There are two types of problems in the world: when something doesn’t work correctly, and when it does.

1. This product doesn’t work because of X problem. These are Class 1 problems.

2. This works perfectly, and that causes X problem. These are Class 2 problems.

Class 1 problems come when the product doesn't work properly. Class 2 problems come when it does.

These problems are highly relevant in the context of web3 (which is a word I’m using as shorthand for blockchains like Ethereum). But they are only relevant if you believe that one day, web3 will work perfectly and become mainstream.

This essay is an argument for why that day will come.

In particular, I want to argue that web3 will become mainstream because it resembles a ‘world computer’. A virtual computer that everyone can use and share.

Elsewhere, people have argued that web3 will become mainstream because it enables an ‘internet of value’, or because it enables true ownership online. On their own, I don’t think these narratives are useful or convincing. Instead, we need to consider web3 as a world computer to understand why it will go mainstream, and to understand what the Class 2 problems will look like in the future.

Let’s start by exploring what these other narratives are, and why they’re incomplete.

The first idea is that blockchain enables an ‘internet of value’. The internet we all know today is an ‘internet of information’, where anyone can publish anything, like sending an email, and you can transfer money, but only through companies, not directly via the internet. This means that sending money is limited to what their tech supports and where they operate.

Blockchain fixes that because it works almost like a big database, documenting what everyone owns, from cash to art to property deeds. To prove what we own online, we usually rely on middlemen, who extract benefits from us to prove it and also limit what we could own. Lots of exciting things are meant to spawn from bypassing those middlemen, like NFTs for art, like DAOs for community funds, or like the reward points on Starbucks’ new loyalty programme.

Why this is useful makes sense to me. How we send money now works pretty well within countries, but sending cash abroad requires you to jump through hoops, and it’s much harder to transfer anything else. But it’s not obvious to me why this will make web3 mainstream. Being able to send assets digitally does not feel like a big enough ‘pull factor’ on its own. It makes existing activities easier but feels too intangible to illustrate the new things it makes possible.

The second idea is that blockchain enables true ownership online. This is where the ‘web3’ moniker comes from. Web1 let people read things online, but it was technically hard for most people to create their own content. Web2 fixed that, by letting people write. With platforms like Myspace, Facebook, and YouTube, people could publish their own content on the internet. The problem was that to circumvent the technological difficulties of doing this, internet users had to hand over their content to those platforms, which took large portions of the advertising revenue, sometimes all of it.

In turn, web3 fixes that by letting people own their content. By storing content on the blockchain, instead of in a centralised server, web3 turns YouTube into one platform amongst many. It’s like if you could carry your videos and subscribers across YouTube, Vimeo, and everywhere else. Now platforms have to compete to attract creators, meaning creators are going to get a much better deal when it comes to ad revenue.

That ‘read, write, own’ evolution is simple enough to understand. The problem is that it only explains one small part of web3: the creator economy. In reality, there is much more going on.

Internet of value and digital ownership matter, but we need to take a step back to understand how it all ties together.

In reality, it’s all about data.

Blockchain gives data on the internet three properties that it did not have before: ownership, programmability, and trust.

IT’S ALL ABOUT DATA
Ownership. People can make and own assets without relying on a middleman.

This extends the read, write, own thesis. Ownership is certainly a big part of web3, but this is not just for the sake of ownership for the reasons illustrated above, it is also for the sake of visibility. Ownership and visibility are two sides of the same coin here. If you own something, but nobody else can see that you own it, or you cannot prove to anyone that you own it, then the relevance of your ownership is limited.

Letting people prove they own something online is a major innovation. And it extends well beyond the creator economy. The impact of blockchain is that all the assets and data you put on the blockchain, and who owns them, can be seen by everyone, and unlike in web2, there are no structural obstacles to that.2

Before we move on, it’s important to stress that whoever creates the data can easily choose to do so anonymously. Blockchain gives people the option to prove that they are the creator, which allows for more privacy, not less, because you share only what you want and when you want.

Programmability. Assets can be exchanged and automated.

Being able to own digital assets is meaningless if you cannot do anything with them. This is the usability layer for web3. It enables exchange. It is the ability to programme assets on the blockchain that lets you transfer a video, receive a recommendation from someone, or be acknowledged for engaging with a platform.

Or to automatically set the conditions for when an exchange occurs. When digital ownership happens on the blockchain, instead of on an organisation’s computer that most people cannot access, it becomes far easier for assets to be exchanged and automated:

Real estate is a $200 trillion sector, and it’ll become much quicker when property deeds are transferred on the blockchain rather than on paper.

A blockchain prototype made by the UK Government suggests that the time required to do this will fall from months to minutes.

UEFA put their tickets on the blockchain for the Euro 2020 football tournament, meaning they would only activate when users were near the stadium, preventing scammers from faking tickets.

Trust. Everyone can see and trust the relations between those assets.

Assets can be owned and exchanged, but equally critical for understanding the full scope of a blockchain is the idea that everyone else can verify what is owned and exchanged.

A big aspect of this comes from the fact that data on the blockchain is relational. It relates multiple parties. Leo sold a video to a collector. Leo received a recommendation from his colleague. Leo bought a ticket from UEFA. These events represent the exchange of assets, but without blockchain, they could only occur via a middleman to facilitate the exchange, who kept the knowledge to themselves. In web3, those relationships can form directly.

In web1 there were users, but only in web2 could they reach out to each other.

But in web2, we still needed middlemen to connect us. We couldn’t do this on our own.

That changes in web3, where connections can be made directly. There are still middlemen, but users are not reliant on them. YouTube becomes in competition with other middlemen.

Blockchain creates a network where anyone can directly create and exchange assets, and where everyone can see what is happening.

Last week I interviewed Justin Waldron, one of the co-founders of Zynga, the social gaming company that was bought for $13 billion during the 2022 stock downturn. He’s also the co-founder of PlayCo, a Sequoia-backed gaming unicorn (they raised a $100m Series A), and it’s with PlayCo that he is building in web3. When I asked him why, he summed it up in one word:

“Interoperability.”

Interoperability refers to the ability for anyone to see what people are doing on, for example, the Starbucks loyalty programme, and use that data in their own application. All those exchanges and relationships, across the entire network, can be tied together:

• Reddit, the internet forum website with half a billion monthly users, integrated the points that users award each other on some forums with the Ethereum blockchain, meaning those points could be sold or traded, whilst anyone can build apps that, eg, ‘only users with a certain amount of Reddit points can access’.

• There are 14 different services involved in making a financial security (according to PwC); the $14 trillion industry would work a lot more smoothly if everyone was working on one interoperable network, rather than different Excel spreadsheets.

• Justin is building a product called Storyverse that makes it easy for (web3) IP owners to license their IP to animated comic book writers.

**A WORLD COMPUTER
**Alone, ‘internet of value’ and ‘read, write, own’ fail to fully comprehend why the future of the internet lies on the blockchain. At best, they are two of three pieces in the puzzle. Owning assets online matters, but being able to programme these assets to facilitate exchange and make them useful is equally critical. Meanwhile, letting users trust those exchange relationships allows further relationships to be built on top of them. These three features combine to underpin a network that lets users create and exchange assets and build up a web of those relationships. In other words, blockchain underpins the foundations for a complete digital economy.

The best way to understand the blockchain itself is to think of it like a world computer that anyone can use.

Fundamentally, a computer consists of memory, processing, and communication. Data is stored on the hard drive and in other places like RAM. It is processed for different applications by different processing units, and the motherboard lets all these units talk to each other and makes the computer work.

• Ownership is our memory, and everyone can add assets to it.

• Programmability is our processing, facilitating asset exchange.

• Trust is our motherboard, which lets these exchanges build on each other.

This is not dissimilar to how a blockchain actually works. Indeed, Vitalik Buterin, co-founder of Ethereum, has oriented that blockchain around the ‘world computer’ idea for many years. It overlaps very closely:

• Consensus, to agree on what data is stored on the blockchain (like memory)

• Application, where applications occur (like processing)

• Networking, which handles communication (like the motherboard)

This lines up really nicely, and reflects the importance of each part of the blockchain by tying it to a critical reason for why the future will be built on blockchain.

Understanding blockchains as world computers that underpin digital economies is the best way to understand why it is the future of the internet.

Blockchains represent a virtual computer that anyone can verifiably store and exchange assets. These ownership, programmable, and trusting features create strong network effects. When one asset or piece of data is stored usefully on the blockchain, it becomes more appealing for the next piece to be stored there, because you could form a relationship or exchange with it.

And because everyone can see and trust that relationship, it becomes even more valuable for more data to join it, because you can programme more relationships on top of those existing relationships, creating an ever-increasing web of interactions that makes the network ever more useful and ever more valuable. Much like how an economy works.

There is a cold start problem for a network like this, but that cold start is already overcome. Already, billions of dollars worth of activity happens on blockchains, and more established organisations, from Reddit to Ubisoft to Nike, are already here. They’re closely followed by the likes of Starbucks and Disney. It is only heading in one direction.

And so we are now in a position to understand why the future of the internet lies onchain. The future of the internet will be built on the blockchain because it underpins a world computer that anyone can add to and interact with, building on top of the billions of dollars worth of value that already exists there. The future of the internet lies onchain because it is where the digital economy is already being built.

Let’s dive back into the Class 1 / Class 2 problems.

This distinction comes from Kevin Kelly, who uses cars as an example. A Class 1 problem, back in the 1920s, would have been something like ‘it only comes in one colour’. We’ve fixed all those problems, and now almost everyone drives. Cars work as intended nowadays: there are no class 1 problems, but the fact that everyone now drives has caused other issues, like environmental pollution, which is a Class 2 problem.

Today, electric cars are in their Class 1 stage. They don’t have enough range and are hard to recharge. But entrepreneurs will soon solve these problems and electric vehicles will be everywhere. This creates Class 2 problems, like strain on the electricity grid. If everyone shifted to electric vehicles, electricity demand would grow by 10% and vary in ways that power providers aren’t used to predicting.

Currently, web3 (blockchains and the ecosystem around them) is in the Class 1 stage. It’s too easy to be hacked, wallets are too complicated, self-custody is too difficult, gas fees are too costly.

But we don’t need to worry about Class 1 problems…

We don’t need to worry about Class 1 problems because web3 is such an attractive technology that entrepreneurs are rushing to solve them (as I argued last month). In doing so, they will unlock some of the value that web3 has to offer, web3 will become better for users, people will start using blockchains, NFTs, and crypto wallets, and entrepreneurs will end up with a successful company.

But even brilliant entrepreneurs cannot make brilliant products from bad tech. This is why I wanted to write a piece exploring exactly why web3 will go mainstream, and thus one worth working in. You don’t need to worry about the Class 1 problems if you believe that web3 has legs.

The stats bear this out. Developer activity in web3 has increased by 69% every year since 2009. Web3 startups have risen in number by 63% yearly over the same period.

Class 2 problems come from the same reasons that mitigate the Class 1 problems.

In solving the Class 1 problems that hold web3 back today, we risk creating new Class 2 problems that could be even bigger.

What could they be?

Kelly suggests that Class 2 problems are those the market doesn’t incentivise entrepreneurs to solve. You can build a successful business by solving a Class 1 problem. Doing so by solving a Class 2 problem is more complicated: for individual people, everything works well; it’s the fact that everyone else uses it that creates the problem. Nobody wants to pay for that.

In other words, this is a type of market failure. Specifically, a market failure in which we have ‘too much’ of whatever is causing the Class 2 problem — a tragedy of the commons. We solved the Class 1 problems, everyone started using the technology without thinking about how it might affect everyone else, and now that the technology is mainstream, we have a Class 2 problem.

But notice where this situation comes from. Class 2 problems arise because a technology is appealing. This appeal encourages people to solve the Class 1 problems, encouraging everyone else to flock to it. In other words, the factors pushing web3 into the mainstream will be the same ones that create the most significant Class 2 problems.

Recalling the reasons I outlined last month, what makes web3 appealing comes down to three different things:

• Ownership: people can create and own assets without relying on a middleman.

• Programmability: assets can be exchanged and automated.

• Trust: everyone can know the relations (like transactions) between those assets.

Web3 enables owning assets online, exchanging those assets, and proving that those exchanges happened.

How could this be a bad thing?

OWNERSHIP.

You can provably own assets on the blockchain without relying on a monopolistic middleman.

These are not exhaustive lists, but two major issues of ‘everyone being able to own assets without relying on anyone else’ are what it means for existing elites in society and the new elites who replace them.

Undermining the State’s Monopoly on Money

Everyone has heard the schtick about how blockchain is meant to replace the Facebook’s and YouTube’s of the world, but there is another type of organisation we rely on to verify what we own: national governments, which provide the ultimate backstop for national currencies.

But as blockchains become mainstream, the cryptocurrencies which power them will progress from the £700 billion valuation they hold today to many trillions. Unlike traditional currencies, or other assets like real estate and company stocks, the government has no role to play. Governments mint traditional currencies and back up ownership of real estate and stocks in courts of law, but they have no role to play for the blockchain. There’s a reason that countries like Egypt, Turkey, and India have banned crypto. Contrary to actually banning it, China has simply brought blockchains firmly under the state’s control.

As money flows into crypto, governments will progressively lose control over the supply of money, and therefore interest rates. What does that mean for the power of governments to manage the economy? What does it mean for the ability of governments to bring in tax revenue?

I expect cryptocurrencies operated by central banks to play a big role in the future, but more is needed to answer these questions fully.

Entrenching New Elites

Whilst existing elites like governments will feel their power under threat, new elites will rise as web3 becomes mainstream. There is a small group of people who own sizable portions of cryptocurrency. As the value of crypto grows, so will their wealth, making many some of the wealthiest people in the world and creating a new entrenched elite.

How they use the influence that comes from that position remains to be seen.

PROGRAMMABILITY.

Blockchain lets you exchange assets, and do so programmatically.

Blockchain makes it easier for individuals or groups to unilaterally make transactions, or create code that does the same thing. This empowers bad actors and good ones.

Empowering Bad Actors

Terrorists and other bad actors already have a range of ways to funnel money into their bank accounts, but crypto is another good one. Sanctioned organisations like the Iranian government and ISIS already use crypto to receive income because, for their purposes, it is more robust than existing payments infrastructure, such as Swift. But today, this is limited because crypto has little liquidity, making it harder to find people who will accept it as payment.

As web3 becomes mainstream, this will change: cryptocurrencies will become globally acceptable. This benefits everyone, including bad actors. Indeed, ISIS news sites already fund their operations with crypto donations. Few people are affected by this today, but web3 going mainstream makes this sort of illicit funding possible at significant scale. No matter how regulated web3 is, there will always be places on the internet where people who want to fund terrorist groups can use crypto to do so. The answer is to ask questions about stopping that before the source and mitigating it after delivery.

Empowering Bad Code

The critical thing about web3 is that you can write code (smart contracts) that perform actions without anyone being involved after the smart contract is published. Beyond people sending money to others using crypto, you can have people sending money to a smart contract, which could also be malicious. Part of what is propelling web3 mainstream is that anyone can launch any smart contract — but anyone includes bad actors. That’s the whole point of all every Class 2 problem: these problems happen when the good of something goes too far.

TRUST.

Everyone can verify who owns what and who exchanged what.

Environmental Cost

2022 was a big year for blockchain because Ethereum improved its energy efficiency by 99.9% by moving to the more efficient proof-of-stake consensus mechanism. Blockchains spend a lot of energy ensuring that every computer on the network agrees on who owns what. By moving from proof-of-work to proof-of-stake, Ethereum now has a lower carbon cost per year than PayPal.

Great, for now. The problem is, Ethereum and other blockchains will likely increase in scope by 1,000x or more. Despite being much more energy efficient than before, that does not make blockchain energy efficient enough for the future. And whilst further changes are planned, like sharding, to make Ethereum more energy efficient, and whilst most other chains, albeit smaller, are also cleaner, it’s worth bearing in mind that Bitcoin reached trillions of pounds worth in value, despite having enormous energy usage: energy efficiency is usually not a priority.

Indeed, environmental costs are the classic tragedy of the commons. As blockchains spread across the world, their energy cost will become significant. Whether that becomes a major environmental cost depends more on how quickly the world transitions to renewable energy than how quickly blockchains can become more energy efficient.

Privacy

Private blockchains will become common, but in the past, privacy has often been sacrificed for usability, just like the environment. As web3 becomes mainstream, there is good reason to expect these trends to persist. And whilst your name won’t be tied to the blockchain every time you buy something at Starbucks, one of your crypto wallets will be. When web3 is mainstream, and many more transactions are stored on the blockchain, there will be enough to build up an informative picture of an individual, despite associated privacy issues.

Society still has not developed new norms for privacy in the internet age, or has accepted that 20th century privacy ideals are impossible in the 21st. Questioning and updating those norms will only become more critical.

Kelly’s Class 1 / Class 2 concept is interesting and useful, but it mainly gives us a different lens to evaluate tragedy of the commons issues, rather than being a completely original concept. It is also a helpful prompt to think about how the features which make something useful can be the same reasons that make it problematic.

In applying the concept to web3, I’ve outlined a non-exhaustive selection of problems that could become increasingly problematic during the rest of the 2020s. These problems will begin as a sideshow, resulting from blockchain becoming increasingly valuable for society. And like there are significant problems with the internet, or any other new technology, we can embrace the overwhelming good and navigate the new issues at the same time.

It is crucial to rise to those problems: we will only face them because blockchain is good for the world, and for that reason, it is vital that we do.