FYI.

This story is over 5 years old.

Tech

Could a Terrorist Hold the Internet for Ransom with a Pair of Scissors?

After the advent of wi-fi, web-enabled cellular networks, and cloud computing, I suppose it's not really a surprise that most of the populace thinks of the Internet as some sort of magic technology where your Google is beamed into your computer from a...

After the advent of wi-fi, web-enabled cellular networks, and cloud computing, I suppose it’s not really a surprise that most of the populace thinks of the Internet as some sort of magic technology where your Google is beamed into your computer from a golden mountaintop near San Francisco. Even during the fight for a free Internet during the SOPA days, the rhetoric employed in protecting the web from corporate-political branding and censoring seems to ignore one hell of a giant elephant in the room: the Internet itself is made of thousands of miles of cables and satellite links, all owned by large telecom companies.

Advertisement

So, even as the SOPA fight and others like it have been successful for web crusaders bent on keeping the Internet the Wild West of information it’s long been, there’s a sinister question lurking: If large companies own the physical Internet, can it ever be as ‘free’ as Web activists want? What happens if they go bankrupt? Or, what about the nuclear option: If some Internet terrorist is hellbent on taking down the ’net, can he take a knife to some important cable and hold us all for ransom?

The handy Wikipedia chart above highlights just how convoluted the route is for information passing from one user to another over the web. The key thing here is how traffic is passed from one network to another. Some networks ‘peer’ with each other; in other words, they provide access to their network for free in return for free access to another network. So, for a hypothetical example, if Company A owns all of the wires and cables creating Asia’s Internet network, it might offer access to that for free to Company B, which owns all the infrastructure in North America. This is indeed how the web works: the networks that only peer with other networks, essentially getting access to all of the Internet for free in return for access to their own network are Tier 1 networks. They’re the grandaddies of the web, controlling the top-level networking that reach all parts of the Internet.

On the other hand, most networks that aren’t large enough to be able to pay for access with access end up paying what’s called ‘transit’. Transit is usually paid for in terms of data volume: you pay your ISP for a certain amount of data (for your cell phone) or bandwidth speed per month, while your ISP pays higher-level networks for access to their system, and so on. Only Tier 1 networks, by virtue of their largesse, are able to access the whole Internet — all of the cables and computer hooked up to them worldwide — for free. (Aside from the massive maintenance and development costs that made them so large in the first place, of course.)

Advertisement
This is a helpful primer if you’re totally confounded by this Internet thing.

So if Tier 1 networks are a major part of the physical backbone of the global Internet, what happens if they go south? According to a well-sourced Wikipedia page, there are 10 companies managing Tier 1 networks, six of which are U.S.-based. Only four of the ten — AT&T, Century Link, Sprint, and Verizon — don’t have a global footprint, according to the companies’ published network maps.

Now, “global” doesn’t mean “everywhere,” but that’s where other networks come in. There are a number of massive, global Tier 2 networks, as well as a wealth of smaller, regional Tier 2 networks. In essence, much of the global web isn’t necessarily redundant, but there are backups. If Level 3 goes under, or somehow someone renders their entire global network useless, there’s likely someone who can take up the slack.

Look at all of those lovely connections.

An absolutely brilliant way to visualize this global network of Internet tubes is TeleGeography’s Submarine Cable Map, which highlights just how many connections there are across continents, not counting satellite links and regular-old landlubbin’ cables. There are also a whole hell of a lot of companies managing those cables, reducing the likelihood that dispute, bankruptcy, or negligence would disconnect large parts of the world — although limited bandwidth could potentially bog things down, as we’ve seen when transcontinental cables get snipped.

But can one simply cut the cables of the global Internet to shut it down? Well, no, not really. There are just too many redundancies. But look again at that cable map. Notice any vulnerabilities? (The Svalbard Undersea Cable System in Norway stuck out to me.) Regional networks, especially those lacking competition, are indeed more likely to crash. This can be caused by anything from simple money-grubbing transit-peering disputes between ISPs to governments like Egypt and Syria throwing the kill-switch, shutting down traffic from every network that connects in and out of the country.

There is no such a thing as a single Internet kill switch. But if the U.S. government wanted to impose some kind of Internet martial law, say in the case of a cyber war, that could impact much of the global network, considering how much of the Internet — infrastructure, storage, companies — is based in the States. But that’s still an extreme case, and only one of the ways the Internet could be compromised. But if you’re planning to destroy the web’s “central mainframe” or slice some important trunk line like a James Bond villain, you don’t understand how the Internet works.

Follow Derek Mead on Twitter.

Connections: