Vitalik: The benefits of complete decentralization and "limited" centralization, blockchain has both

Ethereum founder Vitalik Buterin recently shared his views on decentralization and centralization on Twitter. Complete decentralization is of great value, but "limited" centralization is also beneficial. Vitalik believes that blockchain has both.

Here is what he shared on Twitter:

“Yesterday I talked about the benefits of full decentralization. Today I will discuss the benefits of intentionally “limited” centralization.

The best argument for centralization is usually about centralized incentives: if 1 million people each have a tiny stake in the outcome of something, and little ability to influence it, then they will not care about their tiny influence (or worse, be easily led by the incentives of others attached to them)... but if one person has a large stake and influence in an outcome, they will need time to learn, optimize, and exploit economies of scale to make + execute good decisions.

The best arguments against centralization are the fragility of centralization, the risk of abuse of power, the lower “bus factor”, etc.

Can we combine the benefits of both (decentralization and centralization)? Well, blockchains already do this. Think about the block production mechanism: basically, every 15 seconds we randomly select a "dictator" who has full control over the transactions (tx) included in a block and, from an economic perspective, is the "perfect marginal claimant" of the block reward.

This architecture actually has a lot of benefits and is by far the most resilient to bribery/"under-the-table payments" of all the architectures we know of so far.

Miners and transaction senders do sometimes perform private transactions with each other, but the benefit is so low that it hardly ever happens.

And it has certain drawbacks: if you don’t like a particular block producer, you have to wait 15 seconds for the next one.

Once we have more complex applications on the chain, we will face challenges. Optimal tx inclusion, including capturing “miner-extractable value” through frontrunning (https://arxiv.org/abs/1904.05234) and so on, will become complicated.

This introduces risky centralization of complex block producers.

This problem can be solved by carefully designed "limited centralization".

If we divide consensus into two tasks:

1. Select which txs go into blocks (complex, therefore prone to centralization)

2. Use these txs to verify other people’s blocks (routine work is therefore more decentralized)

The *kinda* in the current eth1->eth2 roadmap features the following: eth2 validators must explicitly register to become “eth1-friendly” block proposers.

Proposing a block consumes more resources, validating a block consumes less resources, and a block validator does not necessarily have to be a block proposer.

If there is a concentration of block proposers, the risk to users is much smaller than if there is a concentration of validators: even if 90% of proposers are bad, the worst they can do is delay transactions by 10 block slots (~2 minutes) on average. If they propose bad blocks, validators will ignore them.

Another more explicit example is the Miner Extractable Value (MEV) auction.

Basically, in a rollup, once a day the right to aggregate and choose the order of transactions is auctioned off, e.g. to the highest bidder.

This completely separates the economies of scale heavy centralized work (tx picking) from the decentralized work (tx validation), and also generates revenue!

This has limited risk, as anyone can post a TX directly to the chain and force it to be included in the next batch of blocks.

Thus, a mechanism where one region is concentrated and contains economies of scale (and thus becomes centralized) can exhaust economies of scale in surrounding regions (and thus lead to more decentralization).”

Next is Vitalik’s previous sharing on the benefits of “full decentralization”:

People who say “full decentralization is overblown and you just need to be non-custodial” are missing the point. A fully decentralized (or “serverless”) application is valuable because it gives users confidence that “it will always be there” so you can build on top of it.

This is why projects like Uniswap, Tornado Cash, etc. easily inspire people to build ecosystems around them, while more centralized alternatives fail.

If you build your dapp on fully decentralized/serverless infrastructure, your dapp can continue to use it even if the rest of the community around that infrastructure disappears. It may become weaker due to smaller network effects, but it will still work.

Link to this article: https://www.8btc.com/article/580621
Please indicate the source of the article when reprinting