Not everyone who works with blockchain knows the famous blockchain trilemma, but they definitely should. The expression was phrased by Vitalik Buteri, founder of Ethereum, in a formula that demonstrates that the blockchain cannot be fully scalable, secure and decentralized at the same time, but can at most achieve two of these potentialities over the other, as a trade-off.
Blockchains are decentralized networks based on peer-to-peer communication. This imposes restrictions on vivacity, routing, and fault tolerance that are not easy to overcome without losing some other attribute in the process. You cannot simply grow a network by adding more nodes or speeding communication between them without consequences. In a public network there is a maximum limitation on the speed of communication to maintain decentralization and network security.
The level of decentralization is correlated to the degree to which actions and transactions become possible and effective without the control or authorization of an intermediary or group of them. Control, decision making and delegation capacity are distributed among network members/nodes, without the presence of a central authority.
Users, validators, and programmers need to work together within an encrypted protocol. The more decentralized the network is, that is, the more people involved in its operation, the harder it is for a single party to control or overturn it.
Safety is measured considering short term resilience and also long term immutability. This means that the protocol must be able to recover from short-term attacks without changing the previous states of the distributed ledger. Network nodes tend to be asynchronous, so the number of transactions per second (TPS) increases the time for information to fully propagate between network nodes. This time lapse exposes the network to attacks. Increasing scalability therefore reduces security when maintaining a high level of decentralization.
Scalability, in turn, is a measure of how a network to handle an increasing amount of work, maintaining throughput to accommodate the growing number of transactions, users and nodes. Block size and TPS are key dimensions for scalability as well as decentralization and governance security. The larger the network, the slower and less scalable, but usually the security trends to increase as well.
In the case of Bitcoin, this is one of the most secure networks, while maintaining a high level of decentralization. But in the way it works today, it is far behind in scalability compared to a giant like Visa in terms of transactions per second. This means that the network is too costly or too slow to scale globally. Even faster blockchains like Ethereum and Litecoin have limitations regarding scalability.
The most scalable networks are those that use less secure and more centralized blockchains, such as Ripple. This coin has more scalability with reasonable security. However, this comes at the cost of a large centralization. Similarly, with EOS, with Delegated Proof of Stake (DPoS) where the influence on the new block generation process is proportional to the amount of tokens in the network participant’s portfolio. Since the primary consensus needs to be reached within just 21 nodes to be considered by the rest of the network to be true in the state of the blockchain, something that has already enabled network members to exploit it.
Decreasing network size and decreasing decentralization gives good scalability potential. But fewer nodes may handle these systems proportionally more insecure, susceptible to malicious attacks. It is because of this security aspect that networks such as Ardor, Dash and NEM are still viewed with suspicion among many operators in the crypto world.
The most important thing about the trilemma is knowing that there will be no project that meets all needs simultaneously, and that must be dealt with. What we have today is a large universe of distributed consensus algorithms where some believe that they may one day overcome the blockchain trilemma. At the moment, however, this solution is impractical. Therefore, it is best to deal with the limitations of technology using the gamma of possibilities that already exists, considering each use case and the development skills available. The effort consumed in solving the trilemma is infinitely greater than the effort required to know how to apply the right technology for each use case, balancing its pros and cons according to the real needs of the network users.
