wanted to understand the relationship between new construction and air quality on your block. You might start by going to your city’s website to see if they have data on building permits and air quality. Perhaps you stumble upon a few websites that share data for your city, but the data you find is organized differently — building permits are recorded by street address in an Excel spreadsheet, whereas air quality is recorded by district on a Google Map. You may also wonder what other data is available that could help you understand your street, such as noise complaints, traffic jams, tree counts, and crime. You’re not a developer or a data scientist, so hunting down multiple data sets, creating a standard data format, and running it all through a visualization tool is just not feasible.
At Stae, we’re working towards a data commons: a shared, open exchange of civic data that is universally standardized and accessible so that everyone can query and build with civic data. Our vision is to create a tool that connects our physical cities and the data they generate in ways that are accessible, open, and responsive, enabling simple searches to complex app integrations. To mark the launch of Stae’s new developer portal, we chat with Eric Schoffstall, co-founder of Stae and head of engineering, on his formative, civic-data experiences that inspired him to build Stae.
Tell us how you first got interested in city data.
Growing up in Phoenix, I knew my neighborhood and the metro region as a whole could be a lot better. In 2014, Code For America decided to do a fellowship in Mesa, a suburb 20 miles east of Phoenix. I was part of a group of about 30 plus engineers who started the Code for Phoenix Brigade and we saw it as our chance to build up the region’s civic data. Our initial meetings were full of ideas, but the data we needed to build with was nonexistent. I spent weeks hunting down year-old exports on hidden GIS servers, wrangling CSVs, writing ETL jobs, scraping web pages, and trying to get enough of anything to keep our momentum going. In the end, there just wasn’t enough data to keep everyone interested and the brigade eventually disbanded.
Were you able to build something for Arizona?
I ended up building a pollution monitoring map that pulled data from county well reports, state pollution violations, and federal superfund data. You could type in your address and see nearby pollution incidents on a map. I got the data from different levels of government, in different formats, and sometimes, only by scraping a webpage. The cities that comprise the Phoenix Metro Area have separate open data efforts from each other — so getting data for the region as a whole was extremely difficult. Since we didn’t have buy-in at the government level, we didn’t get support finding the data or testing the tool. Despite the difficulties of the project, I learned a lot about the complexities of working with city data and saw the need for a tool that could facilitate better collaboration between developers and city governments.
How did you first get the idea to build Stae?
After my experience in Code for Phoenix, I left feeling really frustrated. I wanted to make something that would be positive for Arizona, but to do that, I had to almost hack the State to actually make something of civic value. I saw the deep disparity between how the technology industry had benefited from building open and accessible platforms and by comparison, the isolated condition that cities and their data were in. In fact, cities have often been kept out of the loop and even negatively impacted, left to operate in a reactive manner and catch up to what is happening around them. I had the idea to build an open platform for developers and cities to work together. This would shift cities into a more collaborative, proactive role — setting a vision for the future instead of reacting to one created by Silicon Valley.
Now that the developer portal of Stae is live, what kinds of things do you hope people build with Stae?
I’m really excited for developers to use a single API for the entire country’s data, in a consistent format, and in real-time (instead of encountering hundreds of CSVs in a myriad of formats, which was my experience!). When you build one app with data from Stae, it can work in any city, anywhere in the world. This can enable a tighter integration of civic data into any application and also augment existing app functionality.
I want people to finally see it as possible to integrate and build with the data generated by our surroundings. For example, your bike route to work could automatically re-route you if the bike lane pavement conditions are dicey. Or, you wouldn’t need to circle the block for 20 minutes trying to find a spot if the city is reporting parking availability in real-time. And your taxi could be re-routing based on real-time traffic jam data.
Why is it important to get more developers working with city data?
If we stay down the same path, we’re going to end up with communities falling even farther behind and technology companies monetizing even more of our public lives.
If you had time away from building Stae, what’s something you’d build with Stae?
Lately, I’m feeling very passionate about creating new tools to address urban blight — litter, abandoned buildings, trash in the streets, that kind of thing. I think something that could help keep streets clean is higher street sweeper frequencies, but that’s not feasible because people already have to move their cars multiple times a week in dense cities like New York. If we used traffic cameras to count parked cars, we could keep real-time parking availability in Stae to notify the Department of Sanitation that a street could be swept without the overhead of having people move their cars. It’s basically an opportunistic street sweeping strategy.