#3: Blockspace - best product of the 2020s
Exploring the heartbeat of Web3
A16z’s head of crypto, Chris Dixon, has called blockspace the best product of the 2020’s. A bold statement, particularly as this may be a decade where we need to develop solutions to prevent global warming reaching a point of no return and techbio revolutionises drug discovery and development. So what is blockspace? In a nutshell, blockspace is to Web3 what Visa and Mastercard networks are for payment processing. Chris describes blockspace as “a space on the blockchain that can be used to store information and run code. Critically, it differs from traditional computing space because the hardware is subordinate to the software, the blockchain code”. Its promise rests on a future where all economic activities on public blockchains settle on the blockspace, with consensus producers supplying blockspace and every transaction demanding blockspace.
Being described as the best product of the decade by one of the most prominent web investors got me intrigued to dig deeper. So in this post, we’ll look at:
Status of blockspace today
Scaling blockspace
Future of blockspace
Status of blockspace today
Blockspace is often referred to as the commodity that powers the heart of all crypto networks. Blocks are important because they are the global banking ledger, and as a consequence act as a civilizational trust anchor in the world of Web3. Blockspace are sold by blockchains, whose single purpose is to make its blockspace more valuable. The most valuable blockspace today is by far Ethereum. Based on 7-day avg. gas fees (see below) from block sales, the market cap of Ethereum’s blockspace is almost 9x higher than Bitcoin and Solana combined.
How does the blockspace market work? The most common participation dynamic of blockspace markets is a repeated first-price auction. The principal actors are the users (demand), miners (supply) and mining pools (auctioneers). Mining pools connect users and miners together. Users submit a bid (gas fee) for their desire to include a transaction in a block, whilst miners compete for blockspace (e.g. by solving a complex mathematical puzzle) and receive the fee for processing and including the transaction on the blockchain.
Gas fees are typically a function of i) a base fee (minimum amount of gas required to include a transaction in the next block) and ii) a tip (i.e. priority fee, to get ahead of the queue). Though, empirical evidence suggests that over 75% of transaction processing follow a default strategy without any prioritisation - meaning that users rarely bid up only to get to the front of the queue. The base fee is determined by four factors: i) miner supply, ii) user demand, iii) block size (transactions per block) and iv) the average block time (how long it takes to generate a new block). Generally speaking, the higher the user demand the higher the base fee. However, the latter two factors are why the supply of blockspace does not increase as more miners participate (side note: they do still help improve the security and reliability of the blockchain). Moreover, the transaction speed is different for every blockchain and depends on how the blockchain has been constructed. The transaction speeds for some of the most popular blockchains today are:
Ethereum - 20 transactions per second, avg. confirmation time 5 mins
Expected to increase to 24,000 txn per second with Ethereum 2.0;
Bitcoin - 70 transactions per second, avg. confirmation time 10 mins;
Solana - 50,000 transactions per second, avg. confirmation time 5 mins;
Algorand - 1,000 transactions per second, avg. confirmation time 5 secs;
Visa (for reference) processes ca 1,500 transactions per second and is reportedly able to handle 65,000 transactions per second.
Scaling blockspace
Why aren’t blockchains constructed in such a way that they can process an unlimited amount of transactions every second? The challenge is due to what Vitalik Buterin calls the blockchain trilemma; when designing a blockchain developers have to choose two of the three elements; decentralisation, scalability and security.
Decentralisation is about to what extent the control of the blockchain is distributed to the public vs centrally controlled;
A protocol also must be secure and protection against Sybil attacks (51% takeovers), minting attacks, and DDoS attacks etc. to keep the network running 24/7;
Scalability is about the processing power and the limits on the transactions a blockchain can process.
Decentralisation is a fundamental promise of Web3, whilst security is a must have as no user or miner would be interested in spending resources and time on blockchain applications or tokens if the risk of theft is high. This often means scalability is left out, since generally higher transaction speeds means higher security risk.
The inherent scalability constraint of popular blockchains like Ethereum and Bitcoin is often used as a reason by detractors to explain why Web3 will never be able to replace Visa/Mastercard networks. However, there exists ways to circumvent the scalability constraint. One of the more common ways today is by creating L2 solutions. This method is used by Ethereum. These solutions typically process transactions off-chain, allowing them to be processed faster before being brought back to the mainnet. According to CEX.IO, ca. 35% of all daily ETH transactions are now facilitated by two L2s: Optimism and Arbitrum - both use an off-chain computation to roll up transactions.
One of the main reasons as to why L2 solutions are popular is because these solutions typically inherit the security properties of the L1 blockchain, i.e. solutions like Arbitrum and Optimism inherit the security and decentralisation of the Ethereum blockchain. However, as seen in the image above there are different approaches to build L2 solutions for addressing the scalability constraint in Ethereum (note: these are general approaches and work for most, if not all, blockchains). The most popular ones today are rollups, sidechains, state channels and validiums.
Rollups
Solutions that rollup/bundle multiple transactions into one transaction on the L1 chain. It does this by performing the transaction operations off the L1 chain whilst posting the transaction data on the L1 as a proof / record of the transactions. There are two types of rollup solutions: zero-knowledge (ZK) and optimistic. They differ in their security measures. ZK rollups do computation off-chain and submit a validity proof to the main-chain. In contrast, optimistics rollups assume the transactions are valid by default and only perform computations in the act of challenge. This provides increased transaction speed and lower gas fees compared to ZK rollups. However, ZK rollups offer faster transfers (e.g. withdrawals), since the computational verification is done upfront. Optimism and Arbitrum are both optimistic rollups.
Sidechains
Sidechains are completely separate networks connected to the L1 blockchain. They process transactions on its own chain before relaying the transaction to the L1 network. Unlike other L2 solutions, sidechains do not inherit the security from the L1 chain. As they are independently operated chains, they have their own consensus protocol, block size and security. An interesting consequence is that this structure enables sidechains to often also serve as bridges for connecting two separate L1 chains.
State Channels
State channels are primarily used in relation to payment applications. Similar to a rollups, it tracks multiple transactions from users but instead of reporting every single transaction back to the L1 chain, state changes preprocess and aggregate the all the transactions into just two transactions. This is different to rollups who aggregate and report every transaction into one block. To illustrate, if $100 is sent back and forth between two people 10 times, rollups will record 10 transactions onto the main chain. However, with state channels, only the final amount each user possesses is reported back to the L1 channel - requiring only two transactions to be reported back.
Validium
Validium is a zero-knowledge proof solution that processes transactions off the main chain. Validium solutions enable faster and cheaper transactions by not storing data on the L1 chain. Instead validiums validates transaction operations with smart contracts on the L1 chain. This sacrifices some network security, but the existence of validity proofs, gives validiums higher security guarantees than other pure off-chain solutions like sidechains.
Other approaches
Beyond the L2 solutions presented, there are also other alternatives for scaling blockspace; sharding, building additional blockspace directly on the L1 chain (like Solana), and creating more L1 blockchains or multi-chains.
Sharding is the process of splitting a database horizontally to spread the load. By splitting the main chain horizontally into smaller side-chains (shards) transactions can be verified in parallel. Also, each shard only needs to validate part of the main chain rather than the entire main chain. This helps improve transaction speed and reduce congestion. Ethereum is expected to implement sharding in the near future, which should help increase its transaction per second speed to ca. 100,000, up from 20 transactions per second today.
Solana is an interesting case study as they take a different approach to solving the scalability issue. Solana is solving the blockchain trilemma by employing a proof–of-history consensus for processing and validating transactions. In other consensus mechanisms miners are required to communicate with each other to verify the timestamp of a transaction. This takes processing power and a lot of time. In contrast, a proof-of-history protocol creates a historical record that proves that an event has occurred at a specific moment in time. Per Anatoly Yakovenko, cofounder of Solana, “it chains messages from miners about the validity of blocks together to provide a relative chronological order of events that is not dependent upon local clocks or timestamps.” This ensures predictability and allows miners to use the information encoded to determine if a transaction is valid or not, allowing Solana to process transactions very quickly. Moreover, it eliminates the need for sharding or L2 solutions and enables applications & blockspace to be built onto the L1 chain directly.
Another solution to scale blockspace is by creating more L1 chains. In fact, if we look at the short history of web3 this has been the main approach to date. First, we had Bitcoin, then we got Ethereum and today we have 100s of different L1 chains offering blockspace for validating and processing a multitude of different transactions. However, a downside with this approach, and also a limitation of the above mentioned solutions, is that they are all monolithic-chain solutions. That is, they enable scaling for one specific chain (and admittedly helps make that chain more valuable) but does not enable communication with other L1 chains.
Multi-chains are a solution to this. A multi-chain is a ‘meta’-chain where you can build other L1 chains that are interoperable by default, allowing scaling across multiple interacting L1 chains. The most prominent multi-chain today is Cosmos which uses a proof-of-stake consensus algorithm called Tendermint. It can be connected to using a protocol through the Cosmos SDK. At the centre of the Cosmos blockchain sits the Cosmos Hub, the first Tendermint chain. It provides services and security to Cosmos’ extended network of blockchains. Thus, applications do not need to compete for blockspace in an increasingly congested marketspace. Instead they can create their own chain for deploying and processing their services, and which is interoperable by default.
Future of blockspace
What does the future of blockspace hold in store? There are currently two competing views; i) all activity will converge onto a single monolithic blockchain or ii) the future is multi-chain consisting of many interoperable but specialised blockchains tailored to different purposes (payments, minting NFTs etc.). Up till now, monolithic blockchains have received the most attention and investments. They have also generated the best returns, in particular Ethereum and Solana. Benefits of the monolithic approach are i) a lower cost of deployment - only need to create and implement a smart contract, and ii) synchronous composability - the ability for applications / smart contracts to interact with each other within the same block (e.g. borrow and repay a flash loan in a single transaction).
Benefits of the multi-chain view are i) customizability and ii) more predictable gas costs. To launch an application on a monolithic chain one must accept the L1 chains design decisions on the validation model, security, runtime and so on - this can all be customised in a multi-chain world by simply deploying your own chain (which would be interoperable by default). Additionally, in a monolithic world, applications are competing for finite blockspace with other dApps. This can result in high uncertainty in gas costs, particularly if blockspace demand fluctuates significantly (compare gas fees between crypto summer and crypto winter). A multi-chain view avoids this by expanding the blockspace universe by simply creating more L1 chains.
Though monolithic blockchains have enjoyed the most attention to date, there is an increasingly growing view that a multi-chain world is the future. Ethereum in particular is already today struggling with blockspace congestion, which has given rise to 100’s of new L1 blockchains, including Solana, Avalanche and Algorand. At some point in the future, there will be demand for connecting all these chains to help services on the chain increase availability. Applications are today already starting to launch on multiple chains in an effort to increase their reach. The rise of L2 solutions such as rollups and side-chains on Ethereum is also a first step of Ethereum move towards a multi-chain future. There have also been developments in cross-chain bridges like Axelar to help address this issue. A cross-chain bridge solution would enable multiple monolithic chains to co-exist, however, significant security concerns (evidenced by recent hacks) have resulted in scepticism against this approach.
Multi-chains look like a better solution, and its disadvantages to monolithic chains - higher cost of deployment and lack of synchronous composability - are becoming less profound. The latter is arguably only necessary for a few DeFi use cases and the former is becoming less of a disadvantage as multi-chains matures. Cosmos is a multi-chain that is starting to get increased attention from developers and investors, and Vitalik has also publicly thrown his lot into the multi-chain world view.
A Soviet astronomer, Nikolai Kardashev, proposed a three stage scale for measuring civilisations technology by energy use:
In a Type I civilization can harness the energy of its home planet;
In a Type II civilization can harness the energy of its home star;
In a Type III civilization can harness the energy of its home galaxy.
Ryan Sean Adams presents a analogous measure for blockspace, calling it the Polyna scale:
In a Type I chain individuals buy blockspace;
In a Type II chain applications buy blockspace;
In a Type III chain other chains buy blockspace.
Almost all of today’s blockchains are Type I chains - their blockspace demand comes from individual users. Ethereum is the exception, with applications consuming the most blockspace. Though, the most advanced block-chain structure is a multi-chain world. If blockspace is to become the new oil, muli-chains may be the key to unleash the potential of web3 - creating a vibrant ecosystem of networks allowing users and builders to interact irrespective of the L1 chain of entry. If realised, blockspace may indeed become one of the best, if not the best, products of this decade.
Better Reads :)
Blockspace: An Introduction with Chris Dixon - The Generalist
What to watch in crypto - The Generalist
The best product of the decade - Bankless
Etherum Blockspace - Who Gets What and Why - Paradigm