Thanks for all the love on our new bike accident app and deep data analysis--we wanted to help deepen the street-safety conversation here beyond "cars vs. bikes," and clearly tons of you are interested as well. I wanted to share a few things I'm really proud of from this project:
* The relationship between the text story and the interactive map. A lot of "news apps" like this one get cast in a supporting role, as afterthoughts to a fully conceived story. We didn't start this project with either a story or an interactive "container" already in mind--we started with a question (do bike accidents happen in predictable ways?), and then got data, and then figured out what to do next, and then figured out what to do after that. The story and the map developed in tandem. Our transportation reporter Zusha Elinson was fully integrated into the tech process of developing the app, and Zusha's reporting was driven by the data findings. I think it was Brian Boyer of the Chicago Tribune's apps team who told me that good hacker-journo projects represent "the best of what reporters can do on their own, plus the best that hackers can do on their own, smooshed together" (I'm paraphrasing). Going forward, I'd like to better mix in "the best of what community-outreach folks can do," too. On that note . . .
* The "Report Your Own" feature. We realized early on that displaying only officially reported SFPD data wasn't good enough. In talking to bikers, it was obvious that many bike accidents go unreported, even serious ones. We got nearly 30 user-submitted reports in our first full day after launch, and we're working on getting them in the map. We're serious about incorporating these reports into our analysis going forward, and I'm excited about developing a real picture of the safety landscape on our streets over time. Still, I think we could do better in this area, especially with smarter leveraging of social media.
* Our pre-launch "test users". A week before launch, we sent a password-protected link to a fully-functional bike app to half a dozen bike lovers--leaders of local bike coalitions, bike bloggers, a MUNI board member--and asked them to weigh in. What didn't make sense, what left them wanting more (in a bad way), what annoyed them? Their feedback was fantastic--for instance, Leah Shahum of the SF Bike Coalition suggested we add some intro text to the app help explain what's at stake to people who aren't bike nuts like her. A simple change, but it made a big difference. Thanks also to Michael Helquist of Bike NOPA and Cheryl Brinkman of the MTA Board for their great feedback.
So what's next? Adding data from Berkeley and Oakland are high on the list, plus incorporating the user-submitted reports we've already collected (big thanks to the senders, I couldn't believe how much some of you wrote us about your accidents, and I'm sincerely sorry it's taking us a little while to add them in).
I'll let others from our team weigh in on the "superpowers" they contributed to this project--beat reporting, Fusion mapping--and I sincerely want to hear what you think of this work. What worked for you, and more importantly for the purposes of this blog, what didn't?