I crossed the border between China and Korea fourteen times. Each crossing required a different route, a different assessment of patrol patterns, a different calculation of risk. The border was not a line. It was a problem, and the problem had solutions, but each solution was temporary, and the cost of getting it wrong was death.
I describe this not because the details of 19th-century border crossing are directly applicable to the present. I describe it because the structure of the problem has not changed. What has changed is the terrain.
The border is no longer only physical. It is digital. And the principles that governed my crossings – redundancy, trust, adaptation, operational security – are the same principles that govern the movement of information in restricted environments today.
The Digital Border
When a government restricts internet access – through firewalls, content filtering, platform bans, or network shutdowns – it is erecting a border. Not between territories, but between its citizens and the world’s information.
The purpose is the same as any physical border designed for control: to determine what enters and what does not. To create an information environment that the state can manage. To isolate the population from the ideas, contacts, and organizing tools that might threaten the state’s authority.
The methods are different. The principles are not. And the countermeasures – the tools and strategies that allow people to move information across digital borders – follow the same logic as the countermeasures I used to move across physical ones.
Three Principles
Principle one: Redundancy. I never relied on a single border crossing route. When one was discovered, I needed alternatives already established. The same principle applies to digital communications in restricted environments. Relying on a single tool – a single VPN, a single messaging platform, a single channel of communication – is a single point of failure. When the state blocks it, you must have another ready.
This means maintaining multiple channels, kept in various states of readiness. Not all active at once – that increases the surface area of detection. But prepared, tested, and known to the people who will need them.
Principle two: Trust precedes technology. The most sophisticated border-crossing route is useless if the people at each end cannot be trusted. I could not cross the border alone. I depended on guides, contacts, safe houses – a human network that predated and enabled the physical route.
The digital equivalent is the trust network that underlies any secure communication system. Encrypted messaging is only as secure as the people using it. A perfect protocol in the hands of a compromised contact is not security. It is a trap with better encryption.
Build the trust network first. The technology comes second.
Principle three: Adapt continuously. The patrols changed their patterns. The routes became known. What worked in March did not work in June. The only constant was that the environment would change, and the network had to change with it.
Digital environments change faster than physical ones. A VPN that works today may be blocked tomorrow. A platform that is secure this month may be compromised next month. The capacity to adapt – to detect that a tool has been compromised and to shift to an alternative quickly and quietly – is not a feature of the network. It is the network.
The Internet Shutdown
The most blunt instrument in the digital border toolkit is the internet shutdown. This is the equivalent of closing the physical border entirely – cutting off all movement, all communication, all contact with the outside.
Shutdowns are increasing in frequency globally. They are deployed during protests, elections, and military operations. Their purpose is not security. Their purpose is control: to prevent the documentation, coordination, and international visibility of events the government does not want observed.
The countermeasures to shutdowns are harder than the countermeasures to filtering. You cannot route around a shut-down network with a VPN. You need alternative infrastructure: mesh networks, satellite links, radio-based data transfer, physical transport of data on portable devices.
These are the modern equivalents of the mountain passes I used when the main routes were blocked. They are slower, more difficult, and more dangerous. But they work, and their existence means that the total shutdown of communication is never truly total, as long as people are willing to do the work of maintaining them.
What I Ask
I ask two things.
Of the people building tools for secure communication: remember that the people who need your tools most urgently are not the ones who can attend your conferences or read your documentation in English. Build for the person who has forty minutes of intermittent connectivity, an outdated device, and a government that will imprison them if your tool fails. That is your user. Design accordingly.
Of the people living in environments where information is free: do not take it for granted. The border that restricts your neighbor’s access to information is maintained, in part, by the indifference of the people whose access is unrestricted. Your freedom of information is not a natural state. It is a political condition, and political conditions can change.
Support the tools, the organizations, and the people who keep digital borders crossable. Not because it is your crisis. Because the network that connects us is only as strong as its most restricted node.
The border is a network problem. And network problems require network solutions.