The Trump administration’s recent decision to kill the Open Internet Order has a lot of net neutrality advocates fearing the worst. Foremost among those concerns: that last month’s Federal Communications Commission vote might embolden broadband providers to manipulate how their customers access and use the internet. Although it remains to be seen whether that will happen, a small but growing number of users are taking matters into their own hands by exploring community- or municipal-owned, operated and funded internet access as a cheaper, faster and more neutral alternative to using large commercial internet service providers.

NYC Mesh, which typically gets 20 requests per month for membership to its community-based network in New York City, has received connection inquiries from more than 200 people since the FCC’s December 14 vote, says Brian Hall, a software consultant who helps manage the network. Self-organized, community-run systems including this one and Detroit’s The Equitable Internet Initiative have for years offered alternatives to corporate ISPs such as Comcast and Time Warner Cable.

The community approach typically involves organizers renting internet access from a local data center, and installing rooftop antennas and wi-fi routers that together act as access point for nearby residents. Unlike a home or office router that provides wi-fi service for a dozen or so square meters, a community network can provide a wi-fi signal for several square kilometers. Residents connect to the access point by mounting their own antennas on their buildings’ rooftops or outside their windows. These antennas receive the network’s signal and send it through a cable to wi-fi routers located inside members’ homes or offices. The setup is called a “mesh” network because any member can act as an access point, or node, for other members.

“People are so used to ISPs that they think that’s the only way to access the internet,” Hall says, adding ISPs have built their businesses by acting as gatekeepers between customers and the internet in order to extract as much money as possible. NYC Mesh pays about $1,000 per month to use the data center but does not charge members for access to the mesh network. The organization currently uses a grant from the Internet Society New York Chapter to pay the rent, and suggests members donate $20 per month to help cover ongoing rental and management costs.

NYC Mesh has 70 wi-fi router nodes, many of which are connected to two “supernodes” that provide internet access at about 100 megabits per second, or Mbps—more than twice as fast as the average commercial ISP connection—throughout Lower Manhattan and parts of Brooklyn. The original supernode is run out of a Lower Manhattan data center. In that setup the group’s antenna is mounted on the roof and their computer servers are a few floors below in what is known as the “radio room,” Hall says. The servers run software that manages the network. They are also linked via optical fiber cables to a “meet-me room”—the location in a data center where networks interconnect directly with the internet backbone, without needing an ISP to serve as an intermediary.

To access one of the supernodes, a resident needs to get a wi-fi router node and install an antenna—either on the roof of their building or in a window with an unobstructed line of sight to the supernode’s transmitting antennas. Members whose view of the antennas is blocked by a building or bridge can connect to other NYC Mesh members’ nodes. Residents can also access the network using the public LinkNYC kiosks that the city has installed throughout Manhattan and Brooklyn. “We’re basically a wi-fi network,” Hall says.

Hall and others involved in NYC Mesh have recently been meeting to figure out how they can scale up the network to meet rising demand. One option is renting more antennas at supernode sites, and installing more large antennas atop tall buildings spread throughout the city and surrounding area. He estimates it would take between 20 and 50 supernodes to cover the city’s five boroughs.

Hours after the FCC vote Motherboard—an online magazine that Vice Media launched in 2009—announced plans to set up and operate its own community mesh network in Brooklyn, and to publish a how-to guide for others interested in doing the same thing in their own areas. The project is still in the early planning stages. “A lot of nonprofits and community groups and even just meet-up groups made of private citizens have built networks like this across the country,” Motherboard Editor in Chief Jason Koebler says. “Every time we write about one of these community groups, people ask, ‘What are the actual logistics of making something like this happen?’ We don’t have a good answer for them often, so we thought we’d start from scratch.”

PUBLIC BROADBAND

Another approach, called municipal broadband or public broadband, offers residents internet access via a network supported by their city or town government—often with outside help from tech companies or public utilities. This allows local governments to provide the service at low prices—or for free. Starting in 2010 Google’s Fiber division began offering broadband service directly to consumers. Within a few years the company was delivering high-quality internet access to nine U.S. locations—including Kansas City, Austin and Atlanta. Unfortunately, a reorganization at Alphabet (Google’s parent company) prompted the company to announce in October 2016 it would not extend its Fiber services beyond those locations it already served, plus three locations where fiber installation was already in progress and completed in 2017.*

The thinking behind publicly owned networks is that municipal governments have greater incentive than privately owned ISPs (which have little competition) to provide citizens, government agencies and local businesses with free or affordable high-speed internet access. Large ISPs, in an attempt to keep competition at bay, have lobbied hard—and with some success—to block or limit municipalities from offering broadband services. Several states, including Tennessee and Colorado, now have laws banning municipally run networks. Despite those restrictions, the city council in Fort Collins, Colo., voted on January 2 to move ahead with plans to build a public broadband network. In November an additional 19 cities and counties throughout the state approved financing to study the feasibility of their own municipal broadband projects.

The FCC’s refusal to enforce net neutrality will “catalyze” more city hall conversations about the best way to deliver broadband to both citizens and businesses, says Lev Gonick, Arizona State University’s chief information officer and former CEO of the nonprofit broadband service provider OneCommunity. “At a minimum, [municipal broadband] efforts put incumbent providers on notice that a city is looking for a competitive marketplace,” Gonick says.

Municipalities can fund the set up and management of public internet service in a few ways, such as issuing bonds or collaborating with local broadband providers. Publicly owned broadband networks have a mixed track record, however. Some municipal networks are owned by public utilities, such as Iowa’s Cedar Falls Utilities or the Electric Power Board of Chattanooga, Tenn. These fiber networks deliver citywide gigabit-per-second speeds (a gigabit is one billion bits) that far surpass even the best ISP service in the U.S. The country’s largest commercial ISPs offer average fixed (as opposed to mobile) broadband download speeds of about 64 Mbps whereas the average upload speed is under 23 Mbps, according to a report published last year by internet speed test company Ookla. (The FCC has for the past several years defined broadband as connection speeds of at least 25 Mbps for downloads and three Mbps for uploads.) There are, however, also many examples of failed municipal broadband projects, including iProvo in Utah and the City of Monticello in Minnesota. Mismanagement and high costs doomed both projects.

Whether community- or municipal-run networks can handle a potential influx of tens of millions of disenchanted ISP subscribers is an open question. “It’s not appropriate at this time to say that we have the solution to the problem that the FCC’s decision has handed us,” says Nathan Schneider, a media studies scholar in residence at the University of Colorado Boulder. Still, these localized options are important at the very least for the pressure they put on ISPs to deliver good service at a reasonable price, Schneider says. “The fundamental problem with the situation that we’re in now is that [many] broadband providers are monopolies, but are not being recognized as such by the government,” he says. “If we’re going to do away with the principle of [net neutrality], on whose terms do we want to do that—a big corporation or a company owned and operated by its customers?”

*Correction (01/12/18): This paragraph was edited after posting. It originally referred to Google Fiber as a municipal-broadband subsidy program and referenced only nine of Google Fiber’s 12 locations in the 

Excerpt from Can We Build a Better Facebook

a perceived opportunity to build a better Facebook: one that shows more respect for user data and privacy, and that avoids the perils of manipulative algorithms and fake news.

MeWe, the platform Sifry wrote about in his recent post, offers a “Privacy Bill of Rights” that includes stipulations like “You own your personal information & content,” “You will never receive a targeted advertisement or 3rd party content based on what you do or say online,” and “We do not manipulate, filter, or change the order of your content or what you see.” Ello, which launched itself as the anti-Facebook a few years ago, similarly stresses its privacy policy: “Ello does not make money from selling advertising on the site, serving ads to you, or selling information about our users to third parties, including advertisers, data brokers, search engines, or anyone else.”  Path is a social network that initially handled the noise and privacy problems by limiting the size of each user’s network: as the launch post put it, “Because your personal network is limited to your 50 closest friends and family, you can always trust that you can post any moment, no matter how personal.”

But all these would-be Facebook competitors face a fundamental problem: a social network is only as useful as the people who are on it. In “Information, Innovation, and Competition Policy for the Internet,” Howard A. Shelanski digs into the dynamics that make it hard for new upstarts to compete with established social networks, concluding that “each user creates a positive externality for other users by enlarging the scope of the service.” In other words, any Facebook competitor faces an uphill battle: with so many people already on Facebook, it’s extremely hard to offer a compelling alternative.

When you look at the Facebook alternatives that have been around for a while, you can see that’s exactly the problem they’ve run into. With its vision of small-scale networks, Path baked growth limitations right into its business model; no surprise that it raised (and then removed) its limits on network size, before eventually selling the platform. Ello initially positioned itself as a network for anyone who doesn’t like Facebook’s privacy policy, then settled into a much narrower mandate as a platform for artists. MeWe is apparently trying to address the growth challenge with a sign-up workflow that tricks new users into inviting everyone they know to sign up; I’m not alonein noting the irony of joining a privacy-oriented social network, only to have my first experience consist of spamming every single one of my email contacts with a MeWe invite. (Sorry, friends!)

In a discussion of Micah Sifry’s piece that unfolded on—you guessed it!—Facebook, tech writer Amy Vernon argued that the only way an emergent social network can get over the scaling problem is if it’s substantially new:

Any platform based on what we already have with Facebook (even if it’s supposedly privacy-focused) is doomed to fail. Facebook succeeded and surpassed MySpace because it was a different animal. MySpace took over from Friendster because it was different….In order for us to give up Facebook, something different needs to rise up. Thing is, every time they do, Facebook buys it or copies/buries it. Instagram, WhatsApp, Snapchat (tried to buy it, arguably has decreased its influence through copying its features in Instagram). Everyone needs to stop trying to be the new Facebook, and instead build something that people want to use.

The problem startups can’t fix

If the pace of scaling and expectations for innovation make it hard for a new startup to compete with Facebook, maybe that’s not such a bad thing. After all, most of Facebook’s would-be competitors are not unlike Facebook itself: they’re for-profit companies, ultimately accountable to investors or (if they go public) to shareholders. As Jacob Hasler puts it in a a 2014 article, “[w]ithout clear authority to pursue a social mission, directors are likely to protect themselves by maximizing shareholder value.” That means they’re under huge pressure to generate revenue, or at least, to present a business model and growth pattern that promise not-too-distant riches.

That creates a strong intrinsic pressure to collect and sell user data, and/or to use that data as the basis for a thriving ad business. If any of those networks actually attracted as many users as Facebook has, it would also find itself in the position of overwhelming its users with the sheer volume of posts: it would need some kind of mechanism—an algorithm, say—to determine who sees what. That algorithm would also open up another revenue opportunity: charging companies money in order to make their content more visible.

Does this sound familiar yet? Yeah, I’m pretty much describing Facebook as it exists today.

That doesn’t mean I’ll let the platform off the hook for fostering a business culture that consistently places dollars ahead of its users’ privacy and well-being, but I do recognize the market pressures that drive that kind of culture. Other for-profit startups may talk a good game—now, while they’re small—but I’m skeptical of their ability to resist the same pressures should they ever get large enough to pose a meaningful challenge to Facebook itself.

A kinder kind of startup?

My concerns are exactly why some companies—Ello among them—have incorporated as Public Benefit Corporations (PBCs). The PBC is a relatively new corporate structure that require companies to have some kind of public benefit mission alongside the profit motive: as Hasler notes, “[h]aving a specific public benefit allows the corporation to pursue a personally selected mission, which could be anything.” Ello’s own website says that it incorporated as a PBC to “assure that Ello could never sell member data and that we would always remain ad-free.” But setting up as a PBC is no guarantee that a startup will resist the very real business pressures that have made Facebook what it is. As Rae André writes in a 2012 assessment of public benefit corporations, “benefit corporations follow accountability practices that serve particular private interests, and because of this, the probability that they will be responsive to the citizenry as a whole, to society, is low.”

Call me old-fashioned, but there’s only one kind of Facebook alternative that I’d actually trust to look after its users’ data, privacy and information access: one that is collectively owned.

Imagining a cooperative Facebook

A socially owned social network — let’s just call it a socially owned network — could follow the lead of the consumer co-operatives that exist in a number of countries and sectors. Unlike worker or producer co-ops, which are owned by the people who work in them, consumer co-ops are owned by the customers — or in the case of a social network, by the users. In “A Theory of Co-Operatives,” R. Carson describes these co-ops as follows: “a consumer co-operative, or consumer-managed firm, is managed by elected representatives of at least some of its retail customers, the latter constituting its membership.”

As an organizational model, the modern co-op traces back to the founding of the Rochdale Equitable Society of Pioneers in 1844; as Joel David Welty puts it, the founders of that society “simply wanted to create an institution they could control themselves for their own benefit.” In the nearly two hundred years since, the co-operative has proven itself as a viable model for combining enterprise with social good. Carson writes, “In some instances, a co-operative may be the least-cost way of avoiding potential external effects.”

As a business model that’s explicitly designed around member participation and benefit — and backed with a proven legal framework — a cooperatively owned social network would be a lot more credible when it comes to offering long-term protection of user privacy, data and information access. And indeed, we’ve seen a number of efforts at building social networks on co-operative principles. The #BuyTwitter campaign is busily lobbying for a user takeover of the flagging social network. Diaspora is a decentralized, open source social network that turned itself into a “community-run project” in 2012, housed under a non-profit. Social.coop is an actual cooperative that provides accounts for a server that’s part of Mastodon, another decentralized social network.

The tough choices for a cooperative social network

None of these networks have even come close to the level of success that would allow them to seriously challenge Facebook. And I suspect that’s because there are hard trade-offs involved in creating a broadly appealing social network: a social network that is comprehensive enough, in its membership and content, to replace Facebook in the lives of its users.

Take one obvious example: I’d rather not be part of a social network that enables hate speech. I also want to be on a social network where I can reach and connect with a very wide range of people. But some of the people I want to connect with (let’s say family members and former colleagues) are people who (sigh) may want to connect with people who belong to white power groups, or anti-gay groups. If I were part of a social networking co-op that reflected my principles of social equality by banning hate speech, those white power and anti-gay groups wouldn’t be on the network. The people who belong to those groups would stay on Facebook. And the family members I’d like to stay in touch with on my social network will stay on Facebook, because that’s where their friends are.

If we want to have values-driven social networks, we’re going to create social networks with walls. Precisely because those walls keep some people out, our values-driven networks will never be as big as Facebook…and indeed, may never reach the scale that makes them sustainable. Facebook’s lack of values is what makes us want to leave; it’s also what makes Facebook elastic enough to contain — or more accurately, to capture — everybody.

To have a sustainable, user-centered alternative to Facebook, we’ll have to wrestle with a great many contradictions like this one. We’ll have to wrestle with the tension between co-operative values of progressive social inclusion, and cyber-libertarian values of openness and free speech. We’ll have to contend with our simultaneous desire for wide-open access to information, and for insulation from misinformation. We’ll have to choose between getting flooded with irrelevant updates, and trusting some algorithm to make those choices for us. Some of us might even have to choose between chasing the Silicon Valley dream of making a billion dollars, and the human dream of making a social network that’s actually good for its users.

In other words, we’ll have to make all the difficult, awkward choices Facebook now makes for us. We probably won’t get it right. We’ll fight, we’ll quit, we’ll try agin.

But Facebook’s pathetic response to the fake news epidemic has made one thing abundantly clear. As hard as these choices are, they need to be our choices—not Facebook’s.

NO REVOLUTION WITHOUT RELIGION (2012)