How To Build A Better Internet

How To Build A Better Internet

Living in the US as I do, I pay Time Warner Cable every month for broadband that works 70 percent of the time. Just for fun, I recently upgraded to the company’s best, fastest service — 300 Mbps — and guess what. After weeks of speed tests, I can confidently say that I don’t get those speeds. Ever.

This is all too common in the United States, the country that invented the internet. It’s not that Time Warner Cable’s customer service is bad. (It really is, though.) It’s not our pipes can’t handle higher speeds. (They can.) The reason I ended up getting stuck with slower internet for a higher price is simple: I didn’t have a choice.

I’m in Brooklyn, New York, just three miles away from America’s internet is historically bad and slow. As Bernie Sanders recently pointed out, a shocking 63 per cent of Americans have only one choice for a high-speed broadband provider. This isn’t how the internet was supposed to work.

But it doesn’t have to be like that. Around the country, there are new internet service providers (ISPs) stepping up and local governments stepping in to ensure more people get access to better, faster internet at lower prices. And creating a better internet really is as simple as that: more reliable connections with higher speeds for less money. This is a story about some of the innovations that might allow it to happen.

A Brief History of Internet Architecture

If you drew a picture of the internet on December 5, 1969, the day it was born, it would look like a little bit like a house turned on its side. The four corners of the house were the original nodes which spanned the American West. Over four decades later, the internet hardly looks like a house any more. It’s an infinitely complex metropolis, built of genius and light.

How this incredible piece of infrastructure came to be is a very long story, long enough that some have written entire books on the subject. The U.S. Department of Defence’s Advanced Research Projects Agency (ARPA), laid the groundwork for the internet as we know it with its ARPANET. In the beginning the internet was a tool for the Pentagon as well as research institutions. Only over the decades that followed this genesis did hooking regular people up to it become a business — and a very lucrative one at that.

How To Build A Better Internet

Following the success of discussion networks like Usenet, several technology companies began offering online services in the ’80s. America Online started as a service for Commodore computers called Quantum Link, or Q-Link. In 1989, the first internet service provider, The World, offered dial up to the public for the first time, and things really opened up.

The next two decades of technology history would be defined by the internet. Yet there are still frightfully few companies that sell internet access to the general public. Comcast and Time Warner Cable are effectively natural monopolies, with little incentive to improve their service. As such, America’s internet will improve if and only if we find a way to enable more competition between ISPs.

There are a few ways that might happen. One option involves a bold company building a new internet infrastructure all at once. The other, more optimistic option involves enlisting cities to build a patchwork of new or improved infrastructure. The third is a hybrid model, something that hasn’t really been attempted on any sort of scale. It involves cities working closely with internet carriers to guarantee better service to everyone.

That’s the ultimate goal. Now let’s talk about how we might achieve it.

The Internet Is Out There, If You Can Reach It

By the 1990s, with millions of people on the internet, it became apparent to network engineers that the existing infrastructure of copper telephone cables would soon buckle under the weight of ever-increasing internet traffic. Some cable companies, like Comcast, realised that the relatively larger bandwidth of coaxial cables could bring faster speeds to consumers. Others, like Level 3 Communications, realised that building the best networks would require the very best technology. So in 1998, Level 3 started laying fibre optic cable all across the U.S. — lots of it.

“We got started with the premise of: If you built a telecom company from scratch, what’s the basic infrastructure how would you build it?” Level 3’s chief technology officer Jack Waters told me in an interview. “In 30 months, we built 16,500 miles on a combination of the railroad right of ways, state and interstate high way and local municipalities.”

The bet paid off. Level 3 now operates a Tier 1 network and the third largest fibre optic network in the U.S. based on area, one that participates in the internet backbone. (A Tier 1 network can send traffic to any part of the internet, and since there are only a handful of them in the world, Level 3 plays a key role in making sure the everything works.) Unlike other massive networks like Verizon that carry traffic directly to consumers, however, Level 3 specialises in connecting businesses and has chosen not to be a last-mile provider. This has made the company a key player in controversial interconnection agreements between content providers and companies that don’t operate Tier 1 networks, like Comcast. In fact, Level 3 was caught in the middle of the controversy when Comcast strong-armed Netflix into paying a fee so that its customers could watch movies without stuttering or buffering. In other words, Netflix had to pay Comcast to ensure that its network capable could handle traffic — which is basically Comcast’s number one job.

Those charges trickle down to the consumer — which is really shitty, as so many Americans have no choice over their ISP. Waters, for one, would rather that ensure consumers can choose whom they pay for internet service. “If we believe, in the long run, that internet infrastructure is becoming more and more critical to the world economy, then it’s gotta be a fair place to compete and do business,” says Waters. “The technology is awesome.”

The technology is awesome, but the vast majority of Americans don’t see the benefits in their home connections. The U.S. isn’t even amongst the top ten countries in terms of average connection speed, according to Akamai’s latest State of the Internet report. The U.S. is in 12th place with an average of 12-megabits-per-second, just a tiny bit faster than Romania. Despite being home to advanced networks like Level 3, the average American suffers from last-mile providers who just can’t seem to upgrade their infrastructure. Why would they when many of them don’t have any competitors?

How you ensure a fair, competitive marketplace is a matter of debate, however. The Federal Communications Commission (FCC) is doing its part by putting strong open internet rules into place. The market itself is also playing a role as consumers demand faster, cheaper internet service, and companies like Google are stepping up with fibre services of their own. Level 3 has strived to build a future-proof infrastructure that could serve their clients and, ideally, the end user for decades to come. When the monopolies that control the last mile of infrastructure refuse to upgrade their technology, though, consumers suffer from spotty service and slow speeds.

A City’s Successful Gamble

For all the behind-the-scenes work Level 3 is doing, someone still has to fight for the end user. Somebody has to fix the last mile.

Some cities are simply taking things into their own hands and building their own networks, treating the internet more like a public utility. With the support of President Obama, this municipal internet approach might just be the best path forward. Take Chattanooga: This medium-sized city in East Tennessee was once a railroad and manufacturing hub, but suffered in the 1980s, when it de-industrialized. The next three decades saw various revitalization efforts, one of which included upgrading the city’s power grid to run on a fibre optic network.

In the mid-Aughts, the city’s Electric Power Board (EPB) started strategising how it would upgrade to a smartgrid and also bring more business development to the city. Like Level 3 before it, the city realised that fibre was the future of infrastructure, and the more advanced technology would perform better than any alternative for decades. Furthermore, the smartgrid would enable the city to deliver energy more efficiently, since it had much exponentially more data points. In the event of a major storm, for instance, the power company could see exactly what caused the outage in a matter of seconds and make the repairs in hours rather than days.

The EPB also realised that because fibre optic cables can carry such vast bandwidth, Chattanooga could fix the last mile problem and start offering its citizens high-speed internet — like a gigabit-per-second fast. The proper equipment was already being installed across the entire city. Why not offer internet service?

“The simple answer is because we can,” EPB chief Harold DePriest told The New York Times not long after Chattanooga rolled out the nation’s first municipally owned gigabit network in 2010. “The overriding consideration is that this is a real tool for economic development for our community.”

How To Build A Better Internet

The city’s gigabit internet wasn’t cheap at first. After all, it cost a whopping $US330 million to build Chattanooga’s smart grid — $US111 million of which came from federal stimulus funds. In the years since the launch, however, the EPB experimented with pricing and now offers gigabit connections at bargain prices. By 2014, the citizens of Chattanooga could get the fastest internet service in the country for less than $US70 a month. (By comparison, 25-megabits-per-second service from Comcast costs about $US70 a month in the area.) At gigabit speeds, you can download a two-hour-long high definition movie in a little over 30 seconds.

Chattanooga’s insanely fast, cheap internet doesn’t just help the residents, either. The so-called Gig network also attracted dozens startups. In an interview, Chattanooga mayor Andy Berke told one story about a company called South Tree that digitised old video tapes and could upload higher quality files to the cloud for customers to download because of the Gig. Another company called Quickcue was able to boot strap a service that let people make restaurant reservations via text message. OpenTable bought the company for $US11.5 after just 18 months.

“[The Gig] changed our conceptions of ourselves,” Berke told me. “Six years ago, we just weren’t on the list of tech communities now no matter where you go people are looking at Chattanooga as being ahead of the curve. We have a vibrant startup scene.”

But it wasn’t easy. The city of Chattanooga faced fierce resistance from big cable companies who didn’t want a new competitor in the internet business. Back in 2008, a year before the fibre rollout, Comcast sued the EPB in an effort to prevent the Gig from being built, claiming that the city had funded the project illegally. The case — which DePriest called “a frivolous lawsuit designed to slow us down” — was eventually dismissed. Fast forward to May 2015, and Comcast announced that build its own gigabit network in order to compete with the EPB.

Chattanooga is now trying to expand its gigabit network, but again, it faces opposition from big cable. Earlier this year, the city successfully petitioned the FCC to change state laws that prevented municipal broadband networks from expanding. Now, Chattanooga and the EPB’s Gig serve as models for cities all across the country to build their own high speed networks.

A Compromise Worth Considering

Not every city can do what Chattanooga’s done to address the last mile problem. Building a fibre optic network from scratch costs tens or hundreds of millions of dollars. It’s not just the money either. Political and logistical hurdles complicate the process, even for the most bottomless pockets.

Take Google Fibre, for instance. The private gigabit internet service is now available in three cities nationwide, but Google can’t yet brag about being available in every part of the city, like Chattanooga can. This is due in part to restrictions Google’s faced problems with expansion since it doesn’t have easy access to utility poles. In Austin, Texas, the search engine giant faced fierce resistance from AT&T which controls about 20 per cent of the utility poles in the city. Google needed to pay extra high prices to use the utility poles, AT&T said, because Google was not a telephone or a cable company. (Utility poles are typically owned and controlled either by big telecom companies or local power companies, and you need permission from the pole owners if you want to string up your cables.)

The FCC’s new open internet rules reclassify broadband providers as common carriers under Title II of the Communications Act, a meaningful change that Google said would resolve its utility pole problem. Even then, access to poles remains an on-going battle with big telecom companies, enough so that Google is building its own poles. That costs money, a luxury that many small American cities don’t have.

Some cities and private companies are working together to offer customers a choice for better, cheaper internet service that doesn’t come from a company called Comcast. A wireless service provider called Ting, for instance, is carving out a new business to offer gigabit internet service in partnership with municipalities. This involves buying carriage on larger networks like those run by Level 3 and its competitor Zayo, and then connecting cities directly to those networks. Finally, Ting installs fibre and the equipment needed to handle gigabit connections directly in people’s homes. In essence, Ting wants to help cities and their citizens plug right into the internet backbone, skipping over the slow coaxial cable that still powers the last mile in many of the big telecom networks.

For now, Ting’s footprint is pretty small. The startup ISP is starting in Charlottesville, Virginia and Westminster, Maryland, about an hour north of Baltimore. Ting is partnering with cities as well as larger networks in order to bring fibre speeds directly to homes and businesses. The key for success here is the ability to streamline the challenge of building out the fibre optic network in a given city.

“You tend to be able to build much easier much faster if the city is positive about what you’re doing and helpful,” Elliot Noss, the chief executive of Tucows, told me in an interview. Tucos is a Canadian telecommunications company that started Ting in 2012. Noss added, “Once the network is built inside of the city and once it’s connected to the back bone the rest is about hooking up homes and lighting them up.”

Ting did just that on June 26, when it turned on the fibre optic network in Westminster. The city will now enjoy gigabit speeds and more choice when it comes to picking a provider. Westminster Mayor Kevin Utz boasted about how the new fibre network would attract new businesses and residents.

Noss took a more idealistic angle in his statement. “For too long, people and businesses have had no choice, or at best the illusion of choice, as to who provides them with access,” Noss said in a press release. And the federal government has the stats to back up that claim. A better internet means more reliable connections, faster speeds, and lower prices. Experts and politicians seem to agree that we must have more competition if this is actually going to happen.

Anything But Another Monopoly

This year has been a good year. The FCC passed the strongest open internet rules in history. Comcast’s god awful merger attempt to acquire Time Warner Cable and become an even bigger monster failed. And President Obama laid out a damn sensible plan to increase competition and give Americans more choice over their ISP. He even did so in Cedar Falls, Iowa, yet another city that built its own municipal broadband.

Nevertheless, this is just the beginning. The FCC’s rules still face an assault of litigation from big cable companies that want to keep the status quo. Charter Communications is now trying to buy Time Warner Cable and become the second largest cable company in the country. The election in 2016 also stands to unravel all the hard work done in recent years, especially if some idiot who doesn’t understand how the internet works wins. And inevitably, the fact that individual states have restrictions that make it hard for startup ISPs — from the well-funded Google Fibre to startups like Ting — to enter the market and compete with the monopolies.

For me and my crappy Brooklyn internet, the situation seems a bit hopeless. Sure, Time Warner Cable could turn a corner and start offering me the speeds I’ve been paying for. And maybe, if I’m really lucky, another ISP could edge its way into my neighbourhood and give me another option. Neither seems very likely in the very near future.

If America wants a better internet, it’s going to take work. We’re going to have to dig holes and install new equipment. Local governments will need to help startup ISPs compete with big telecom in a fair way. Companies like Comcast and Time Warner Cable need to be held accountable for selling crappy service for high prices.

After all, the US invented the internet. It should make it better for itself.

Illustration by Jim Cooke / Photos via Flickr, AP

The Cheapest NBN 50 Plans

It’s the most popular NBN speed in Australia for a reason. Here are the cheapest plans available.

At Gizmodo, we independently select and write about stuff we love and think you'll like too. We have affiliate and advertising partnerships, which means we may collect a share of sales or other compensation from the links on this page. BTW – prices are accurate and items in stock at the time of posting.