Subscriptions live - Check your email if you subscribed during the fundraiser ~
Image: Alina Grubnyak, 2018, Unsplash License.

Who Is “Web3” For?

Blockchains can't decentralize the internet – but the cooperative Fediverse can.

In a world where a single company, which controls the conversations, news feeds, and personal connections of almost two billion people, considers it a good idea to base its post promotion algorithm on how angry a post makes its readers, we can perhaps conclude that the time has come to decentralize our digital communication spaces. Users of a recently-bought social network seem to agree.

Those with vested interests in the cryptocurrency space claim to have a solution ready: web3. What is web3? Anything you can dream of — as long as its proponents need not define it too clearly. Any web3 scheme does seem to have a certain necessary element, though: blockchain — along with tradable crypto-assets (coins or tokens).

Incentives and Investment

Ostensibly, blockchain is used because it enables certain traditionally centralized tasks (global identity management, for example) to be performed in a “decentralized” way, without single points of control or failure. But proponents of web3, and more broadly of cryptocurrency and blockchain-adjacent technologies, are also rather explicit that it’s just as much about “investment opportunities.”

Different blockchain projects approach this differently, but in the end, those who invest early in the given system ostensibly gain crypto-assets for their trouble. Or as Mike Masnick puts it in his seminal text “Protocols, not Platforms” : “a token-based cryptocurrency is the equivalent of equity in a company — but rather than the value being tied to the financial success of the company, the value of a crypto token is tied to the value of the overall network.”

No wonder, then, that the likes of Elon Musk and Jack Dorsey — privileged, moneyed, well-connected, extremely influential white technologists — think “blockchain” whenever they hear “decentralization.” Some have been involved in building the centralized walled gardens we so urgently need to break out of, others own them; and now they are pivoting to blockchain-based projects. They recognize, with startling clarity, what Cory Doctorow unpacked earlier this year: whatever the particular web3 project’s service is, it is there to generate a need for a specific crypto-asset in order to drive up its price, and to ensure pay-off for those who invested early.

Results for the search term “decentralization” are saturated with these vague, mysterious, hype-building computer renderings. If a peer-to-peer, decentralized internet becomes widely adopted, perhaps we can visualize decentralization with other images and metaphors, like flocks of birds or the like. – Ed

Image 1: A glowing cube of power made from smaller blockchain blocks, Shubham Dhage, 2021, Unsplash License;

Image 2: Blockchain blocks, set against matrix-esque arrays of data, Shubham Dhage, 2022, Unsplash License;

Image 3: An array of red/Green, on/off, toggle buttons with mystical blockchain symbols embossed on their golden surfaces, Arthur Mazi, 2022, Unsplash License

Regardless of how much cryptocurrency enthusiasts insist they are “truly democratic forms of money,” blockchain based systems echo and amplify the power dynamics of centralized walled gardens. Ethereum’s Gini coefficient — a measure of income or wealth inequality in a financial system — is on average considerably worse than the most unequal places on Earth. So is Bitcoin’s.

In other words, web3 is less a technology project for decentralizing the internet, and more an economic project for a select few to profit from: those who acquire crypto-assets early or have the resources and knowledge to run Ethereum validators (or the analogs in specific blockchain systems used by given projects).

Chain of Trade-offs

Like any technology, blockchain-based systems are not without significant trade-offs. These revolve around block size (and thus the number of transactions that a single block may contain), how often a block is added to the blockchain, and what information is kept directly on-chain (what is included directly within a block).

The more transactions per block and the more often a block gets generated, the higher the global transaction rate that can be supported by a given blockchain, but also the faster the whole blockchain grows.

Ethereum — perhaps the most popular blockchain system used as a base for many web3 projects — adds one block every twelve seconds. With around 140 transactions included in a block on average, its global transaction rate is just shy of twelve transactions per second. The average block size of 70 kilobytes means that Ethereum blockchain grows by half a gigabyte each day, and its current size is almost half a terabyte.

If half a gigabyte per day does not sound like much, consider that within most blockchain systems (including Ethereum) as many people as possible are expected to run their own nodes; the more people running nodes, the more decentralized the system is. Each node requires the whole blockchain to be stored locally, with some additional available storage space required. The bigger the blockchain and the faster its growth rate, the harder it is to run a full node, and the fewer people will do it, leading to a less decentralized system.

It’s a fine balancing act for web3 projects: use a smaller block size, and transaction rates become far too low to support a global system serving anywhere near the estimated five billion Internet users today; use bigger blocks, and the blockchain size and growth will make it impossible for enough people to run nodes that the system remains truly decentralized.

Then there’s the on-chain/off-chain dilemma. Data that is kept directly on-chain is cryptographically signed and embedded in the history of that blockchain. But including this information adds to block size, and thus to the rate of growth of a given blockchain, making it more difficult to run nodes. Keeping important data off-chain lowers the blockchain size and speed of its growth, but necessitates their centralized storage somewhere. Where? Under whose control?

This is painfully obvious in the Non Fungible Tokens (NFT) space. NFTs are cryptocurrency tokens that are supposed to enable the “ownership” of digital artworks (they do not, but that’s a separate issue), by recording ownership on a blockchain (often Ethereum). However, including a whole image (often megabytes in size) as part of a transaction would bloat the blocks well beyond the average transaction size of a few dozen kilobytes, and would drive up the size of the blockchain immensely.

Often, instead, only an artwork URL is recorded in the transaction. But URLs can break, leading “owners” to see 404 Not Found errors instead of their digital art when trying to view it on major NFT exchanges. As a result, the system is neither reliable (because the link between the NFT and the artwork can break), nor meaningfully decentralized (as the artwork is still referenced via a “centralized” web URL). Such an unreliable, centralized system could, of course, easily be built without any blockchain – but then how would the validators get paid? 

After musing about “blockchain Twitter” (where each tweet was to be included on-chain as a transaction and would have a price-tag associated with it), Musk himself quickly noted that it simply would not scale well enough. In a message to investment banker Michael Grimes, he wrote: “Blockchain Twitter isn’t possible. The bandwidth and latency requirements cannot be supported by a peer to peer network, unless those ‘peers’ are absolutely gigantic, thus defeating the purpose of a decentralized network.”

Imgix data servers. Image: imgix, 2017, Unsplash License.

In other words the basic choice for any web3 project that is supposed to scale to large numbers of users is this: either faux-decentralization, because the nodes must necessarily be “absolutely gigantic” (and thus cannot be run by enough people to be considered decentralized); or keeping important data off-chain (which usually relies on centralized external services, defeating the purpose).

Alternatively, one can drop the blockchain entirely. BlueSky, which was incubated in Twitter’s “blockchain team” and aims to create a decentralized social network, has finally published a draft protocol specification which, notably, does not include any blockchain-based technology. Instead, they appear to have concluded that a centralized “Placeholder” the system (the actual name) is more practical. Still, BlueSky’s architecture, despite being decentralized without blockchain, seems designed to leave a window for a “winner takes all” monetization (and control) strategy. 

An Actually Decentralized Social Network in Word and Deed

Meanwhile, a decentralized, federated social network is flourishing which is built within a protocol that does not require blockchain, and does not convert every human interaction into a financial transaction. It focuses on communities and building useful, thoughtful, safe tools, instead of fetishizing buzzwords and pitching free speech absolutism in order to sell snake oil.

The Fediverse is built around a standardized protocol (ActivityPub) with a number of independent but compatible implementations, client apps, and thousands of instances run by individuals and communities. In other words, it fully realizes the “protocols, not platforms” idea.

There are no tokens to invest in. There is no cryptocurrency to “hodl.” No blocks to mine or validate and no blockchain to grow.

Image: Christina Kirschnerova, 2019, Unsplash License.

Instead, there are servers (called “instances”) run for and by specific communities — like There are geographically-themed instances (say,, instances for programmers (, general-purpose instances (like, where I have my fedi-home) and many more. There are large instances with hundreds of thousands of users; tiny one-person instances; and anything between.

There are even instances run by and for governmental institutions, like the official Fediverse instance of the European Union or the server for German federal authorities.

Depending on the software these instances run they can feel more like Twitter, Instagram, or Facebook. There are Reddit-like instances, video hosting instances, and Eventbrite-like instances. Once you have an account on one of them you can follow and interact with users on all of them: imagine being able to use your Twitter account to respond to your family on Facebook, engage in a thread on Reddit, follow your friend on Instagram, and subscribe to an Eventbrite event.

Servers are not free, so I am donating to help my instance admin cover the running costs. Many instances are community-supported. On most instances, this is optional (if welcome) — but it is easy to become a member of the Fediverse without spending a dime. In fact, you’re encouraged to! Some “fediversians” decide to run their own servers, or pay for managed instances from service providers like Mastohost.

Social Media is Better in Moderation

Compared to web3 platforms, the Fediverse does not claim to be “censorship-free,” or a “free speech zone.” Each instance has their own rules, and decides what is allowed and what is not. There is a granularity in the moderation tools on “fedi” (as Fediverse is often fondly referred to) that is simply unavailable on big tech nor web3 platforms. Each user can block or mute other users and whole instances. Instance admins and moderators can mute and block users from other instances, or “defederate from” whole instances (in addition, of course, to moderating users registered on their own instance). Alternatively, they can decide that content from a given instance is allowed, but will by default end up behind a content warning.

This makes it easier to deal with abusive users. If an instance enables abusive behavior and ignores abuse reports coming in from other instances, it will be blocked (“defederated from”) by instances who choose to protect their users from abuse. This effectively quarantines toxic groups, limiting their reach to instances that call themselves “free speech” zones (yes, these exist too). Their users can, of course, set up accounts on other, less permissive instances (which are thus not as widely blocked). But to remain there, users need to respect the preferences of these instances.

Deplatforming toxic users from big tech platforms is so contentious partly because it’s an all-or-nothing decision: a monopolistic provider of a centralized service decides whether or not a user is able to post to that global platform at all. On fedi each of the thousands of instances makes their own rules and decisions, so it’s hard to argue that anybody is truly “censored.” If a thousand different, independent community sites refuse to carry somebody’s posts, perhaps it’s no huge conspiracy? Perhaps people just don’t want to be exposed to certain kinds of abusive behavior?

This makes moderation decisions easier for instance teams and more meaningful for the users, as they focus more on specific behaviors and less on specific people. And if you don’t like the moderation on your instance, you can set up an alternative account somewhere else, or even migrate your account, keeping your social graph intact. This is also a good solution in the event that an instance is shutting down.

Values vs. Valuation

In the end, it’s all a question of who has power in a system, and the values the system is imbued with. Where the Fediverse focuses on community and communication, web3 focuses on narrow, somewhat naïve notions of “ownership” and “free speech.”

“Ownership” in the web3 context is almost entirely financial and transactional (“I own something if I can sell it”), and surprisingly often based on wishful thinking. After all, legally speaking, users “own” the content they posted to Facebook without any need of a blockchain record. On the other hand having an NFT transferred to your Ethereum wallet doesn’t make you the “owner” of the digital asset which the NFT refers to, unless a binding legal contract has been signed (which NFT-holding “anti-right-clickers” refuse to believe, much to their chagrin).

“Free speech,” on the other hand, seems to be understood by the web3 community mechanistically, along the lines of: no technical interference in the act of speech (for example, posting). It completely ignores the very real chilling effects targeted harassment has on speech (online and off). Perhaps it’s not surprising, then, that the biggest proponents of such mechanistic “free speech absolutism” happen to have immense platforms and be in a position to inspire harassment campaigns themselves.

When radium was first discovered at the end of the 19th century, a whole slew of snake oil products emerged capitalizing on the sensationalism surrounding the new element and its radioactivity. Perhaps the most absurd product was the Doramad Radioactive Toothpaste, whose promotional materials used naïve and distorted notions of “energy” and “radioactive rays,” to market radioactivity as a solution to the very real problem of tooth decay.

The analogy is quite compelling. Like radioactivity, blockchain as such can be a useful tool in solving certain kinds of problems. Like dental hygiene, the decentralization of global communication platforms is an important problem, but not necessarily the right application for the instrument. Like Doromad Radioactive Toothpaste, web3 has little to do with solving the stated problem, and everything to do with profiting off of a buzzword, resulting in more harm than good in the process.


  • Michał Woźniak

    Michał "rysiek" Woźniak is a hacker and a digital human rights activist, focused on information security, privacy, copyright reform, and digital media literacy. He worked at OCCRP, where he was responsible for infrastructure and the cybersecurity of investigative journalists. Today he manages the information security of the Icelandic national domain (.IS) registry.

Strange Matters is a cooperative magazine of new and unconventional thinking in economics, politics, and culture.