When most of us use the internet, we don’t tend to think too hard about what’s going on behind the scenes. We tap a button, a video of a cat running around in a sombrero appears, and no further thought is required. But for data like that to reach our phones, it first needs to pass through one or more internet data servers, which are huge data centers located all across the world. In order to connect these servers to one another, and have a truly global internet, a series consisting of hundreds of cables collectively measuring more than 700,000 miles has been laid at the bottom of the world’s oceans and seas. Around 99% of the world’s data passes through these cables, which means pretty much everything you see on the internet, including this very sentence, has passed through at least one of these cables. The data also passes through fiber interexchange points, which allow each of these networks to interconnect directly. Gil Santaliz is the founder and CEO of one such center, New Jersey Fiber Exchange (NJFX), which he describes as “the fiber interexchange point for domestic North American cables as well as European and South American transatlantic cables.” When packets of data are exchanged through NJFX, “they go on to those countries and they find places like ours in Europe and South America, and they go out to the rest of those continents.”

 

The first underwater cable

In the 1830s and ’40s, various inventors developed electric telegraphs, which could transmit messages over long distances at speeds that were previously impossible. Networks of land-based telegraph cables were very quickly built and expanded, but it was still impossible to send messages across large bodies of water. In the 1850s, that all changed. A New York paper manufacturer by the name of Cyrus West Field decided to undertake the challenge of connecting North America and Europe with a giant underwater cable. The plan was to connect Ireland and Newfoundland by having two boats meet in the middle, each loaded with half the length of cable. They would connect the cable in the middle of the Atlantic, and then sail to their respective shores. After multiple failed attempts, leading to high levels of discouragement and cable manufacturing, the first transatlantic cable was finally laid in 1858. The very first message to be sent was, “Laws, Whitehouse received five minutes signal. Coil signals too weak to relay. Try drive slow and regular. I have put intermediate pulley. Reply by coils.” Pretty profound stuff I know, but it was quickly followed by a much more formal exchange between President James Buchanan and Queen Victoria. Although these messages were exchanged considerably faster than had been previously possible, it still took hours for a single message to be completely transmitted.

 

The birth of a network

Three weeks later, after an attempt to improve the signal by increasing the voltage, the cable failed. It wasn’t until 1866 that another cable was successfully laid, and since then we’ve been laying progressively more and better cables across the world’s largest bodies of water. Between 1955 and 1956, the world’s first transatlantic telephone cable system, TAT-1, was laid, allowing a whopping 35 simultaneous telephone calls to be made. When the first communication satellites were launched in the 1960s, they initially took some of the weight off the shoulders of their underwater counterparts. As Santaliz puts it, “In the old days satellite communication did compete with these original copper cables going across the ocean. 40 years ago the technology of uplink satellite and the subsea cables were kind of on the same playing field.” In 1988 the very first fiber-optic cable, TAT-8, was laid across the Atlantic, which metaphorically blew satellite communication out of the water. Nowadays, there is a relatively continuous stream of new cables being laid and old ones being decommissioned, so the exact number of operational cables is constantly changing.

 

Then and now

While a lot has changed in 150 years, modern cables still share some unusual similarities with their predecessors. For example, cables haven’t gotten much thicker, and are only about the width of a garden hose. They also function using many of the same components, just made with different materials. Conductors that used to be copper are now made of glass, insulators now use polyethylene instead of gutta-percha, a natural latex from the plant of the same name, and strength members that used to be made of iron are now steel-based. This means that cables are now built to last for 25 years, as opposed to a few weeks, and currently the world’s fastest cable can transfer up to 26.2 terabits per second. That’s enough to stream 5,000,000 high-definition movies simultaneously.

Germany Mecklenburg-Western Pomerania Baltic Sea - undersea cable to Hiddensee island
Image by Jens Köhler/ullstein bild via Getty Images

Who foots the bill?

Underwater cables can be incredibly expensive, usually costing hundreds of millions of dollars, and someone has to pay for them. “Traditionally, 20 years ago, it was consortium telecom companies. TAT-14 was a consortium of 70 carriers, U.S. and Europe. Everyone chipped in to create a consortium and they all collectively collaborated to get the system built, and they all had an ownership in the system, and they all used the system on the percent ownership they had,” explains Santaliz, but according to him, this is starting to change. “The users are revolting. They’re saying I need control of my own destiny, I want owner economics, and they’re not waiting for the telcos to get together a consortium to replenish a cable, they’re taking their destiny in their own hands and starting these projects and running them.”

 

Sabotage

You might think that with all these cables just lying around, carrying the world’s data to-and-fro, they would be prime targets for sabotage, but you don’t see agents from different countries zipping around in submarines equipped with industrial-grade cable cutters. For one, cutting a single cable doesn’t usually have that much of an effect. The network has become so extensive that in most cases the internet traffic can simply be rerouted through a different set of cables. Additionally, in most cases it would be difficult for a saboteur to cut a certain region’s internet without also having an adverse effect on themselves. On the other hand, tapping these cables as a form of international espionage is pretty much par for the course for most countries.

 

Accidents happen

While they’re rarely tampered with on purpose, underwater cables do get damaged and need to be repaired on a regular basis. Trawling fishing vessels, anchors, and natural disasters such as rockslides and earthquakes all have the potential to cause significant damage to underwater cables. There have even been occasional cases of sharks biting the cables hard enough to cause a disruption. If a cable needs to be repaired, ships built specifically for that purpose are sent out to fix it. These ships are crewed by a large team of specialists and have a giant remotely operated vehicle on board that can be lowered to the seafloor in order to retrieve the broken cable.

 

The future of the internet

Although not wanting to make any specific predictions himself, Santaliz did outline where some people believe the industry is heading. “Some theorize that you could see bundling one day of services. You could see companies that will say when you use my cloud services it comes with the connectivity that you need. You could see someone that goes out to a large enterprise and says what do you guys want to accomplish. They’ll say we want to have all our employees connected on a platform. They will use these assets to deploy that connectivity globally. And they will be in a better position than a traditional carrier’s ever been in before. I’m not sure if I think that’s inevitable, but I think it’s a definite possibility.”