Gone are the days when blockchain didn't scale. A modern blockchain can handle a huge number of transactions per second (TPS). Blockchain that does not scale well is lost and will soon disappear. Sharding and parallelization are the way to go. The future is bright. This narrative is getting stronger. Many influencers overlook decentralization in their obsession with high TPS. Thus, the essence of the blockchain industry is being forgotten. How to combine scalability and decentralization? Let's think about it.
To be clear, the modern blockchain needs to scale better than the first generation of blockchain. It makes no sense for the network to divide users into first-class and second-class citizens because of outdated technology.
All users must have the same status. The division between rich and poor creates inequality in the system that will be an obstacle to adoption. The poor will not adopt a system in which the rich have advantages.
If the blockchain ecosystem does not provide cheap onboarding of new users as well as fast and cheap payments, it has no chance to compete with traditional financial systems. For DeFi services and all other use cases, scalability is essential.
Knowing the above, it is necessary to return to the roots of the crypto industry and remember that the key feature is decentralization.
Decentralization is often forgotten in the effort to deliver a network with huge TPS to the market. The priorities for each blockchain should be in this order:
Each team balances these characteristics differently. That's fine. One team prefers decentralization over scalability. The other prefers scalability over decentralization. Blockchain can be highly scalable and only moderately decentralized. It will still be useful.
The problem is that teams that neglect or underestimate decentralization at the beginning may not be able to improve it later.
Both decentralization and scalability are properties that are not static. Teams, along with communities (node operators), will work to maintain and improve them.
If we wanted to describe decentralization and the conditions that a given project must meet, it could look like this:
- A high number of block producers.
- Consensus participation must be inclusive.
- The initial coin distribution must be fair.
- There must be some form of governance, ideally based on resource ownership (coins or hash rate).
Let's describe the points in more detail.
Components of Decentralization
When Satoshi launched Bitcoin, he was the only block producer (miner) on the network for a while. The network was centralized at the beginning, but gradually new miners joined. Bitcoin's decentralization has gradually grown.
Although Bitcoin only has 20 block producers these days, everyone has the opportunity to join mining. It is possible to buy ASIC hardware and delegate the hash rate to one of the pools.
An important component of decentralization is inclusiveness. People must be able to participate in the network's consensus and be rewarded for doing so.
Decentralization is primarily about increasing the number of participants. This is about distributing decision-making power among the largest possible number of people who are interested in participating.
Networks with high TPS tend to have high demands on block producers in terms of hardware. It is often combined with the need to hold a large number of coins. The team wants the network to have a smaller number of block producers with high-quality equipment.
The exclusivity is a result of the high TPS requirement. Communication between fewer nodes is faster.
Every network user cannot be a block producer. Nevertheless, this option should be open and relatively accessible. Nowadays, some networks have on the order of hundreds to thousands of block producers. If a delegation mechanism is used in the given network, such as in the Cardano network, the possibility of becoming a delegator should be available to everyone. The barrier to entry must be low.
The initial coin distribution is essential. A blockchain can in principle be decentralized only if the initial distribution of coins is fair. When a team and VC funds hold the vast majority of coins then the blockchain can hardly be considered decentralized.
In the image below you can see the initial distribution of coins from Messari.
Decentralization is about the assumption that users are in control of the network, or the goal is to iterate them to this state. If insiders, teams, and VC funds hold the vast majority of coins, they control the network.
They can sell the coins one day. There is a possibility that the distribution of coins among people will increase which will improve decentralization. But it doesn't have to happen.
If the network is successful and has the potential to grow, it will be beneficial for insiders to keep the coins and profit from staking.
This is a problem for many projects with high TPS that were essentially sponsored by VC funds. Teams delivered highly scalable networks but neglected decentralization at the very beginning.
While the number of independent block producers may grow, improving decentralization, the network will still be largely controlled by insiders.
Whales can run block producer nodes, but they can also delegate coins to independent producers with the ability to change the delegation at any time. They can theoretically dictate the conditions under which they will delegate the coins to the chosen producer.
I am certainly not saying that it is always like this. Of course, a highly scalable network can be created with a fair initial coin distribution.
The last component of decentralization that we will talk about is governance. A project may not have any formal governance. The team can improve the protocol at will and does not have to consider the community. In that case, the coins play no role.
However, if a project implements governance based on coin holding, the initial unfair distribution of coins is a problem. If the majority of the coins are held by the team and the VC fund, it is obvious who will make the decisions.
It is often said that PoW networks like Bitcoin are the fairest because all coins are mined (earned as a reward). BTC coins are not used for governance. The problem is that PoW mining is exclusive. Decision-making power is concentrated in the hands of entrepreneurs who own huge mining halls. Decentralization has a downward trend both from the point of view of the number of block producers and mining.
It is hard to improve the scalability of a well-decentralized blockchain (let's neglect L2s). Improving the decentralization of a highly scalable network can be more difficult.
If there are already a large number of nodes in the network, computing resources (CPU, memory, storage, etc.) are available. If the team manages to streamline consensus, scalability increases. It is possible to implement sharding or some form of parallelization.
The IOG team plans to improve the scalability of Cardano through Input Endorsers. Three versions of blocks with different minting frequencies will be produced in the network. The blocks will be hierarchically connected in a certain way.
Input Endorsers will make more efficient use of available computing resources on the network.
It was smart from the IOG team to put decentralization first before scalability and governance on the roadmap. Since the Shelley hard fork, the decentralization of the system has been growing. So far, the scalability is sufficient, although sometimes the network reaches the edge of its maximum throughput.
However, in the short term, it can be solved to some extent by increasing the block size. Plutus V2 applications are more efficient than Plutus V1 applications.
Since the team planned from the beginning to put power in the hands of ADA holders, the initial coin distribution had to be fair.
Teams that have built a network with high TPS have a difficult task ahead of them if they plan to decentralize the network more. These networks usually have large blocks (for example up to 128 MB) or very low block time. It requires nodes with higher computational power.
The network has sufficient computing resources on several nodes, so it does not necessarily need additional nodes. Adding new nodes to the network to increase decentralization may increase the complexity of communication, which may be undesirable for a particular consensus.
It can be difficult to modify the consensus so that it is possible to increase decentralization, for example by being able to join the consensus with a cheaper (less powerful) node. I dare say that teams will not go this route.
Rather, the way is to randomly draw a smaller group of nodes (from the total number of nodes) that will participate in the production of blocks at a given time, or some form of parallelization (multiple groups responsible for a specific part of the consensus).
A delegation system is a form of decision-making power distribution and is a relevant path for high TPS networks. Although there will be fewer block producers, the number of stakers may be high. Decentralization can be solid.
However, it always depends on a fair initial coin distribution. This is a necessary basis for high decentralization. If insiders sell a large number of coins at the end of a bull market, people usually sell them too, so decentralization does not grow. If VC funds start buying coins at the end of the bear market, it will cause hype, which can lead to the growth of decentralization. However, it may still be true that whales and VC funds hold most of the coins. These cycles may tend to repeat themselves, so the pursuit of greater decentralization may be thwarted forever.
From a project image perspective, it is more attractive and credible to be decentralized as soon as possible and gradually improve scalability, rather than the reverse order.
Many texts have already been written about why decentralization is important. If we adopt networks with high TPS, but which are not decentralized (and without governance), we will create a similar to PayPal and banks. The only difference is that instead of one CEO, there will be a larger group of 'CEOs' (whales). Such a system may be a good FinTech company, but not a blockchain as Satoshi envisioned it.
Scalability is a property that users can perceive when interacting with the protocol. It is nice to use networks with cheap and fast transactions. Users love it. Decentralization is difficult to measure and users have no chance to come into direct contact with it when submitting transactions or using DeFi. They therefore tend to overlook it. This can be a problem in the long run if we want people to own the infrastructure on which new money or new banking systems are created. By choosing the network people use and the coins they hold, they decide whether our future will be decentralized or remain centralized.
The desire for profit can destroy decentralization. This is happening and we will see it in the next bull run. People like to ignore problems with decentralization if the numbers go up. Decentralized networks will exist only for fans, or they will ultimately win and defeat all VC projects. Let's believe in the victory of decentralization.
As mentioned in the introduction, scalability is a must, but not at the cost of sacrificing decentralization.