The Battle for the Soul of the Web

Long before the NFT boom or the Web3 backlash, an unglamorous movement was under way. Where does it stand now?

Erik Carter / The Atlantic

Recently, at a fancy arts complex in Manhattan, the billionaire Frank McCourt led a three-day series of talks and workshops about the future of the internet—part of his expensive effort to “fix technology, save democracy.”

In the lobby, attendees networked in a cocktail bar created by the superstar restaurateur Danny Meyer; in front of the main stage, they held up blue and orange glow sticks to record their votes in polls like “Which will kill us first?” (AI or climate change) and “Who would you rather take care of your children?” (a surveillance robot or TikTok stars). The agenda for this conference, Unfinished Live, was almost random in its diversity: Attendees could learn about how Indigenous communities were using decentralized technology to create their own maps, and they could also learn about importing products into the metaverse, starting with a sweater that has a microchip in it. (This allows the sweater to “accrue value based on who owned it last.”) The Facebook whistleblower Frances Haugen invoked several times the necessity of a Mothers Against Drunk Driving for social media. Nadya Tolokonnikova, a founding member of Pussy Riot, vaped onstage and responded coolly to questions about how blockchain tech is used to fund the defense in Ukraine. “There’s nothing particularly magic about cryptocurrency,” she said. “It’s a tool, like a road or a gun.”

The disparate threads could all be tied to the same point of origin: It feels like things have gone wrong on the internet. Decades removed from the gonzo highs of blinging GIFs and wacky blogs, the web is now a place where many people feel exploited, manipulated, and tracked; where freedom of speech is being tugged around in a strange culture war; and where the rich get richer.

Among this set, one solution seems to be the consensus favorite. If these problems are intrinsically linked to consolidated tech giants like Meta, Google, and Amazon, why not embrace technologies that decentralize power? This has become a key issue for Brewster Kahle, the 61-year-old founder of the Internet Archive, a nonprofit and digital library created in the late 1990s. (You might know it from the Wayback Machine, which has crawled and snapshotted billions of webpages for posterity.) When I introduced myself to him at a morning workshop on water scarcity, he was wearing a Jansport backpack and black shoes that appeared to be nonslip, possibly appropriate for work at a high-volume restaurant, and he was open to sitting down immediately for a 90-minute conversation about the major problems facing the web.

Kahle would remake the web as an endless library, which could hold copies upon copies of everything you’d want to know and mimic “the robustness that we have in the physical world.” Today, the tech giants have tremendous authority over the information that passes through their platforms. These platforms can remove any data on a whim; they might do this intentionally, for their own obscure purposes, or they can be subject to copyright takedowns, geopolitical demands, and other outside threats. They collect and track personal details and target content against them. Centralization, the logic goes, makes all of this possible, and leads to all of the problems that result.

In 2015, Kahle put out a call for a “decentralized web,” or a web that looked more like the one that early visionaries such as Tim Berners-Lee had imagined. “The way we code the web will determine the way we live online,” Kahle wrote at the time. “So we need to bake our values into our code. Freedom of expression needs to be baked into our code. Privacy should be baked into our code. Universal access to all knowledge.” Pivoting to a decentralized version of the web—evading massive platforms, sharing peer-to-peer—could enable much of this by giving ordinary people control of their own data and a broad range of options for publishing or accessing information. His manifesto instigated a movement called “DWeb,” which began with a star-studded summit, and has continued with annual retreats in California, which are referred to as “DWeb camp.”

At Unfinished Live, Kahle and several of his colleagues from the Internet Archive were finally participating in a major decentralized-web conference with big money behind it, and though the focus was on repairing social media—not Kahle’s personal bailiwick—he found it exciting. “We have structural problems in how this technology works,” he said. “We need another run at it. And I’m encouraged. I’m optimistic. Look at this—people coming together in New York and not to talk about how to make their next deal.”

The decentralized web that Kahle and others have envisioned for years has yet to receive major mainstream attention for an obvious reason: It never promised to get anyone rich. But the Web3 movement certainly did. A close relative—or dark foil, some would say—Web3 is defined by blockchains and cryptocurrencies and NFTs. Like all kinds of things that are marketed to adults of my generation, much of this was sold by celebrities, or in the language of empowerment, or with sleek branding that promised to make participants part of a stylish future, nevermind what else that future might entail (as long as it included a vertical profit line). Not an insignificant fraction of the population was repulsed by this, I should say. Nevertheless, people jumped on the bandwagon, leading to hype and then a crash.

It reminds Kahle of the dot-com boom, when the internet first moved out of an academic-government space and into the commercial sphere. (“Scams galore!”) “The hype has gotten ahead of things,” Kahle said. But it’s also given the movement an opportunity to readjust—and, perhaps, realize its dreams.


The difference between DWeb and Web3 is material or semantic, depending on whom you ask. The terms are “in some ways superficial,” according to Glen Weyl, an economist and a co-author of Radical Markets: Uprooting Capitalism and Democracy for a Just Society. “On the other hand, they’re very important.”

In a recent webinar series, the Internet Archive defined the decentralized-web movement as an effort to break apart “all the layers” of the current online experience. It’s helpful to think about this idea in terms of what it opposes: Meta, for example, centralizes messaging, media sharing, data collection, and much else, so users are subject to its content-moderation policies and can’t help but submit their information to its sprawling marketing apparatus. Amazon owns so much of the infrastructure that the internet runs on that you could hardly function without it.

The DWeb movement is interested in subverting this status quo through tools that would give individuals greater control over their online identities and information. “I’m trying to channel the confusion that you’re looking at me with right now,” Kahle told me, when I asked him to explain it. “How do I help other people understand what the heck is going on here?”

Some things would be really different for the average web user—she might no longer rely on Facebook or Google to verify her identity when logging in to various sites or be followed around by advertisements that know all about her. Other changes would ideally be unnoticeable to her. For instance, a decentralized version of The Atlantic’s website might look the same, but the underlying machinery would be quite different—it might be hosted on any number of independent servers owned by users around the world, rather than through a major provider controlled by a big tech company.

Web3, according to the Internet Archive’s series, has a narrower technical definition: It’s “the ‘blockchain-ification’ of the web, using blockchain technologies and cryptocurrencies to verify transactions, pay for services, and certify content such as NFTs.”

The two communities—or perspectives, or movements—overlap, and the boundaries are fuzzy. Filecoin, a decentralized storage project, is funded and funds other decentralized projects with a fortune raised through an initial coin offering, or ICO (the crypto equivalent of an IPO); the Ethereum Foundation, a nonprofit established to support projects related to the titular blockchain, is a sponsor of the Internet Archive’s annual DWeb camp; and Twitter is experimenting with crypto and working on a new decentralized social-network protocol simultaneously.

At the conference, speakers used “Web3” interchangeably with “decentralized web.” They come up in the same breath because crypto is fundamentally decentralized itself, and because their proponents have a shared cause. “There’s an underlying agreement between Web3- and DWeb-aligned people that they care about user self-determination,” Mai Ishikawa Sutton, a fellow at the Commons Network and one of the organizers of the past two DWeb camps, told me. “They care about being able to control your data, at some level having information that is transparent, that the system is transparent, and that it’s not about having one entity controlling everything, in theory.”

Sutton was part of a group that published a set of “organizing principles” for the DWeb movement in 2021—a process that took several years. By dint of having principles at all, DWeb became a smaller tent than Web3. People who disagree on a lot of things can agree on a technical system, Nathan Schneider, a media-studies professor at the University of Colorado at Boulder and a frequent writer on collectivism and tech, told me. “DWeb asks more,” he said, and dwells on two key questions: “What do we actually want socially, and how do we center those values in our technical designs, so the technical becomes a means to an end, rather than an end in itself?”

This ethos is expanded on in the new digital magazine Compost, edited by Sutton. Compost is also an example of the decentralized web in practice; you can access it on your regular browser, Google Chrome or whatever you use, and you can access it via Hypercore or IPFS, peer-to-peer protocols you probably have not heard of. Those protocols offer advantages: If you like what you read, you can download the magazine in its entirety, ensuring that you can read it later in a decentralized browser like Brave or Agregore without being connected to the internet. By making this copy, you can also become a node in the decentralized network, so that when someone else asks to see the magazine, your computer may be the one that serves it. As long as individuals keep sharing the content, it remains accessible—not so with a PDF hosted on a company’s web server.

This technology wouldn’t fall under some definitions of Web3, because it doesn’t use a blockchain. In fact, the underlying mechanics have been around for a long time—it’s the same basic idea that made Napster work in 1999, or BitTorrent in 2001. And there are other elements of the DWeb that definitely wouldn’t be called Web3, like community-owned mesh networks, which serve as last-mile infrastructure for internet service or help groups maintain their own private intranets without relying on a service provider like Verizon.

These decentralized technologies, it could be said, are focused on empowering individuals; “pump and dump” cryptocurrencies and NFT projects, perhaps less so. Sutton has been saddened by the conflation of the two ideas. “When I say I’m working on a crypto project or I’m working on the decentralized web, it sounds like I’m part of this movement to build a series of exploitative pyramid schemes,” they told me.

This was the discrepancy that motivated my original curiosity about DWeb and Web3 as entwined but separable strands. How must it feel to start something idealistically, in obscurity, and then watch it become a global phenomenon for reasons that you would never have chosen? But this question turned out to be more complicated than I thought. Some rejected the question itself. Weyl, for one, acknowledged that the projects and products that got the most attention over the past two years were “horrible, hyper-capitalistic, financially exploitative idiocy,” but also argued that all of the other DWeb stuff I was talking about “wouldn’t have gotten anywhere near the attention they’ve gotten, including your interest, if the whole Web3 thing hadn’t happened.” The moral distinctions I was buying into were, at some level, petty tribalism, he said. DWeb may have existed first, but “as a practical matter they’re riding on the coattails of Web3.”


The polar-opposite perspective would be that Web3 acolytes appropriated the decentralized-web movement, corrupted it, and are ruining it. Danny O’Brien, a senior fellow at the Filecoin Foundation, has espoused that point of view fairly literally; his Twitter bio currently contains the text “THEY STOLE OUR REVOLUTION // NOW WE’RE STEALING IT BACK.”

He describes a “dramatic split” between DWeb and Web3 as a recent phenomenon—back in 2019, at the last DWeb meetup before the pandemic, they had a lot more overlap, but they’ve become like “tectonic plates” separating in the past year or so. He attributes this to the growing concern about the ecological impact of Bitcoin, the conspicuous consumption of the NFT craze, and some aesthetic and cultural quibbles: “The traditional decentralized-web community is uncomfortable with the idea of the tech bro,” and with people who “were coming in because of the delicious smell of money.”

Capitalism has crushed idealism in many of the historic fights for the soul of the web—see, for instance, the battle between “free software” and “open source” in the ’90s—but O’Brien feels up to the challenge. “You start something idealistically in the tech world, it gets taken from you,” he said. “You have to steal it back. You can’t just sit there and let them do it. Right?”

The DWeb people have long been aware and wary of “dark forces,” as Schneider put it, “who see an emerging technology as an opportunity to consolidate their own power and wealth.” Although the rhetoric is dramatic, the work of the DWeb movement steadily continues as it always has: The community takes funding and attention where it can, moves from ideas to demos, and hopes its projects become useful to someone.

It’s not dazzling stuff. Kahle analogizes the effort to rebuilding the internet’s “plumbing.” Especially after the crypto boom rightfully put skeptics on guard, he sees the challenge as convincing people that it’s worth it to push forward—that the internet really can be different, and better. “When there’s a new technology, people gravitate towards it with their existing anxieties,” he told me. “And unfortunately, you roll time forward, and people are disappointed by the technology, right? It didn’t fulfill their dreams. But it’s not because their dreams were wrong. It’s because we didn’t build good-enough technology.”

How do you inspire the boring work of iteration? In a recent essay about digital real estate for the Austria-based art magazine Spike, the sociologist Ido Nahari wrote about a longing for a return to the original ethos of the web, which at its best “demonstrated the democratic desire not only to idly exist within the world, but also to take part in its collective creation.” It doesn’t much matter what you call it, so long as you can get people feeling that way again.