Coord makes San Francisco curb rules available for community use.
In the past few years, the humble curb has become one of the most hotly contested pieces of real estate in cities. In addition to long-standing demand from private cars, taxis, buses, trams and delivery trucks, now ride-hail services, bike lanes, bike-share stations, car-shares, scooters and e-bikes are also competing for use of this precious piece of concrete between the street and the sidewalk. As Aarian Marshall from Wired aptly put it: “to see the future of cities, watch the curb.”
This increased demand often translates into congestion at the curb, which can have rippling effects on traffic flow and safety. Cities such as Washington, D.C. and San Francisco have begun looking for ways to better manage their curbs by designating select blocks for passenger vehicle pick-ups and drop-offs. Other cities have launched programs to digitize their public street inventory but, as Willa Ng at Sidewalk Labs has previously discussed, this is typically a very costly endeavor. This is why we embarked on a journey last year to explore more efficient ways to digitize city curbs. Our Curbs API makes this information widely available for multiple cities in a standard format, allowing software developers to easily integrate it into their applications.
Today, we are excited to publicly release the digitized curb rule data we’ve collected in San Francisco! We’ve made this information available for non-commercial use by researchers, local agencies, mobility companies or anyone else in the community, and we’ve also built a Curb Explorer so you can see the data on a map. We hope this will help jumpstart the conversation on the importance of the curb in the rapidly evolving transportation landscape.
How we brought the curb to the cloud
There are two components to digitizing the curb. The first is collecting information about the curb features on the street. These assets — parking signs, curb cuts and paints, hydrants, and bus stops — define the allowed uses of the curb. Conventional surveying methods rely on measuring curb feature locations with a surveyor’s wheel, and recording sign text by hand — a time- and labor-intensive effort. “Why not just use street imagery?” many have asked. Well, we explored such newer data sources, but we found that curb features were often obstructed by parked vehicles or trees in the images, and that it was very difficult to get precise measurement and location information.
To address this challenge, we decided to take to the streets ourselves with a next-generation surveying tool: a smartphone app that leverages augmented reality technology to help “code the curb.” With this app in hand, surveyors snap photos of curb features as they move down the block at walking pace, averaging just three minutes to survey all the features along the curb of a single block. This method preserves the accuracy of traditional techniques but is an order of magnitude faster.
The second part of digitizing the curb required translating the collected data into rules. We do this to transform the raw, point-based curb feature data into user-friendly information on what rules apply to which curbs at any given time. Remember the last time you stared at a parking sign, trying to make sense of the complicated (and sometimes conflicting) information, while also figuring out which stretch of the curb the rules applied to?
In this translation process, we take into account municipal-specific definitions and idiosyncrasies (for example, San Francisco loves its motorcycle parking!), and attempt to resolve any conflicting rules. In addition, we incorporate the city’s parking meter data to add price information. The result? A linear-referenced dataset that details how a person or vehicle is allowed to use every stretch of the curb on any given day and time. We have intentionally built it in the GeoJSON format, making the data easily queryable and mappable for as many software systems as possible.
If you’re going to San Francisco…
For San Francisco, our data encompasses the commercial districts of the city where curbs see the most demand: North Beach, Chinatown, the Financial District, Russian Hill, Nob Hill, the Tenderloin, Union Square, Marina, Western Addition, SoMa, Mission, and Castro, and commercial corridors in Inner Richmond and Inner Sunset.
Land of the free (parking)
Even at 9 a.m. on a Monday, most of San Francisco’s streets in our dataset allow parking for private vehicles, with the exception of SoMa, Financial District, and Union Square. Paid parking is implemented along the main commercial corridors, but is free for two hours at a time pretty much everywhere else. After 6 p.m., the vast majority of curbs in our coverage areas turn into a free-parking haven (though some require permits) until the next morning. Keep in mind, car-share users!
Where are the dedicated passenger loading zones
According to the San Francisco County Transportation Authority’s “TNCs Today” data, the Financial District, Union Square, and SoMa see the most intense ride-hail pick-up and drop-off activities, with peaks at 8 a.m. and from 6 to 7 p.m. on typical weekdays.
Though these areas have a much higher concentration of no-parking zones than elsewhere in the city, there also happens to be many no-stopping zones — which by law forbid taxi or TNC vehicles from picking up or dropping off passengers along these curbs. There seems to be an opportunity to rethink the curb-use designation and management in these areas to enhance safety of the streets.
“Knowledge is power,” Francis Bacon famously said. We hope the information we are making available today will not only help the public and private participants in the mobility market interact with the curb more efficiently and safely, but also enhance cities’ capacity to reshape their regulations and management of curbs. We believe that widely distributing accurate information about mobility services, such as the use of a curb, is a key step in accelerating the adoption of a digital mobility market that more effectively matches mobility supply with demand.
We certainly can’t accomplish this alone. If you share our vision, we look forward to hearing from you at firstname.lastname@example.org.
This article was drafted in collaboration with Corinna Li and Amy Kyleen Lute. Illustrations by Stephen Kennedy.