6 comments

[ 3.0 ms ] story [ 25.2 ms ] thread
As replicas are easy to create, it doesn't matter how many of them exist in a system. Bitcoin could have 99% of the nodes being malicious and it wouldn't matter as long as the participants are able to access the longest chain
"Permissioned" ledgers have existed and failed many times before the invention of bitcoin. This is not decentralized. None of the BFT protocols are.
You can combine BFT with dPOS for full decentralization.
dPOS (Delegated Proof of Stake) is just another word for "more centralized proof of stake".

I am not saying dPOS is bad and even think it might be a good tradeoff for building real world applications with moderate decentralization, but claiming you can achieve "full decentralization" with dPOS is kinda ridiculous.

dPOS for full decentralization.

Any form of Proof of Stake isn't really decentralized, often as a result of a single or small group of users who have minted the entire supply. Other users are then expected to purchase the supply rather than generate it themselves. The sale price on exchanges is often spoofed due to the nature of the low liquidity and ease of wash trading.

Is there any code available to replicate these findings? I couldn't find a link in the PDF, only other repo refs e.g. the relic lib they used.