Michael Bottjer

A Primer on Web3


Digital assets are at an inflection point for mass adoption and represent a rare opportunity for investors. Tim Berners-Lee brought the internet to life in 1989, introducing the world wide web. At the time, some believed it to be a fad whilst others believed it would be yet another distribution channel, in addition to newspapers, radio and television. As we know, the internet became the infrastructure through which media is distributed, and in fact gave birth to new forms e.g. social media.

Web3 technology represents a natural progression, enabling digital property rights and the distribution of public goods, critical pieces of global infrastructure, via the internet. These technologies, NFTs and fungible tokens, permit economic activity to flourish natively online. Labor and capital will continue to flock to the internet and the apps built on top will look completely different from Web 2.0 businesses. With this adoption comes the need for infrastructure layers to support it and this is where investors should focus, leveraging the expertise of software engineers, quantitative researchers and tech investors to identify the crucial building blocks underpinning this technology’s continued growth.

As of January 2022, there exist 16,000+ cryptocurrencies, digital tokens that leverage the decentralized and tamper-proof environments of blockchain networks to facilitate highly secure transactions. These represent a total market capitalization of ~$1.5T. These tokens vary considerably in both quality of their design and utility. Most importantly, the pace of innovation necessitates active management. To best illustrate this let’s compare 2017’s top 10 cryptocurrencies by market capitalization to 2021:

 #  2017 2021
 Bitcoin Bitcoin
 Ripple Ethereum
 Ethereum Binance
 BitcoinCash Solana 
 Cardano Cardano
 nem Ripple
 Litecoin Terra Luna
 TRON Polkadot 
 Stellar Avalanche
10   OTA Dogecoin

The transition represents a shift from highly speculative and mostly non-functional tokens to time tested functional chains: Bitcoin and Ethereum. In fact in 2017, Bitcoin accounted for 70% of market share and has subsequently reduced to 40% whilst Ethereuem increased from 14% to 20%. As we enter 2022, Arceau will capture exposure to projects that prioritize user ownership of data and/or assets and interoperability between distinct applications.

We believe Web3, an internet owned by the builders and users, orchestrated by tokens, will far exceed the internet as we know it.

Introduction to Web3

Web 2.0, a term coined by Tim O’Reilly and others around 1999, captures a progression from static desktop web pages delivering users content via costly servers, to user-generated content and interactive experiences like Youtube, Facebook, Uber and AirBnB. Web 2.0’s growth can largely be attributed to the growth of three innovations: mobile, social media and the cloud. The beginning of the ‘always connected’ state was marked by the launch of the iPhone in 2007, with mobile internet access, came broader adoption and our reliance on the internet for everyday life; our browsers and applications could now be found in our pocket. Social Media coaxed users into generating content and provided an online identity, an identity which we use to entrust unknown travelers with our homes or grab a ride with a stranger. Cloud commoditised the production and maintenance of internet pages and applications. No longer did companies need to buy and maintain their own expensive infrastructure, they could rent storage, compute power and software tools as they saw fit. Entrepreneurs have been empowered to experiment with low-cost resources that scale with their operations, laying fertile ground for innovation. Whilst Web 2.0 will continue to sprout opportunities, the next paradigm shift in internet applications is emerging coupled with ~$27B in venture capital in 2021.

Web3 (originally coined the Semantic Web by Tim Berners-Lee), is a fundamental disruption to the status quo; it is a leap forward to open, trustless and permissionless networks. Open in the sense that these networks are built from open source software by open and accessible communities of developers, executed in the public domain. They are trustless in that no third party is required to interact either publicly or privately with the network. Permissionless by the fact that anyone can participate, there is no need for a governing body. Web3 is being driven by the advent of edge computing, decentralized data networks and artificial intelligence. A multitude of computing devices across phones, sensors, computers, appliances and vehicles are augmenting the legacy data centers we rely on today. Decentralized data centers are enabling these data generating devices to transact with no loss of ownership, privacy or reliance on an intermediary.

Web3 permits a future in which users and machines are able to interact with data via a substrate of peer-to-peer networks without the need for third parties.

The movement we have seen build up over the past few years goes beyond Bitcoin and other crypto-assets, or even open source software and blockchains. Taking a broader view, it is the victory of peer-to-peer data networks based on open standards, it reflects the power of properly aligned economic incentives, and it begins to harness the individual data centers in everyone’s pocket, desktop, car, living room and wrist. The movement is made possible by the proliferation of access to high speed wireless broadband, rapidly maturing cloud-native software and a surge of recent machine learning advancements. Over the past few decades, technology architecture advancements have commoditised operating systems and software packages by making them globally accessible via data centers and cloud infrastructure. In this new wave, the data center is being spread to the very edge of the network and the data itself is being ‘open-sourced’, commoditised into reusable trusted building blocks. Distributed users and machines interact with this data via a substrate of peer-to-peer networks. These peer-to-peer data networks become the foundation that validates and curates information inputs without the need for third parties, while empowering individual users with their own data in a usable, secure and scalable manner.

Web3 Current Pitfalls

For a decentralized application (dApp) to communicate with the Ethereum blockchain, the development team needs to run and manage an Ethereum node. To avoid the complexities of doing this, two companies: Infura and Alchemy, have abstracted these difficulties to enable developers to focus on their application logic. These companies run Ethereum nodes-as-a-service, providing both analytics and the ability to introspect historical transactions. When dApps make an API call to these services, there is nothing to verify the blockchain state, they simply accept the response; putting into question the trustless distributed consensus mechanism we expect when interacting with the Ethereum blockchain. The analogy that can help us best understand this is what if everytime you interacted with Chrome, your request first went to Google and was then routed to the desired location and then back to you. To see why this matters, let’s consider NFTs. Not all NFTs are created equal; most would be too expensive to store on-chain and hence, a creator must decide how to reference their raw NFT asset. They create a link between the blockchain native token and the associated data (a JPEG for instance). Historically, creators might have used location-addressed URLs pointing at centralized servers. These “normal” URLs however, can be susceptible to issues such as broken links, 404 errors, or a bait and switch. The creator of Signal (an innovative privacy chat app) went as far as to create an NFT that changed dependent on the url the user visited:

His NFT was later removed with no warning by OpenSea. Once OpenSea removed the NFT, it no longer appeared in his wallet, alarming. Wallets like MetaMask are non-custodial and in fact provide a way to view data via an API call to mostly centralized entities, in his NFT case it was OpenSea. These calls aren’t authenticated nor are they signed; OpenSea is free to execute its discretion and remove the image from existence. A purchaser of said NFT would be left with a reference URL and nothing more.

There are solutions to the above problems and we expect as this space matures, to see further abstractions to the complex realities of launching and maintaining decentralised infrastructure. In the case of NFTs, creators have turned to IPFS which uses content addressing to reference data. When an NFT Creator uses IPFS to reference data, they create a unique fingerprint (CID) of the content itself. This unique fingerprint can replace traditional URLs – anyone can find the data associated with this fingerprint just by using the CID to request the data. When using IPFS content-addressed URLs in NFT metadata, creators can ensure that there is an immutable link to their content embedded on the blockchain.

Immutability is often touted as a core feature to blockchains, once something is written to the blockchain it can’t be removed or modified. A transaction is final. However, we need only look at the early days of Ethereum to understand why this is not necessarily the case. The first Decentralized Autonomous Organization, aptly named ‘The DAO’, was exploited by attackers in June 2016, stealing 3.6M ETH in the process. To avert this crisis, the Ethereum blockchain was hard forked, which permitted the lost funds to be recovered. In December of 2021, Polygon, a hugely popular layer-2 solution performed a ‘silent’ hard fork after discovering a significant vulnerability, declaring their decision after the fact, highlighting the centralized nature of the protocol. More worryingly, it is believed that Polygon is reliant on the skill, honesty and integrity of 5 of the 8 multisig keyholders, a far cry from programmable trust.

Another core concern is the over-reliance of the crypto ecosystem on Amazon Web Services, which accounts for 32% of the cloud computing market and hosts 90% of the largest gaming companies servers. On the 8th of December, at around 10:30 am EST,  AWS went down for almost 8 hours. The decentralized derivatives exchange dYdX was down for most of the day. Though claiming to be decentralized, and illegal for U.S. citizens to access, core servers in the dYdX protocol run on East-1, the network of AWS servers that halted. Events like these reveal the critical infrastructure flaw of centralization that spans across all industries. In fact roughly 60% of Ethereum nodes are running on centralized cloud services of which 25% are operated by AWS. While Amazon may have an edge on cheaper and better cloud services, moving away from these centralized companies will result in a more robust tech ecosystem, less conflict of interest (advertising) between consumers and producers of information, and simply, a better version of the internet for all. 

These platforms are nascent and constantly being improved, with active participants incentivized to act in the best interest of their communities. However, as we have shown, many solutions today remain ‘band-aids’, sacrificing security and decentralization. At Arceau, we are interested in projects providing long-term sustainable solutions that maintain the ethos that distinguishes Web 2.0 from Web3.

Investment Thesis

In 2021, all risk assets were buoyed by loose fiscal and monetary policy, this was particularly the case for crypto, its market cap increased by 300% YoY. However, this hyper growth has also been driven by cryptocurrencies finding real use cases:

  • DeFi (Decentralised Finance): expanding the use of blockchains beyond the simple transfer of value to more complex financial use cases including: loans, insurance, crowdfunding, derivatives, betting and more. It represents ‘open finance’, replacing intermediaries / gatekeepers with ‘code-is-law’ to improve both the speed and sophistication of transactions.
  • NFTs (Non-Fungible Tokens): NFTs represent verifiably scarce, portable and programmable pieces of digital property. An NFT could be a share of a stock, a virtual weapon in a game, a profile picture or your data record on social media. It’s still early days; however, it’s not hard to imagine a world in which your property deed, a verifiably scarce item, could take the form of an NFT. You will prove ownership to your insurance company by signing a transaction in your wallet that holds this deed. In this vein, Centrifuge is teaming up with Aave to enable the tokenization of real assets dubbed RWA (Real World Assets), bridging assets like invoices, real estate, and royalties to DeFi.

The total value locked (TVL) into DeFi applications has increased ~1600% YoY, that is over $250B in assets deposited in DeFi protocols earning rewards, interest, new tokens etc. The TVL represents the spot value for all assets currently staked in protocols, there are three key types:

Lending and Borrowing Protocols

Users lend assets to liquidity pools and enable users to borrow funds from these pools. Well-known examples include Maker (MKR), Aave (AAVE) and Compound (COMP). For these apps, TVL refers to the total value held in the borrowing and lending pools.

Decentralised Exchanges (DEXs)

These protocols allow users to trade cryptocurrencies using an automated market maker whereby a user is transacting with a liquidity pool that holds the requisite pair of cryptocurrencies. The leading examples of these protocols include Curve Finance (CRV), UniSwap (UNI), and SushiSwap (SUSHI). On these platforms, TVL refers to the total value held in all of the coin-pair pools of the protocol.

Yield optimization protocols

These applications apply algorithms to optimize a user’s investment across lending and borrowing platforms, acting as an automated portfolio manager. Users deposit their funds into the protocol and the algorithm allocates funds to the pools with the best-expected returns. The biggest among these protocols at the moment are Convex (CVX), Yearn.Finance (YFI), and Balancer (BAL). On these protocols, TVL normally refers to the value lent by users to the app for yield optimization.

The market cap for NFTs, solely on the Ethereum Network (ERC721), has reached more than $9.5B. Both DeFi and NFTs have been enabled by second generation smart contract layer 1 blockchains and layer 2s, which leverage Proof of Stake consensus algorithms, and as a result have seen their total value rapidly increase, reaching a market cap of ~$280B.

Moving into 2022, we have seen monetary conditions tighten. The decrease in purchasing power we are facing coupled with tapering will result in a worsening economic system. However, on the Web3 side, we are still early, the tens of billions that have flowed into funding new projects will only start to blossom in a couple of years, as the projects launch publicly. This innovation cycle has happened before, it will happen again and is best summarised by:

This cycle, with its influx of developers, is highly promising. In fact, the number of Web3 developers is captured by Electric Capital’s developer report:

Timing however, is everything and to better frame where we intend to gain exposure we look at the past two waves of exponential tech: internet and mobile. They gave us the heuristic that mass-market applications get major traction once the underlying technology reaches 1 billion users.

Right now, there exist 180M unique Ethereum addresses, a proxy for adoption of Web3. At the current growth rate we anticipate it taking another 5 years to reach 1 billion users. Hence, we believe that at this stage of adoption, outperformers in this area will be:

  1. Infrastructure e.g. enabling secure cross-chain activity, bridging real world assets to DeFi
  2. Niche applications with a high value add e.g. blockchain games

Our Approach

Web 2.0 enabled new methods of monetisation to flourish on top of new tech platforms; however, this time around, decentralized data networks allow monetisation not only within the application layer but also within the base protocols on which these apps are built. Hence, a lot of the early value creation will be concentrated around the infrastructure layers; tools built by developers for developers. Our craft as investors is focused on sourcing projects from developers, inspecting Github commits and tracking developer activity across relevant repositories. Where developers go, new protocols and projects come to life, attracting users and subsequently increasing the value of the underlying network and its token (assuming the token has utility).

Cross Chain Solutions

Bitcoin and Ethereum’s historical dominance have meant that there has been little need for cross-chain asset transfers / interoperability. The advent of new layer-1 & 2 solutions show that this is changing fast. For clarification, a layer-1 network, for example Ethereuem, Bitcoin and Solana, is a blockchain whilst layer-2 solutions are third-party protocols that can be used in conjunction with a layer-1 blockchain e.g. Polygon and Arbitrum. Transportation across crypto ‘nations’ and ‘cities’ will become the next infrastructure challenge; without interoperability we cannot facilitate widespread blockchain adoption and fulfill the promises of Web3 as data remains siloed to a given layer-1 network. This is why we like projects such as Polkadot, for which the number of developers contributing to the ecosystem has grown faster than that of Ethereum.

Proof of Stake Layer-1 / Layer-2

Mass adoption of Web3 is reliant on scalable publicly accessible blockchains. We see this in action already as next generation L1s are reaping the rewards of enabling massive growth of on-chain applications. This will continue into 2022, and newer layer 2s on top of Ethereum will continue to compete. At Arceau, we believe the risk / reward can be sorted as such:

New L1s > L2s on Ethereum > Ethereum L1

Blockchain Gaming

The gamer demographic has a significant overlap with early crypto adopters. Gamers already want player-owned in-game assets, the natural use case for NFTs. Couple this with a weaker global economy, we have an increased appeal for players to make money from gaming via newly created play-to-earn games like Axie Infinity and Star Atlas. High performance games are now possible thanks to these second generation layer 1s and 2s. What is still lacking is high quality blockchain games that true gamers want to play, with high quality games will come sustainable tokenomics.


  1. https://moxie.org/2022/01/07/web3-first-impressions.html
  2. https://blog.mollywhite.net/blockchains-are-not-what-they-say/
  3. https://medium.com/electric-capital/electric-capital-developer-report-2021-f37874efea6d
  4. https://panteracapital.com/blockchain-letter/the-year-ahead-in-crypto/
  5. https://www.theblockcrypto.com/post/128784/polygon-critical-bug-24-billion-matic-tokens-at-risk-hard-fork.
  6. https://defiwatch.net/inquiry-into-polygon-multisig/
  7. https://coin.fyi/news/ethereum/more-than-60-of-ethereum-nodes-run-in-the-cloud-mostly-on-aws-d8e72f#
other posts
Robinhood’s PFOF Simplified
Robinhood pioneered a new methodology, by partnering with a wholesaler, for example Two Sigma Securities LLC (TSS)...
learn more