Search This Blog

Friday, January 16, 2015

Where Cellular Networks Don’t Exist, People Are Building Their Own

As reported by Wired: Inside the cloud that is perpetually draped over the small town of San Juan Yaee, Oaxaca, Raúl Hernández Santiago crouches down on the roof of the town hall and starts drilling. Men wearing rain gear of various impermeabilities cluster above him, holding a 4-meter-tall tower in place. Braided wires trail from four small circles welded near its midpoint; eventually those will be bolted or tied down in order to hold the tower steady during the frequent storms that roll through this part of Mexico’s Sierra Juárez mountains.

They don’t want it falling over every time it rains. Ninety thousand of the town’s pesos—a bit over $6,000—are invested in the equipment lashed to the top of the tower, in a town where many residents get by on subsistence agriculture.

Raúl Hernandez and Peter Bloom.
The tower—which Hernández, Yaee’s blacksmith, welded
together out of scrap metal just a few hours earlier—is
the backbone of Yaee’s first cellular network.
 The 90,000 pesos come in the form of two antennas and an open-source base station from a Canadian company called NuRAN. Once Hernández and company get the tower installed and the network online, Yaee’s 500 citizens will, for the first time, be able to make cell phone calls from home, and for cheaper rates than almost anywhere else in Mexico.

Rhizomatica’s Peter Bloom helped make sure the tower could support a base station and an antenna.  Strategically ignored by Mexico’s major telecoms, Yaee is putting itself on the mobile communications grid with the help of a Oaxaca-based telecommunications non-profit called Rhizomatica. Its founder, Peter Bloom, is among the men currently getting soaked on the roof of town hall. It’s May of 2014, and this is the third of what he jokingly calls “artisanal cell phone installations” that he’s led in the Sierra Juárez in the past year and a half—the first of their kind in the world.

By the end of the year, he will have installed six more networks all over the state of Oaxaca, bringing the total to nine. Armed with an experimental concession from the Mexican government that grants Rhizomatica access to coveted cellular spectrum all over the country, Bloom is slowly but surely bringing coverage to towns that have been left out of the 21st century’s most important technological revolution.

Too small for profit

Of the world’s 7 billion or so cell phones, a few hundred of them are already in Yaee—they’re just not connected to a network. Kids use them as cameras and mp3 players, and Hernández, like many adults, bought his to use in Oaxaca City, a seven-hour bus ride away.

When he’s there, his cell phone can connect to plenty of base stations, which, in turn, link him to his choice of commercial network. But back in Yaee, there are no base stations and therefore no network. Every time Hernández wants to make a call in his hometown, he hikes for 20 minutes to the top of the highest hill around and hopes to catch some signal trickling in from a faraway base station, installed in a place deemed more profitable for telecoms than small towns like Yaee.
Raúl Hernandez.
Raúl Hernández, el herrero de Yaee, construyó en su taller la torre hecha de chatarra.  Lizzie Wade

We’ve all heard plenty of uplifting stories of the democratizing potential of cell phones, how they’ve brought everything from voice calls to mobile banking to people who have never had access to landlines and laptops. Cell phones “have definitely proven the most ubiquitous piece of communication and digital hardware that people own on earth,” says Bloom. But on its own, “your cell phone doesn’t really know how to do anything,” he explains. All of the utility is in the network. And by and large, that network is provided—and, therefore, controlled—by a company that wants to make a profit.

That profit comes from subscribers, and if there aren’t enough of them in a particular region, cellular providers simply refuse to install their infrastructure there. Some countries get around that economic reality by legally requiring telecom companies to build networks in rural areas, no matter how many people end up paying for a contract. Mexico doesn’t have any such laws, meaning that Yaee, with its 500 residents, doesn’t stand a chance of attracting a commercial provider.

To make things worse, Mexico’s telecom industry is largely controlled by Telmex, a near-monopoly run by Carlos Slim. Ever since a supposed reform in the late 1980s transferred the country’s state telecom into Slim’s hands, Mexicans have paid first world rates for third world service—first for landlines, and now for cell service and internet access. And that’s when they live in a place with a network. Limited access and high prices meant that only 55 percent of Mexicans were using cell phones in 2011, according to the International Telecommunication Union.

Despite Mexico’s reputation for horrendous, Slim-driven telecom service and policy, it’s far from the only country that struggles with providing rural cell phone access. According to the GSM (for Global System for Mobile Communications, the standard technology behind a 2G network) Association, a consortium of commercial mobile providers from all over the world, 1.6 billion people in rural parts of developing countries don’t have access to mobile networks. That’s why Bloom and his collaborators at Rhizomatica say that if you really want to make the benefits of cell phones available to the people who need them most, it’s not enough to democratize the hardware by making the phones themselves super cheap. You have to democratize the infrastructure, the network itself. And that’s a lot harder to do.

Democratizing technology

From a hacker’s point of view, mobile communications came along at just the wrong time. The first commercial systems were deployed in 1991, right before the internet emerged from the academy and started making its way into people’s homes. By the time a strong open source community came into being, cellular networks were locked up behind walls upon walls of patents and proprietary equipment. Even today, “it’s very difficult to get your hands on the technology,” says Harald Welte, an open and free source software developer in Germany who works on mobile communications.

It wasn’t until around 2006 that old base stations started showing up on eBay, giving interested hackers like Welte a firsthand look inside the (albeit already outdated) technology that made 2G mobile networks possible. Out of straightforward intellectual curiosity, Welte snapped up a few and, four years later, he was able to make the first call on his reverse-engineered, open source network, dubbed Open BSC, referring to the base station controllers that coordinate traffic on a cell network.

Now, Rhizomatica is pushing Open BSC to its limits out in the real world. “We’re amongst the first people actually putting it into a live environment and using the shit out of it,” Bloom says. He first got interested in community cell networks when he was living in Nigeria and working with communities that were protesting the presence of oil companies in the Niger Delta. The activists there had cell phones, but thanks to the high cost of service—not to mention political forces that could monitor their communications or even shut down their network on a whim—it was difficult for them to share information with each other or with larger audiences. So Bloom decided to help them build what’s called a mobile mesh network, which connects cell phones directly to each other instead of routing calls through base stations or commercial networks. But the technology, which is mainly used in disaster relief situations like post-earthquake Haiti, proved to be too unreliable for everyday use. Sustained, real-world levels of traffic overloaded them, and the mesh networks frequently collapsed.

When a local man suffered several serious snakebites and needed antivenom right away, there was no signal.

A few years later, Bloom moved to Mexico to be with his now wife, who works with community radio stations in the Sierra Juárez. These villages wanted but couldn’t afford commercial cell service, and Bloom started thinking about a way to continue the project he had started in Nigeria. He decided to ditch the mesh network idea and went on the hunt for serious telecom technology, ultimately settling on Welte’s Open BSC as the strongest open source system. But since Bloom’s background isn’t in programming, he needed help installing the software on open source base stations he buys from NuRAN and another company called Fairwaves. He started enlisting the help of any experienced hacker who happened to pass through Oaxaca City on backpacking trips and the like. (One of them, an Italian, would eventually move to Mexico permanently to be part of the Rhizomatica team.) Today, Bloom spends much of his time personally driving the equipment out to villages like Yaee, getting soaked on as many roofs as he needs to in order to get the networks up and running.

The communities pay 120,000 pesos ($8,000 dollars) upfront for the equipment and installation, about one-sixth of what the commercial provider Movistar charges for a similar rural installation. Ninety thousand of the pesos go to buy the hardware, and the rest covers Rhizomatica’s time and expenses. Subscribers to the community network pay 30 pesos (about $2) per month for all local calls and texts, and the town keeps any profit left over after paying for electricity and maintenance. Thanks to a Mexican company called Protokol, which provides internet access all over rural Oaxaca, Rhizomatica can also hook up the town’s network to a voice-over-IP connection, which allows users to make very cheap long-distance calls to Mexico City and even the US, where many people have relatives. Once the network is installed, Yaee’s residents will be able to call the U.S. for 20 centavos (less than 2 pennies) per minute. A similar call from one of the town’s public landlines runs 15 pesos (about $1) per minute, a prohibitive cost for many residents.

Still, commercial networks have “20 years of headway” over the open source approach, Welte says, and Rhizomatica’s community networks can suffer from their distinct DIY feel. Bloom, Hernández, and the rest of the team must make sure to install Yaee’s tower above one of the town hall’s windows, so they can run an extension cord through it and plug the base station into a wall socket. That means whenever the power goes out in Yaee—which happens frequently, especially during the May-to-September rainy season—they lose the cell network, too. And until the town could raise enough money to move the entire installation to higher ground than the town hall’s roof (which happened in August 2014, three months later), there was no guarantee that Rhizomatica’s signal would be able to reach up the hillside to where Hernández and a good portion of Yaee’s residents live.

It’s this fundamental instability that causes the most frustration for users in towns that have had their community networks up and running for longer. In Yaviche, a similar sized town on the other side of the mountain from Yaee that installed its local network in September of 2013, Abi Martínez Ramos serves as the rural doctor and says that having any cell service at all has been a boon for emergency medicine. But when a local man suffered several serious snakebites and needed antivenom right away, “there was no signal,” he remembers. Someone had to physically find Martínez to administer treatment, just like in the old days.
Back in Yaee, it takes the team about 2 hours out in the rain to anchor the tower to the roof.

Raúl Hernández Santiago and Peter Bloom helped
install the first cell phone tower in the town
of San Juan Yaee, Oaxaca.
But it turns out the rain has caused a more serious problem than merely soaking everyone to the bone. Three cloudy days in a row have depleted the solar panels that power Protokol’s repeater antennas, knocking out Yaee’s internet access. With no internet, Bloom and his tech team can’t get the network online. They make plans to come back the following week to finish the job.

Despite its problems, an increasing number of communities in Oaxaca are eager to be part of Rhizomatica’s experiment, attracted by the low price and the promise of complete control over their networks. Keyla Mesulemeth Ramírez, who helps run the community network in Talea de Castro, a town of 2,000 that volunteered to be a Rhizomatica pilot project in the spring of 2013, fields one such inquiry in her office the day before the Yaee installation. Four men from a town called Yalahui have heard Bloom is in the area and they want to talk to him about installing a network in their village. They grow coffee, sugar cane, corn, and beans, and they’re tired of not being able to call home when they’re out in the fields. It’s annoying not to be able to call when you’ve forgotten your lunch, they say, but it’s downright dangerous when someone has an accident and needs help. Plus, everyone in Yalahui has family in Mexico City and the US, and they want to be able to call them without worrying about how much it costs. They’ve driven for 5 hours to talk to Bloom about a possible solution.

Mesulemeth is frank with the Yalahui men about the cost of the installation and Rhizomatica’s waiting list. But she’s sympathetic to their frustration of being left off the cellular grid. “Before, cell phone service was a luxury,” she tells them. “Now it’s a necessity.” She promises to put them in touch with Peter, who finds the group lingering over lunch a few hours later. He goes over the costs again, and says he’ll get out to Yalahui as soon as he can. Three months later, they’re making calls on a brand new network all their own.