Internet of Islands
Nasugbu, in the Philippines, is a hub for undersea networks, with two major cable landing stations. Photo: Nicole Starosielski


This story is over 5 years old.

Internet of Islands

Global internet infrastructure has created an imbalance of power, and it's due for a reckoning.

Tahiti, a figure-eight shaped island in the South Pacific with a population of around 180,000, received its first undersea fiberoptic cable in 2009—a simple branch to Hawai'i, where traffic would funnel out to the rest of the world.

As Tahiti's only cable connection, Honotua (a name that means "link" and "backbone," or "far ocean" in Tahitian) brought residents high-speed internet. But because of the immense amount of information that fiberoptic systems can carry, Tahiti was left with far more capacity than its residents could possibly use—not to mention the low level of adoption caused by the high cost of access, $70/month for 512 Kb/s, and $180/month for 8Mb/s. More than five years after installing the system, the two internet service providers use only 7 Gb/s of the possible 640 Gb/s.


Still, the Tahitian government is planning another system, one that would make Tahiti more than just an endpoint for transmissions.

The as-yet-unnamed project would make Tahiti the central node in a transpacific network, linking to South America on one side and China on the other.

A network like the one Tahiti is proposing might destabilize the balance of telecommunications power

"The government of French Polynesia is very seriously studying the possibilities of a new connection of its territory to the most pertinent markets in the Asia Pacific region and Latin America," said Teva Rohfritsch, the minister of economy in charge of the Digital Strategy of French Polynesia.

Left unsaid: if Tahiti could gain entry to the elite club of island telecommunication hubs—Guam, Puerto Rico, and Cyprus, the nodes that connect the modern world's staggering undersea cable network—it would supercharge its global significance.


Fiberoptic cables can span the ocean in a single hop. Direct links connect the United States and the United Kingdom, Portugal and Brazil, and Los Angeles and Chikura, just outside Tokyo. The FLAG and SeaMeWe systems include long, uninterrupted routes between the Middle East and India, from India onward to Southeast Asia, and from Australia up to China, with spurs branching off to countries in between.

Guam is a networked island, with more cables landing on its shores than in many nations around the world.

Although it is possible to directly interconnect continents, our global network of undersea cables is, literally, an island network. More than half of the world's undersea cable landings are located on islands. These include island nations (Japan and Indonesia), populous island cities (Hong Kong and Abu Dhabi) and less-intuitive cable hubs like Long Island (where much of New York's traffic comes ashore).


There are more landing points on the islands of Hong Kong and Taiwan than on the entire mainland of China. Fifteen cable systems connect Singapore to the network; only 13 are linked to the rest of mainland Southeast Asia. Six international cables link to the Italian peninsula, while almost three times this number land on Sicilian shores. Continents often depend on islands for their communication. All of Australia's cable traffic passes through islands, which must remain above water and politically stable, before it reaches any other continent.

Some of these island nodes are mere offshoots for global systems, with a simple branch connecting them to the main line. Others, such as Guam, are major hubs of internet traffic. With its three cable stations and 10 cables, extending to 33 landing points across the Pacific, more signal traffic moves through Guam than through many nations around the world. Compare this, for example, to the number of undersea cables linking out from Argentina (5), Canada (4), or Bangladesh (1).

Every ocean and sea has a number of critical intermediary islands. In the Mediterranean, cables stop on Sicily, Crete, and Cyprus. Puerto Rico and the British Virgin Islands are nodes in the Caribbean, and in the Indian Ocean, islands like Mauritius are stopping points for cables between Africa and India. In the Atlantic, Bermuda and Spain's Canary Islands have long been key hubs for communications traffic.


A significant number of these are hubs in part due to the British empire's investment in a globe-spanning telegraph system during the 19th century. Islands were important to colonial expansion precisely because of their insularity. Physically separated from the mainland, they could be more easily managed, insulated from threatening populations, and secured by military forces.

In the colonial era, the British staked out a global archipelago. Today, small islands are trying to harness the flows of internet traffic across the ocean themselves—yet they almost always have to link up with one global power or another. The goal is not simply to secure a single cable line, but to position themselves as intermediary points for transoceanic flows.


The project the Tahitian government has proposed is a completely innovative one—with a route as original as the cable planned to link London and Tokyo via the Arctic Ocean.

In many ways, this system would both reflect and support growing Chinese investment not only in South America, but in French Polynesia itself. This prospective link would not build on the older colonial regime (the French Government funded the Honotua link), but on Chinese economic power.

As an island hub, Tahiti would share the burden of generating both the traffic and the profit for the cable system. With multiple cables, telecommunications companies might be motivated to interconnect there. One potential plan is to give away the capacity on the existing line in order to lure them in. A second cable could open up new possibilities for tech industries that require redundancy (after all, if the existing system goes out, only a small percentage of the capacity could be backed up via satellite).


Establishing another hub in the Pacific, and one outside of the United States, would be a benefit to the network as a whole, especially given the NSA's monitoring practices and the widely known difficulty of landing on US soil. The global undersea network needs more diverse routes. Otherwise, disasters like the 2006 Hengchun earthquake, piracy off the coast of Vietnam, or the digging of an elderly women in Georgia can simply shut of parts of the network. From a technical perspective, it seems to be a win-win situation.

The view from inside the cable station at Deep Water Bay, Hong Kong.

Despite such benefits, the telecommunications industry and investment banks don't usually support these kinds of projects. They tend to stick with the tried-and-true, which is why undersea networks keep linking to Guam, Puerto Rico, and Cyprus. It is also why the network of undersea cables is one of our most durable technical infrastructures. The cable-supply business has historically been dominated by the British, Americans, French, and for a period, the Japanese: the Chinese have yet to make headway in the construction of transoceanic cables. A network like the one Tahiti is proposing might destabilize the balance of telecommunications power, even more so if the Chinese government chooses to fund it (and the French and Americans choose not to interfere with it).

But, every year entrepreneurs propose new undersea cable projects, and every year, many of these fail. They are posted as prospective links on the cable maps, and subsequently disappear. Innovative projects like this one tend to disappear more often, given that the industry is conservative on the whole.

Over five years ago, the South Pacific Island Network (SPIN) was designed to connect North America and Hawai'i to Australia and New Zealand, by hopping from American Samoa, Samoa, Wallis and Futuna, Fiji, New Caledonia, and Norfolk Island. Although the plan never gained traction with the telecommunications carriers, piece-by-piece links have been built out in the area. This is a promising sign for island networks. Tahiti's officials are right to be conservative in their proclamations, but signs suggest the longstanding balance of power in the global information network may be due for a disruption.

This excerpt was adapted with permission from The Undersea Network (Copyright Duke University Press 2015). Hell or Salt Water is a series on Motherboard about exploring and preserving our oceans. Follow along here.

All photos by Nicole Starosielski