A Family Portrait of Web 3

15 min readOct 7, 2022

A Philosophical Homage to the later Wittgenstein

0xFishylosopher

TLDR: Just play the theme song of the “Big Bang Theory” to the following “lyrics”

The whole of Web 3 was in a hot dense state, Then nearly 15 years ago Bitcoin started. Wait… The miners began to drool,
The EVM began to rule,
Smart contracts developed tools,
We built a DEX (we built the stablecoins!)
DAOs, DeFi, NFTs, unraveling all the Ponzi schemes,
That all started with the Bitcoin! (Bang!)

Introduction

What is Web 3? There are too many ways to answer this question — from a technological, economical, anthropological, or philosophical perspective. As with any field of technology, the projects of Web 3 organically evolve from each other. Ever since the launch of Bitcoin in 2009, there have been wave after wave of novel projects, each bringing fresh new ideas into the field.

As the visions of each project is laid on top of the previous like a series of transluscent photographs, the idea of “Web 3” emerges as a photographic montage of these innovations — a composite image that becomes clearer over time as new projects are added to the field. This article will aim to give a project-based examination of the most important technological and ideological innovations in Web 3 as a field, thus examining how the definition of “Web 3” has grown and evolved over time.

Though any sort of timeline is without doubt a hopeless generalization of Web 3 at large, I believe the following is sufficient to give a “family portrait” of Web 3, showing how the members of each generation have gradually grown and evolved, before ultimately discussing the structural problems in Web 3 that these projects have brought to light.

*The Family Portrait of Web 3: A Rough Sketch. Source: Original Content*

Generation 1: Bitcoin and the Blockchain

Bitcoin is the grandfather of all of crypto. Without it, none of the subsequent projects would have existed. It is Bitcoin that brought to life the concept of a “blockchain,” and allowed it to record financial transactions. At the most fundamental level, a “blockchain” is a publicly viewable, append-only data structure [1]. This means that while anyone can view all of the transactions on the blockchain, they can only write by “appending,” or adding to the end of the data structure. Once this data is written onto the blockchain, it can never be erased. This is because the data of the blockchain is available not just on one machine, but on multiple machines all over the world.

So how is new data “appended” to the Bitcoin blockchain? Simply put, Bitcoin uses a mechanism called “proof-of-work,” where participants “mine” new blocks. One miner “broadcasts” a piece of encrypted data, and the other participants must solve complex math problems (reverse a hash function) and eventually prove that the piece of encrypted data is valid. If it is valid, then the miners get “rewards” in the form of Bitcoin, deducted from a central vault called the “coinbase.” Hence, a blockchain is secured and implemented through an intricate system of cryptography, distributed systems, and game theory, all of which ensure that it becomes a permanent database [2].

The “distributed, append-only data structure” underlying blockchain technology has a natural affinity with financial bookkeeping. In an ideal “financial ledger,” you would want to make sure no one can alter previous records (and steal your money), or go offline (and lose your money). Therefore, the blockchain’s “permanent” and “append-only” attributes make it the perfect candidate for creating a reliable “distributed ledger,” as you can be sure that your money will always be there.

Generation 2: Ethereum and Smart Contracts

Ethereum was one small step for blockchain, but one giant leap for Web 3. A blockchain, being an append-only data structure, was fundamentally a way to permanently store different “data states.” Now suppose in the first block, we have the data state “0000.” In the next state, we have “0001.” Then “0010,” so on and so forth. We can create a decentralized computer from these state transitions. Ethereum set forth to do just this — to turn the blockchain into a “distributed state machine”, or a decentralized world computer, that will undergo “state changes” according to pre-defined pieces of code.

Thus, instead of being just a “distributed ledger” like Bitcoin, Ethereum is a “distributed state machine,” implemented using a Ethereum Virtual Machine (EVM) that executes the blockchain’s necessary “state transitions” [3]. Anyone can write code on this blockchain, which in turn gets compiled into binary machine code and executed in the EVM to initiate the state changes. But running this “computer” isn’t free. You have pay a fee to run code — or a gas fee, denominated in Ethereum’s coin, “ether”. As a result, “ether” is more than just a store of value like Bitcoin. It is a form of “computational energy,” a currency that has innate utility in running software on this magic global computer — almost like AWS credits.

*A Depiction of the EVM. Source:* *https://ethereum.org/static/e8aca8381c7b3b40c44bf8882d4ab930/302a4/evm.png*

This “blockchain code” is what is called “smart contracts.” What makes smart contracts unique and different from an ordinary Python script is that it can natively deal with money (cryptocurrencies). Therefore, I can use smart contracts to use code to dictate how to allocate scarce resources (money). Furthermore, this code is permanently and publicly stored on the blockchain, so everyone can see where your money goes. This therefore opens up a whole new world of financial mechanisms secured through trustless code — in a word, Decentralized Finance, or DeFi.

Generation 3: DeFi, DAOs, NFTs

The EVM was the fertile soil that truly enabled the blossoming of Web 3-native decentralized applications, or DApps. While it is impossible to give a succinct account of all of the DApps built on top of various blockchains, I will attempt to give a generalization of the different types of Web-3 native innovations that smart contracts enabled.

Arguably the most innovative and important category of DApps is DeFi, or decentralized finance applications. This is a catch-all term to refer to financial services conducted on Ethereum — lending, exchanging tokens, insurance, etc [4]. Simply put, these are bankless banks. Rather than having to trust Wells Fargo or Citicorp to act responsibly on your behalf, your transactions are handled by a piece of publicly-viewable code, which you can independently view and verify.

A paradigmatic example of DeFi is the Decentralized Exchange, or DEX, which allow users to instantly swap different cryptocurrency tokens, much like how a foreign exchange office allows you to exchange dollars for euros. The leading DEX, Uniswap, is an “automated market maker” because it does not rely on any human operator to constantly monitor and adjust the exchange rates between say Bitcoin and Ethereum. Rather, it uses a mathematical equation to automatically change the exchange rate. More specifically, it uses “x * y = k,” where “x” might be the current balance of Ethereum and “y” might be the current balance of Bitcoin, and “k” is some pre-defined magic constant. [5] Suppose I sell a lot of Ethereum to buy Bitcoin. This means that the balance of Ethereum will increase (thus decreasing Ethereum’s price), and the balance of Bitcoin will decrease (thus increasing Bitcoin’s price). Thus, an automated market maker takes the “greedy middleman” out of the picture, replacing him with a transparent, mathematical equation instead.

Another key class of smart contracts create, define, and maintain DAOs, or Decentralized Autonomous Organizations. Fundamentally, DAOs are an abstract data structure generalizing social institutions. If you look at the social institutions such as churches, governments, corporations, and schools, you can see that there the following fundamental similarities [6]:

There is a way to keep track of membership (citizenship, enrollment, HR etc.)
There is a way to determine how to allocate its scarce resources (money, time, energies)

A DAO essentially implements these two features through coding “rules” onto smart contracts, and using “governance tokens” (community votes). A smart-contract acts as an on-chain constitution, governing how a certain organization will be run, and the people running it are the ones with “governance tokens,” which have voting power. One of the most successful DAOs is MakerDAO, a community-governed lending protocol that also maintains one of the leading decentralized stablecoins, DAI (more on this later) [7]. To borrow, say $1000, from MakerDAO, you first have to pay at least 1.5x worth of Ethereum ($1500) as a down payment. If Ethereum drops and your collateral Ethereum becomes less than $1500, MakerDAO’s smart contract will automatically sell off your Ethereum, ensuring that it will be solvent [8]. On the other hand, if the price of Ethereum becomes $2000, and you manage to turn your $1000 into $1100, after you pay back the $1000 loan (plus interest) you’re able to pocket both the higher-value ETH as well as the ~$100 you earned from trading — a win-win scenario. [9]

The interest on MakerDAO loans are paid in MKR, the protocol’s “governance token.” Governance tokens are essential to DAOs, because holding these allows you to vote on important project policies. For example, using MKR you can vote on MakerDAO loans’ interest rates. The higher the interest rates on the loan, the more your own MKR will be worth (however, if the interest rates are too high, no one will borrow DAI). DAOs’ governance tokens therefore also give a certain cryptocurrency token extra utility — the “votes” and ability to actively participate in maintaining a decentralized project [10].

Non-Fungible Tokens (NFTs) are another core concept that the EVM helped bring to life. Unfortunatey, NFTs get a bad rep as people think they are nothing more than vanity plates for crypto bros. This is both true and false. Many reknowned NFT projects, such as Cryptopunks, was a simple profile picture. But that’s not the unique part of NFTs. Recall that a blockchain is a publicly-viewable, append-only data structure whose data can never be erased. This means that you can prove on the blockchain whether an NFT is authentic or a counterfeit, and who is the rightful owner of this NFT. This has never been the case before — even an expert may have a hard time distinguishing between a real and a fake Rembrandt, but anyone can prove whether your Cryptopunk is authentic or a counterfeit. Thus, NFTs represent proofs of digital ownership without the need for the backing of a centralized entity of violence (eg. property rights are guaranteed by a government).

Generation 4: Stablecoins, Yield Farming, NFT Communities

As DeFi, DAOs, and NFTs matured, they gave rise to another generation of innovative, Web 3-native concepts. One of the most significant of these is the “stablecoin,” or a cryptocurrency that is pegged to the US dollar, such that the value of the currency will always be $1. Stablecoins are important to the overall cryptocurrency economy because [11]:

People don’t need to go to centralized entities (i.e. bank) to “cash out” other crytocurrencies
People can send transactions that are independent of cryptocurrency transactions
Compatibility with traditional financial institutions and vendors (eg. using USDC stablecoin with Visa)

Stablecoins are crucial to mainstream adoption as they act as a bridge between real world commodities and traditional measures of value with the novel principles of the blockchain economy. There are several ways to create stablecoins. The most straightforward, and non-Web 3 way, is to have a centralized entity (such as Tether in the case of USDT, or Circle in the case of USDC) guarantee that every stablecoin you receive is backed up by an actual dollar in their vault. But this fundamentally contradicts the “trustlessness” nature of crypto. These centralized dealers may simply be lying to you. A more crypto-native and arguably more robust way of maintaining a stablecoin is through smart contracts.

Again, MakerDAO is the most paradigmatic example in this field. Remember that previously, you have to give in an “over-collateralized loan” of $1500 worth of ETH to get $1000 worth of DAI stablecoin. Your “over-collateralized loan” becomes the backing of the DAI currency itself. When the price of DAI drops below $1, people will buy DAI on exchanges to pay back their DAI principle at a discount, thereby reducing the supply of DAI and putting the price up. On the other hand, if the price of DAI goes above $1, then people are incentivized to “mint” more DAI by putting up their ETH as collateral. This increased DAI supply and ETH collateral will cause the price of DAI to go back to $1 [12]. DAI’s stability is therefore maintained through intricate game theory, replacing the centralized entity of violence.

Another interesting evolution in DeFi is the growth of “yield farming,” or essentially earning interest on your crypto savings. There are actually many ways to yield farm. One common way is through “staking,” where you voluntarily lock up your coins for a specified amount of time, and getting a specified amount of interest in return. Another method is to become a “liquidity provider,” where you provide your coins as “usable liquidity” to a DEX (such as Uniswap) and in return earn a portion of the DEX’s transaction fees. One of the most interesting “yield farming” projects is Yearn Finance, which some describe as an “Amazon marketplace for interest-bearing crypto products.” [13] As a DeFi aggregator, you have the ability to shift your money around all these different DeFi projects, like a build-your-own-taco for DeFi. The ability for DeFi aggregator projects such as Yearn Finance to thrive attests to the explosive growth in both the DeFi model and the field in general.

Finally, as time has worn on, NFTs too have become more than just a “vanity plate for crypto bros.” Instead, NFTs have become entry portals into niche, like-minded communities. Bored Ape Yacht Club and Azuki, for example, both have a strong online community of its holders. Given that you can always easily prove whether an NFT is authentic or not, there is no way to “fake” membership in these NFT communities. Essentially, NFTs have also become like governance tokens, a way to build DAOs and track community membership. The merging of governance tokens and NFTs is most apparent in Yuga Lab’s Apecoin airdrop, where a governance token (Apecoin) was airdropped to all BAYC NFT holders [14].

Generation 5: StepN and X-To-Earn

Step by step, Web 3 has evolved from a novel data structure to a socioeconomic realm that looks completely foreign to the traditional tech models. One of the most recent developments is the idea of “X-To-Earn,” such as the “move-to-earn” idea popularized by StepN. Personally, I believe that the StepN epitomizes both the perils and promise of Web 3 as an industry. Built on top of Ethereum-competitor Solana, StepN became the top-app on Solana, once accounting for over 20% of Solana’s fee-paying users [15].

The basic model of StepN is this: first, users buy a sneaker NFT, which allows them to enter into the “move-to-earn” game. Essentially, this sneaker NFT is an entry card into StepN’s game. Then, when you are walking or running, StepN uses a pedometer to track your footsteps, and you are awarded with the in-game token, GST, according to how much you have moved. However, over time your sneaker NFT will have “wear-and-tear,” and you have to spend GST to repair your sneaker so that you can continue to earn. You can also use GST to upgrade your sneaker in order to gain more rewards. Once your sneakers reach a certain level, you can earn the “governance token,” GMT, which will can be traded on real-world marketplaces for real money. Alternatively, you can also stake your GMT in order to gain voting power for the governance and ecosystem fund. [16]

In many ways, StepN epitomizes the evolution of Web 3 principles and incentives. It aligns a socially-productive activity (running) with an incentive mechanism from which people can ostensibly earn passive income by doing something they love. It seems to be a pioneering model that solves the problem of “positive externalities,” providing a monetary subsidy to people who do socially productive things. Furthermore, StepN also appears to use the principle of “DAO governance” to incentivize active users with governance potential and thus allow for the gradual decentralization of the community. Indeed, StepN’s immense popularity definitely attests to the innovative appeal of “X-to-Earn.”

But far from being a magical cure to the problem of positive externalities, StepN veers dangerously close to an elaborate Ponzi scheme, where the “earnings” of players essentially come from the sneaker NFT sales of the new participants. The “X-to-earn” model only works if there is more money put in than money taken out [17]. And currently, the largest way to “put money in” is through sneaker sales (the entry of new participants). Of course, StepN could do partnerships with sports leagues and brands. But the premise of advertisement is that the advertising company gets more money than it spends. Herein lies StepN’s dilemma: if advertising does indeed become sustainable in the long run, users’ on net aren’t actually “earning.” Rather, they are spending, whether directly or indirectly, and paying advertisers. On the other hand, if the external avenue of advertising is shut down, StepN’s “earning” either becomes an actual Ponzi scheme, or a “zero-sum game” where experienced players essentially take the money of inexperienced players. Either road leads to a toxic path.

Furthermore, the principle of gradually-decentralizing through the governance token of GMT is also dubious at best. Asking a startup to decentralize power to a community is like asking a dictator to abdicate power to a democratic assembly. Unless there is an external catalyst, this is very unlikely to happen. Giving participants GMT tokens does not equate to giving them power, or even, in fact, value. What is to prevent the founders of StepN and/or similar “X-To-Earn” projects to do a “pump-and-dump,” artificially inflating the token price before selling all their tokens? Can the community of GMT token holders really stop this from happening?

Given that many “X-to-Earn” projects essentially just clone StepN’s economic model to different fields, I believe that StepN epitomizes both the innovative brilliance and profound perils that plague the blockchain economy today [18]. NFTs, DAOs, DeFi and other principles must one day be aligned with real-world activities and offline items in order to succeed in the long run. And I trust that it will.

But the question is, how?

What’s Next?

As shown above, technology never stops evolving. As talent and cash continues to pour into Web 3, there are bound to be slates of new conceptual innovations on the horizon — ones that no one can possibly predict. But Web 3 has a fundamental structural dilemma that must be addressed sooner or later.

There is but one philosophical question in Web 3, and that is one of centralization versus decentralization. All others, whether NFTs are scams or the future of digital assets, whether there should be one or two or five or fifty viable blockchains in the Web 3 ecosystem, come thereafter.

The fundamental reason why centralization exists is that economies of scale are everywhere, especially in the digital economy. It is far easier (for maintenance, debugging, etc.) to put two servers together, rather than keep them apart and decentralized. It is easier for one machine to keep track of all records rather than solve the complex Byzantine Generals Problem of distributed systems consensus. Of course, engineers are known to be lazy and find solutions that work but are not necessarily perfect. Decentralized systems, such as MakerDAO’s DAI stablecoin, are much harder to invent, implement, and maintain, than their centralized counterpart, such as USDC stablecoin. Even when such systems are created, they are often less efficient than centralized equivalents. Compare, for example, Ethereum’s transaction speed to even the cheapest AWS instance.

There is a distinct incentive for scrappy startups to hack together centralized prototypes in the name of speed and utility, capturing market share and worrying about decentralization later. Think about all the companies, even established (and highly decentralized) projects such as The Graph, that put out a centralized service first, before planning a decentralized product. But once you have a centralized entity taking advantage of all these economies of scale, why should it abdicate its power to a decentralized DAO? Caesar Augustus never gave back his power once he became the sole ruler of Rome. So why should StepN? Or for that matter, even Avalanche or Solana? ****Not every project has a founder like Satoshi Nakamoto or Vitalik Buterin who will willingly sacrifice their own gains for a still-blurry long-run promise. Yet the long-run promise of Web 3 is to create a decentralized, inclusive ecosystem where there is not one single player dominating the market. When push comes to shove, decentralization after centralization goes against human nature.

Other secondary structural problems Web 3 faces include the dilemma between utility and speculation, as well as how to deal with the status quo — of governments, laws, and big tech. As time goes on, I will endeavor to embark on a philosophical investigation into the fundamental philosophical implications of both cutting-edge Web 3 technologies, such as zk-Rollups, as well as attempt to explicate the ideology of Web 3 writ large. Too many articles talk either only about the technology and skim over the philosophy, or make empty philosophical talk without explaining how these concepts are realized. In my writing, I will try my best to put these ideals side by side, looking at Web 3 through both the lens of technology and philosophy.

So let’s begin on this journey.

🐦 @0xfishylosopher

📅 07 October 2022

Many thanks to my friends at Moonshot Commons for consulting and sponsorship for this article.