As of this writing, the US Senate is expected to vote soon on a pair of ambitious antitrust bills targeting the dominant internet platforms. The European Union is finalizing its own suite of new regulations. And states around the US are passing laws—some better, some worse—that seek to wrangle a tech industry widely seen as out of control.

To Ben Tarnoff, these developments are woefully inadequate. In a forthcoming book, Internet for the People, he argues that the internet’s problems are fundamentally tied to the profit motive; only a move to public ownership can solve them.

“The internet reformers have some good ideas, but they never quite reach the root of the problem,” he writes. “The root is simple: The internet is broken because the internet is a business.”

Tarnoff sees promise in the successful examples of cooperatively and municipally owned broadband networks throughout rural America. But what would it mean to place the web itself—the websites and apps we use every day—under public ownership? Tarnoff recently spoke to WIRED to lay out his vision for a socialist internet and how to achieve it.

This interview has been condensed and lightly edited.

WIRED: The central argument of your book is that we need to “deprivatize” the internet. That implies that it was once public.

Ben Tarnoff: The internet protocols, which are the rules that allow the networks of the internet to communicate with one another, are invented in the 1970s by DARPA researchers. Then the Pentagon uses those protocols to interconnect various networks, starting in the 1980s. That network of networks then passes into civilian federal control, under the National Science Foundation.

The pivotal year is 1995, at which point the National Science Foundation terminates its backbone, a core artery of the internet up until that time called NSFNET, and the private sector takes over. So that's where privatization as a process starts: in the so-called basement of the internet, with the pipes.

There are many places around the world that have way faster, way cheaper internet than in the US, and it's provided by the private sector. So is the problem here privatization, or is it deregulation? The internet wasn't just handed over to the private sector in the US, it was handed over on super-favorable terms.

You're pointing to something important for people to understand, which is that the US has a highly concentrated market for internet service. We have four companies that control 76 percent of internet subscriptions in this country. As a result, we pay some of the most expensive rates in the world for awful service. I mean, we pay higher average monthly prices than people in Europe or Asia. Our average connection speeds are below that in Romania and Thailand.

This sounds like an argument for antitrust enforcement to increase competition, rather than getting rid of the whole concept of for-profit internet service providers.

You raise an interesting question: Is my goal simply better speed for lower cost? Or is there something else? Research shows that if you were to bring competition to the highly concentrated market for internet service in the United States, it would almost certainly improve speeds and lower cost. That's a very important goal. But it's not quite enough, for two reasons. One is that competition tends to work best for people who are worth competing for, which is to say, competition is best at bringing down prices for higher-end broadband packages. Where competition is not so effective is in bringing connectivity to people who really can't afford it, or who live in communities, particularly rural communities, in which it's not profitable to invest under any circumstances.

A bit of a grander explanation is that, under a private system, people do not have the opportunity to participate in decisions about how their infrastructure is deployed, developed, managed, and so on. And that's where I have a lot of faith in community networks, because as publicly and cooperatively owned alternatives, they have the opportunity to encode and accredit practices that actually give users a say in how the service runs.

It sounds like you believe the antitrust and competition stuff is necessary, but not sufficient.

Yes, that is what I think. Anti-monopoly is actually a pretty rich and diverse tradition. I see anti-monopoly measures as quite useful for curbing the power of these firms, for shrinking their footprint. And there are a lot of specific measures, like demanding interoperability between social networks and breaking up the big firms, that I see as quite useful intermediate measures for a deprivatized internet. But there's a deeper disagreement that I have with the anti-monopoly folks about where the root problems with the internet come from, and what the ultimate horizon looks like.

It's hard to argue with the idea of democratic control over networks that people use and rely on, but I'm skeptical about what that really looks like in practice. I think that people overwhelmingly just want it to be fast and affordable. And there's a pretty strong argument that the the the most direct way people will enact that preference is by choosing in a market.

I think part of what you're asking is, what are the decisions that are worth making in a more democratic, deliberative way? There are a number of issues that come up around where and how to deploy infrastructure. When you're building the last-mile broadband network, for example, or even what's called a middle-mile broadband network, there are a lot of choices. Which neighborhoods are going to be served? Which technologies are you going to use? Are you going to try to incorporate with a smart grid, which can improve energy efficiency? These are matters of concern to local community members, and these are the kinds of questions that come up, for instance, in the rural cooperatives in North Dakota that have been very successful in building out their broadband networks.

We've been talking so far about the bottom of the stack: the plumbing, the cables, and stuff. The conversation gets even more interesting and complicated when we move up to the application layer of the internet, like Facebook or Google. With the cables, the components of the network themselves are scarce. It's a physical thing, and you have to decide whose it is, and on what terms access will be allowed or prohibited. I struggled to apply that framework to the application layer. Are you talking about literally outlawing profit-driven web applications, or is there some other way to drive them out of existence by offering nonprofit alternatives?

I don't see the immediate abolition of profit-making on the internet as a particularly practical proposal. I do have other proposals about how we could nurture deprivatized alternatives and begin to grow a deprivatized sector.

While reading your book, there were a lot of moments when I thought, “What this shows is that we need better regulations.” Government could certainly subsidize the build-out of broadband to people in rural areas who are ignored by a competitive market. You write about Uber being so successful at working the political process because their business model depends on classifying workers as contractors. And there are other examples where it feels like what we need are better policies governing how these businesses operate, as opposed to killing the profit motive entirely.

Well, I agree that regulation of these firms is essential. Where I part ways is that at the end of the day, if we want to build a better internet, I think that we need to transform how it's owned and organized. To my mind, leaving the internet in the hands of private firms, and leaving it organized around the principle of profit maximization, means that there's only so much that public policy can do.

Changing the ownership model is not merely something to do for its own sake. It's really a means to an end, which is an internet in which people have the opportunity to participate in the decisions that most affect them.

The goal of a more participatory internet reminds me of the Web3 movement, which in theory is about putting internet platforms on the blockchain and giving users more ownership over them. I recently did quite a bit of reporting on Web3, and the thought that I had over and over again was: Most normal people just don't care. Most people don't want to vote on proposals about the development of the protocol that they're using. They just want it to work. I know this sounds very cynical. But how do you sell somebody on a deprivatized web? How exactly would their world be better?

Here, I turn to experiments in the so-called decentralized web community—in particular, decentralized social media projects like Mastodon. Mastodon has been around for a while. It's an open source project that enables people to build their own social media instances and federate them together. It's interesting and promising because it enables people to form social media communities in which critical governance decisions, like content moderation, can be made on a democratic basis and in which a cooperative of users can come together to determine how their social media community should be run.

I agree that the Mastodon example is interesting, but as you say, it has been around for a while, and there's just not a ton of demand for it. I think that having pluralism and federation and distributed, community-level control makes all the sense in the world. And yet, it's not what people gravitate toward.

This is where we need to talk about public investment. Mastodon is an open source project. Open source projects always have their challenges in terms of getting enough people to contribute and making sure that they're maintained properly. It's also relatively expensive to run your own Mastodon instance because it's so computationally intensive. And then there are a number of questions, like “Is the UX good enough to draw people from Facebook?” Facebook has a lot more money to play with when it comes to that sort of thing. We can't scale these alternatives up without public investment—and, I should say, without social movements, because the other point of my book is that if we want to transform the internet, we need to create a social movement capable of demanding that transformation.

There's a bit of a chicken-and-egg problem here. It seems difficult to the point of hopelessness to try to galvanize a social movement to achieve something that you can't really describe because it doesn't exist. How do you get people fired up to usher in a new version of the application layer of the web that you can't sketch out very concretely?

You use the term chicken-and-egg, which I like, but I would maybe use the term dialectical in this case. Being able to point to small, yet promising, experiments like Mastodon or like worker-owned ride-hailing services, gives people a sense that another internet is possible, and in turn enlarges their imagination of what the internet may look like. Those can be important starting points for the types of conversations that lead to social movement organizing. We need those alternatives to exist even in their miniature form at the moment, but we also need social movements that can enlarge and strengthen those alternatives in order to inspire more people.

Part of the problem here is that we're still working within an enemy paradigm. My ultimate horizon is not a cooperatively run Twitter. That, to me, is a constriction of our imagination in terms of what's possible. It makes sense that that's where we start because we have to start somewhere. But ultimately, what excites me is the possibility of bringing masses of people together, connecting them with the technical resources they need to build the online spaces and structures that can serve their everyday lives.

Speaking of technical resources: To take a concrete example, I could imagine something like Facebook Marketplace existing on the local level as a cooperative or municipal service. If I'm shopping for a used couch, I don't need to be able to see what's for sale in Miami. I really just want to see what's in the East Bay, where I live. And yet it strikes me that I could never build a Berkeley Marketplace that competes with Facebook. And I don't know that I could find someone to build it for me. The people who have the technical training to do that are off getting rich working for Meta. So it's all well and good to rally my community, but are any of them master coders?

You're pointing to a real problem, which is that there are severe material constraints that exist for groups of people who want to build alternatives to the platforms. This is where I think public policy has a really important role to play. In the book, I talk about an experiment that the Labour Party conducted in London in the 1980s, where they created these spaces called technology networks. These were buildings that people could walk into and get connected with machine tools, a bit like hackerspaces or makerspaces. Today, they could get connected with experts and forms of expertise, and they could build technologies that made their lives better. A lot of energy efficiency technologies came out of these centers, and blueprints for what they built went into this shared databank that anyone else could access. That, to my mind, is an interesting model for how we might use public policy to connect non-technical people with technical resources so that they can actually build the kinds of online tools that would make their lives better.

Just to level-set a bit: Do you think that anything should be for-profit? Is your argument that the internet should not have a profit motive, and neither should anything else? Or that there's something special about the internet?

The former. This book is scoped to the internet, but to answer your question, I am a socialist. I want to see a post-capitalist society. There are dynamics to the internet that require special attention, but I see it as part of a broader political economy that needs to be transformed.