Would centralization undermine blockchain technology’s purpose as a shared ledger?
When we address the topic of centralization or decentralization in blockchain technology, one must be specific. The degree of decentralization could be observed on many levels. Before answering this question, I will distinguish between three levels of decentralization:
First, the political decentralization is the one observed in the decision-making for new features implementation for a given blockchain. It is directly related to the governance of the network, its protocol updates, and the resolution of critical issues that may arise.
When Segwitx2 was suggested as an improvement for Bitcoin, requiring a hard fork, it ended up not being implemented because of the lack of consensus among bitcoin developers and opinion leaders.
Having clear governance systems and means by which a blockchain implementation is updated, is critical to the maintenance of the network.
Now depending on the use we make of a given blockchain, centralization at this level may be a threat but could also be a requirement for success. Imagine a consortium of businesses using a public blockchain and the changes voted by the network operators are not inline with their needs. This could represent a risk for their business process. Therefore, a more centralized governance of the infrastructure will be required to ensure that only the required updates are made to the blockchain infrastructure. Besides, the governance will be responsible for avoiding hard forks that could eventually harm the participants.
Second, the consensus protocol decentralization looks at the validators (miners, minters…).
In more general terms, it is measured by the number and nature of the participants in the consensus mechanism. At this level of decentralization, we could compare the 21 block-producers of EOS, in charge of validating and adding new blocks through a Delegated Proof-of-Stake consensus, to the 10.000 full-node miners working to achieve consensus in Bitcoin Proof-of-Work blockchain.
Centralization at the consensus level could be a threat if there are no means of control over this centralization.
In a consortium blockchain, if most validators represent the same party, this party will have power over the network and may be able to tamper with the shared data.
Finally, the node-level or hardware decentralization would be in regard to participants in the network, users of the network owning addresses, validating transactions, and every other kind of activity allowed by the platform.
This is the physical layer. We could also refer to it as the geographic decentralization. Centralization at this level is definitely a threat from an operational point of view.
The hardware being centralized at the same location reduces the resiliency of the network in the case of a disaster, a power blackout, or a state intervention.
To conclude, I wouldn’t consider that a centralized blockchain would undermine the technology’s purpose as a shared public ledger if the centralization is well designed in terms of security or if the context of centralization is justified. For instance, a delegated proof-of-stake consensus could be considered more centralizing than a proof-of-work consensus. Nonetheless, this centralization has proven to be beneficial for participants of the network in terms of speed and scalability while maintaining the necessary means of control over the delegates.
This been said, the case of Bitcoin blockchain presents a real challenge and potentially a threat because the consensus mechanism relies more and more on the computational power and storage capabilities in a trustless context. The security of the Bitcoin network, a trustless network, is built on its ability to protect itself from 51% attack by probabilistic calculations and the randomness of miners selection. But if small miners can no longer compete, then the security is undermined.
How could blockchains scale while remaining decentralized?
To solve the issue of scalability while maintaining a high level of security and decentralization, many solutions are under experiment, some are already implemented.
For example, Bitcoin has adopted the lightening network, a peer-to-peer “off-chain” network for micropayments, that operates in combination with segregated witness. In another upcoming update that will also improve the security of Bitcoin blockchain, “watchtowers” will be introduced to keep an eye on malicious transactions and detect them before they cause any harm.
Other scalability solutions like “sharding” are looked at for Ethereum blockchain. This solution is still under development and is expected to be deployed over two steps, one to separate the data layer and another to separate the state layer.
Sharding, like lightening, are both layer-2 scalability solutions. Whereas BloXRoute is a layer-0 solution that could eventually be implemented on any blockchain.
