Learning from disaster
As the floodwaters recede from the Carolinas, the U.S. East Coast is still assessing the full environmental and economic impact of Hurricane Florence. At least 44 people have been counted dead, a number that could have been higher if emergency officials hadn’t taken the threat of the storm so seriously.
The toll would likely have been worse, too, if Crowdsource Rescue hadn’t kicked into gear. That’s the map-based, emergency evacuation platform created last year by two developers in Houston in a few frenzied hours during Hurricane Harvey. Cofounder Matt Marchetti estimates the app supported 444 Florence rescues over the course of about a week, representing about 2,000 people. On Monday, I called Marchetti, who was en route to a much-deserved vacation, to hear more.
How did you get the word out about the app in the lead-up to Florence?
We wanted as many rescue volunteer sign-ups as possible, plus a big fleet of trucks and boats. The more people we talked to, the better. Even if someone was like, ‘Maybe I’ll respond,’ we wanted to have their info in our database so that we could include them once we drew the big geofence around the area that was flooded and send out texts alerting them to rescues.
So we ran a bunch of targeted ads on Facebook in the towns we knew would be affected: New Bern, Wilmington, Fayetteville. And we did a lot of outreach to existing groups on Facebook: the four-wheeler groups, the truck groups, the fishing groups. People who usually get together and post links, but in this case, we were telling them that they could really be helpful. And some would be like, ‘Hell yeah, my F-250 is jacked up four feet, let’s do this.’
What about reaching out to potential victims who needed you guys?
It helped that we got a lot of national media attention before the storm. Also, with the Facebook ads, we spent a few thousand dollars targeting folks we knew would be highly vulnerable, in terms of age, income, and location. It’s a good thing, because in Hurricane Harvey, Maria, and now Florence, even when we see the same level of flooding in two neighborhoods, if one was poorer they were more likely to use us. We’re realizing we’re a resource for more vulnerable populations, and we’re trying to lean into that.
How many volunteers did you gather up, all told?
We had 745 local rescuers, 70 percent of which had a home address in one of the Carolinas, which is so important because local volunteers have the best local knowledge. They know the people, they know the neighborhoods, they know the roads. It tends to go so much smoother and effectively when you have local volunteers.
We also had about 4,500 people sign up as remote volunteers, which was actually too many. These were folks who helped dispatch rescues, coordinating back and forth between evacuators and evacuees, and were helping make sense of the storm.
It seemed like pre-storm evacuations went a lot more smoothly with Florence than other recent mega-storms. Would you say so?
We got a lot of pet rescues where there were no people home. That tells me a lot of folks evacuated. So it seemed like orders were heeded, for the most part. But we still had more people rescues than animals. And you know, I always try to tell people, it’s above my pay grade to say why someone does or does not evacuate. Just because some choose to stay doesn’t mean they’re idiots. They might not have, or there are extenuating circumstances. They still deserve love.
Lessons learned for the next event?
There have to be more formalized and official partnerships with spontaneous volunteers and the official emergency management agencies. Otherwise it’s just chaos. We consider our platform one of the best ways to do that, but it’s hard to do during a response. So it’s part of our ongoing work to make those connections with EMs during “steady state” [Ed: i.e., not during a disaster].
We coordinated a lot of successful rescues. But some did not succeed. We did have at least one case that a rescuer followed up on, and the person was dead, and we passed that on to the police. That’s rough. You know, for us, this takes everything personally. You can prepare for a year, and you still don’t feel ready, and it still feels like a personal failure every time someone dies. It sound naive, but with disasters like these, I do think a lot of problems can still be solved with more engagement ahead of time.
Disrupting place data
Google Maps has such a monopoly on geographic knowledge that it’s hard to rely on any service to locate places and routes. Map developers in particular have a love-hate relationship with the world’s most popular map, which charges sky-high rates for the use of its top-notch place data.
But a new startup wants to disrupt that old data model for certain types of information. Founded by former leaders of the defunct open-source map company Mapzen, Streetcred wants to crowdsource a new library of “points of interest”—i.e., restaurants, bars, schools, parks, subway stations—by paying volunteers (in cryptocurrency!) to contribute the data. Eventually, “we’re hoping that the income can be similar to driving an Uber,” Randy Meech, the CEO, told me.
This week, the company launched Map NYC, a competition with Bitcoin prizes, to see who can gin up the most points of interest on a blank map of New York City. Cartographers, start your engines, and read more about the whole project (including how the money works) here.
Mappy links
Fighting to exist: the communities where environmental injustice could flare up in North and South Carolina. 
Leader of the map world: how Jen Fitzpatrick rose up the ranks at Google Maps. Spy games: open-source geographic intelligence is changing government surveillance. Digital divide: bad maps are stymying the spread of broadband. Internet safari: the history of the web, told through maps. Great way to improve a book: a delightful literary map. It flows to the satellites: aerial imagery captured North Carolina’s murky rivers, post-Florence. Bad romance: how Grindr’s location services expose its users.
As always, I want your feedback! How can I do better?
And while you’re thinking about it, forward this newsletter to a friend. They can sign up here.
Laura