People are building technology that could survive the apocalypse

Thinking about the apocalypse has long meant thinking about technology.

The U.S. interstate highway system was built with an eye toward evacuating motorists before an impending nuclear war, and phone companies have long built telecom switching centers to withstand both earthquakes and bomb blasts so networks can ideally stay online even after a disaster.

But in a serious enough situation—the kind that leads to a full or partial societal collapse—tech networks and even hardware supply chains could be cut off temporarily or permanently. It’s not just science fiction: Already, large-scale disasters such as Hurricane Maria in Puerto Rico can leave areas without electrical power and connectivity for long periods of time, and that’s during a time of peace and relative political stability. Refugees fleeing disaster and war around the world often find themselves in areas with limited power and connectivity, even if they have cellphones or computers with them. Digital attacks on power grids and other vital systems have already become a tool in geopolitical disputes that can affect civilian life.

There’s no reason to think these kinds of situations will become any less common as the effects of climate change become more pronounced. That’s why researchers and engineers are thinking about—and building—resilient technology that just might be able to survive if we lose the power grid, the tech supply chain, and the internet.

From grungy sci-fi to an unsettlingly possible future

Scavenging for complex industrial products such as car parts, medicine, and gasoline has long been a feature of postapocalyptic movies, TV shows, and video game franchises like Mad Max, The Walking Dead, and Fallout. And now that digital technology has arguably become as much a feature of modern society as the phone and the automobile, engineers and scientists are thinking seriously about what computational resources might be salvageable, maintainable, and useful after a serious societal decline.

Examples include technology for managing small-scale electrical grids, optimizing barter transactions, and managing information about food and agriculture, according to a paper published in 2012 by researchers from the University of California at Irvine, Indiana University, and the Bureau of Economic Interpretation. If computing resources become more scarce and networks fracture, even deciding what information to store could become an important question, potentially leading to scenarios such as communities dividing up pieces of resources like Wikipedia to try to preserve it across computers, says Bill Tomlinson, a professor and vice chair of the Department of Informatics at the University of California at Irvine.

Although sudden cataclysmic scenarios such as an asteroid strike or nuclear war aren’t out of the question, researchers say that historically it’s common for societies to collapse gradually, with declines in trade, communication, and the standard of living taking place over years or generations rather than all at once. And that makes it more likely that technology beyond basic hand tools (and weapons) could continue to be useful for longer.

“If you think that things will go to hell in a handbasket tomorrow, probably you’d do best to get out of computing and buy beans, bullets, and bandages,” quips Daniel Pargman, a senior lecturer at the Royal Institute of Technology in Sweden who has written about potential post-collapse computing.

But more gradual collapse means preparing for life beyond eating canned goods, fending off marauders, and treating wounds. “It is less The Walking Dead than it might seem,” Tomlinson says.

One possibility, says Esther Jang, a graduate student in information and communications technologies for development at the University of Washington, is that people may still have functioning, familiar devices such as laptops and smartphones. They could also have localized, small-scale electrical grids or generators to plug them into, but not necessarily stable connections to the internet or a stable power grid. That could mean photos and records stored in cloud systems are sporadically or permanently unavailable, but people can still store some data locally and communicate across more local or regional networks.

“There are lots of ways to still use the devices that people have without the larger infrastructure, but it would look very different,” Jang says.

Resilient technology for a disaster-stricken world

Much of what these researchers project is purely speculative, and there’s no way to know if any of their predictions for the ways we might use tech during an apocalypse are accurate. But there are real reasons to study post-collapse technology. Thinking ahead about a world where the internet and modern conveniences are no longer a steady reality can also help prepare for short-term disasters such as fires, floods, and earthquakes that can knock out connectivity. Puerto Rico saw internet and telephone cables, as well as electricity, knocked out for months after Maria hit, with many residents only able to receive communications via AM radio broadcasts.

In a long-term disaster, people would likely also begin to run physical cables from place to place and communicate with handheld radios, says Matt Johnson, another UW grad student, who coauthored a 2017 paper with Jang and others on post-collapse computing. Ham radio is already a common tool used to keep emergency responders and ordinary people in touch when newer communications networks fail. Some devices would also be able to communicate with so-called wireless mesh networks, which operate when devices pass messages to each other instead of relying on a central internet provider. Those have been used successfully after networks were knocked out in parts of the Bahamas this year by Hurricane Dorian and after Hurricane Sandy struck the New York area in 2012.

But keeping equipment up and running for long would still be a challenge: Less stable sources of electrical power would take their toll on hardware, and software could actually fail first, according to the UW paper. Data, including program code, could get corrupted as decaying hardware stores it incorrectly, causing errors when apps are run. Malware could be spread by whatever networks and disks were still in place, and antivirus software could be hard to come by. Cloud-based tools and the data stored in them could be lost permanently. Other software that relies on talking periodically to a central server, such as design software that checks to make sure you’re still paying for a subscription, could fail when connectivity falls away.

“Fake licenses would proliferate—it’s not a problem that’s insurmountable,” Jang says. “Potentially, open-source software might become a little more popular.”

Hardware that’s more easily repairable would likely become more valuable, too. Unfortunately for would-be digital preppers, today’s manufacturers often sacrifice modularity and easy replacement of broken parts to optimize price or keep devices small and light.

“Consumer technology is not designed for collapse,” Johnson says. While it’s possible that disaster prep could lead hardware makers to start focusing more on resilient tech, just as architects and planners have started designing buildings and cities to be resilient in the face of climate change, that hasn’t happened yet.

And while some components such as solid-state drives and solar panels could last for years, the UW researchers estimate in their paper that other parts such as computer fans, batteries, and traditional hard drives could fail and require replacement more quickly, assuming parts and expertise were available.

“Just as important [as spare parts] for successful repairs would be human resources with the skills needed to perform them, such as soldering and use of a multimeter,” they write. “Without intentional teaching and community retention of these skills even in a generation of less computing ubiquity than we have today, the skills could be lost to many communities.”

Building a collapse-proof operating system

An open-source project called Collapse OS, led by Quebec programmer Virgil Dupras, aims to broaden the range of computing devices that could be usefully scavenged in a post-collapse scenario. It’s a simple operating system that can run on basic hardware, including old Sega video game systems, in order to do simple tasks such as reading and writing basic text files and accessing data stored on various old disks and drives.

“With flaky power sources, electronic components will be under more stress and will probably have a shorter life than under present conditions,” Dupras wrote in an email. “This makes those old machines great because they have a simple design and are very easy to repair, much more than modern machines.”

Collapse OS, Dupras argues, can also run on even more primitive computer chips: so-called microcontrollers, which could be scavenged from all kinds of electronic junk. They could prove invaluable for controlling things such as irrigation, power supply, security, and water purification systems.

Whether or not the project proves to be useful, Dupras writes on its home page, depends on what type of societal disasters humanity may be facing.

“This project is only relevant if the collapse is of a specific magnitude,” he writes. “A weak-enough collapse and it’s useless . . . a big enough collapse and it’s even more useless (who needs microcontrollers when you’re running away from cannibals).”

But for a long-term collapse, short of complete devastation, or just a future marked by more conflict, storms, and other disasters, resilient technology can help ensure people still have access to the digital tools and data that we have all come to rely on.