Amid the horror of a natural disaster in which people have lost loved ones and perhaps their homes and livelihoods, it is difficult to imagine that anything as simple as a smartphone or desktop app could make a difference to such a tragedy’s devastating toll.
Unlikely as it may sound, in hundreds of hackathons worldwide, thousands of developers are brainstorming and coding open source software apps that they believe stand a chance of reducing the impact of natural disasters.
Here’s why: hurricanes, floods, earthquakes, tsunamis, and wildfires are predictable phenomena, and scientists know pretty much where and when they can happen. Also, as they have often happened before, there is data on what worked—and what didn’t work—as people struggled to escape, recover, and reconstruct their lives last time around.
To a growing band of software developers, such data provides fuel for a new breed of apps that might just save lives. This is not just the work of a bunch of well-meaning geeks: these developers are working with humanitarian relief experts at the United Nations (U.N.) to make sure it all works.
“This is a meeting of two worlds: those with 20 years’ experience in conflict and natural disasters, but who can’t troubleshoot an email problem, are meeting crowds of developers with loads of technology plans, but no idea what it takes to respond to natural disasters,” said Laurent Sauveur, director of external relations for the U.N. Office for Human Rights in Geneva, Switzerland.
“While technology has a lot of potential, unleashing it means making sure that these two worlds understand each other’s constraints. And that takes investment,” Sauveur says.
Such investment has been forthcoming, in the form of a competitive initiative called Call for Code, which is now in its second year, and which is the brainchild of David Clark, a Boulder, CO-based backer of humanitarian causes.
With its aims backed by Sauveur’s organization, alongside experts from the U.N. Office for Disaster Risk Reduction, Call for Code is sponsored by IBM to the tune of $55 million over five years, for disaster app development and deployment.
Call for Code comprises many hundreds of hackathons in the 170 cities where IBM has offices, and initial ideas will be honed further and further as the year progresses, until a bunch of them are ready to compete for funding.
In 2018, for example, the $200,000 Call for Code competition prize was won by Project Owl, in which five American developers created a buoyant, battery-powered Wi-Fi node called a ‘duck’ that can be dropped from drones or helicopters in disaster zones to form mesh networks that quickly restore communications.
Recalled Clark, “Those five winners were from all over the United States and they actually met on a Call for Code Slack channel, and came together with this ingenious idea.”
Fuelling developer interest in the competition is Clark’s penchant for capturing the imagination of celebrities, with what he calls “the good and the great” on the Call for Code prize jury. He explained, “So if you actually win, it’s not just all about the money, or the incubation, or the introduction to VC firms; it’s the fact that you’ve been selected by such an esteemed panel.” Former U.S. president Bill Clinton was a 2018 jurist, for instance.
For the 2019 competition, supporters include Sting, Cher, and Black Sabbath frontman Ozzy Osbourne. “They’re tweeting Call for Code out to millions and millions of people,” says Clark.
Yet, it’s the millions and millions struck by disaster who really matter. To find out how Call for Code works, in early June I was invited to observe a two-day hackathon at Sauveur’s headquarters, an imposing edifice called the Palais Wilson on the Lake Geneva shoreline that was home to the U.N.’s predecessor, the League of Nations.
There, 20 software developers from a variety of industries, alongside a gaggle of senior IBM engineers and a clutch of U.N. experts in human rights and disaster relief, set about developing open source risk-reducing applications to improve the way citizens, businesses, and local government predict, cope with, and recover from natural disasters.
Sauveur and colleagues wanted the apps to do one more thing where possible: respect victims’ human rights. That would ensure, for instance, that elderly and disabled people have a say in their rescue, rather than being moved around en masse—and perhaps away from family and friends—without their say-so.
In order to ideate such apps, the hackathon’s assembled developers were told, they needed to ignore a common perception about disaster relief: that everything moves along fine once we see food and medical aid arriving, usually in spectacular airdrops, on TV newscasts.
“This goes against a lot of the things you see on TV, especially in the big disasters. But when somebody is throwing bags of rice from helicopters, that’s actually bad news in one sense, as it means that we have failed at every step until that point to implement a series of possible risk reduction interventions that would have been much more cost-effective, and much more efficient,” says Adam Fysh, program officer at the U.N. Office for Disaster Risk Reduction, also based in Geneva.
It is those risk-reducing interventions the hackathon teams need to deliver. With that thought ringing in their ears, the developers set to work, with as much IBM technology at their disposal as they needed – including Watson APIs featuring artificial intelligence, IBM cloud services, and the firm’s Hyperledger blockchain-based immutable database technologies.
To encourage a broad approach to problems, Call for Code selected developers from a broad spectrum of business types, with coders coming from consultancy CapGemini, Lloyds Banking Group, Deutsche Telekom, Royal Bank of Scotland, Irish software house NearForm, Indian enterprise IT services firm Persistent Systems, the ad agency WPP, the Linux Foundation, and IBM-owned Red Hat.
The group, guided by IBM chief developer advocate Willie Tejada, divided up into four five-person app creation teams. “You’re not building complete product,” Call for Code CTO Daniel Krook reminded them. “What you’re going to build will live on past this room. The code will go into GitHub for other developers to work on.”
One direction given the first team to explore was the concept of “building back better,” learning from historic disasters how to reduce risk in future events to create greater community resilience that will aid recovery, rehabilitation, and reconstruction. The app plan hatched by that team was multipronged, as they imagined three apps: one for victims seeking information on how they could help the recovery operation, another for businesses hoping to get back into operation, and still another for local government officials to assess damage and prioritize rebuilding jobs. Like many a computer project, however, it grew; “We thought about having small apps at first, but then came up with a landscape of larger systems as the subject is really quite broad,” says team member Vinaya Shastrakar, enterprise architect at CapGemini in India.
A second team drew up plans for a pair of apps for improving flood and drought prevention. One was an advice-rich mobile app for flood victims offering personalized evacuation routes, the other a dashboard for first responders and local authorities that uses past data to tell them which areas are at greatest risk of a deluge, allowing them to prioritize rescue plans. “The weather and flood-plain data is available on where and when floods will happen,” says David Williams, a developer with Royal Bank of Scotland. “It’s a question of being able to correlate all that data and being able to make visualizations that help planners.”
The third team’s efforts involved development of a humanitarian aid app that could help vulnerable, less-mobile individuals in communities. “The idea is to give all populations equal access to humanitarian aid and assistance,” said Cheryl Hung of the Linux Foundation. People would be able to make an informed choice about evacuation destinations, she said, and would receive help in finding their families when they get to a shelter. “So they’d have access to security and shelter and the right to family reunification.”
Elsa La Pennec, Emergency Response Officer at the U.N. Office for Human Rights, cautioned that such measures could easily fall apart. Supposing a vulnerable person can use an app in a disaster situation is a big ask, she believes: it could be undone, La Pennac says, if the person simply “has lost her glasses and cannot find her phone.”
Finally, a further appalling aspect of natural disasters is that the mayhem unleashed in their wake can provide cover for many kinds of human rights abuse, ranging from extrajudicial killings to gender-based violence such as rape to brutality by militias or law enforcement. To address such situations, the fourth team at the hackathon developed an app that allows a person to record their story, encrypt it securely, and send it to the U.N.
“We want to give people a voice that is confidential and safe, so that the U.N. can bring perpetrators to justice,” says Rabia Mahmood, one of the developers on the team and a senior software engineer at Lloyds Banking Group in the U.K. Conor O’Neill, chief product officer at NearForm, says an emerging secure transmission protocol called Secure ScuttleButt (SSB), which establishes a cryptographic social network between users, looks a likely basis for such an app, one that later Call for Code hackathons can utilize.
Well-thought-out apps that bring human rights and justice to those struck by disaster are certainly needed. Harnessing smart disaster prediction technology, and informed real-time response in the critical 24 hours after a disaster strikes, should be able to reduce the high drama that often surrounds tragedies wrought by nature.
Or, as Fysh put it: “Disasters should be boring.”