Author: Billy Tetrud 2021-05-27 10:08:57
Published on: 2021-05-27T10:08:57+00:00
The conversation revolves around the vulnerability of nothing-at-stake attack in proof of stake (PoS) systems, which allows an attacker to secretly mint on both their secret chain(s) and the public chain. The discussion also points out that the vulnerability exists in proof of work (PoW) as well. Possible solutions for this issue are proposed, such as collaborative randomness and requiring additional signatures on each block from outside validators. The importance of comparing the levels of advantage that can be obtained in PoS and PoW is emphasized to determine which one is more secure. Another topic of discussion is the burden of proof in claiming the security of a design not believed and trusted to be secure. The conversation also discusses the violation of linearity in the power of stakers in the system due to the nothing-at-stake attack and the selfish mining attack in PoW. The conversation concludes with a reminder to reject unsupported claims that PoS is a reasonable alternative to PoW.The debate is centered around the security of Proof of Stake (PoS) protocol in decentralized consensus. The "nothing at stake" problem, where anyone will mint on any chain, is a concern. Although some argue that punishments as part of the protocol can solve this issue, others believe it introduces additional complexity and does not solve the problem. A large stakeholder can gain significance in the system over time, increasing their percentage relative to all normal users who do not stake. An attacker can use a described technique to create blocks faster than honest nodes to gain an advantage, even with just 5% of the stake. This can lead to a 51% attack, causing problems for the system. Quorum-based PoS systems have punishments as part of the protocol, but they also introduce DDOS opportunities for attackers to turn off the network. The burden of proof is on those who vouch for a design that is not believed or trusted to be secure. Ultimately, PoS remains critically insecure until someone invents and demonstrates an actual way of solving these issues. The concept of Proof-of-Burn (PoB) could solve issues with Proof-of-Stake (PoS) and make systems more secure. PoB requires coins to be burned for a specific block height, making it deterministic and removing the "nothing at stake" problem that exists in PoS. Miners are unable to choose which chain to mine on and are time-locked, meaning rollbacks and instability hurt them the most. PoS may have punishment systems, but they are unproven and can lead to collusion between large and small stakeholders. PoB is a simpler system that can be more secure.The bitcoin-dev mailing list is discussing the weaknesses of Verifiable Delay Functions (VDFs) and how they can be improved. One proposed solution is to use the current Proof of Work (PoW) mechanism with lower difficulty, making it easier to find a block in approximately one minute. This would reduce the variation in block times. However, there is concern that miners could potentially overclock and freeze the circuitry to improve their energy consumption and gain an advantage over their competition. In a bitcoin-dev thread, Billy Tetrud argues that there is misinformation and bias against proof of stake (PoS), and while he doesn't suggest switching to PoS for Bitcoin, he believes that it is likely that a PoS consensus protocol could be built with substantially higher security and lower resource costs. He notes that the disagreements around hardcoded checkpoints are critical in solving attacks that could be levied on a PoS chain. He also refutes claims that PoS tends towards oligopolistic control, explaining that there is no centralization pressure in any PoS mechanism he knows of. Tetrud further argues that energy usage is not something to be ashamed of in itself, but we should aim to do better if possible. He refutes claims that PoS is only resilient to ⅓ of the network demonstrating a Byzantine Fault, pointing out that staking systems like Casper can exceed that up to nearly 50%, while PoW is not resilient up to ½ threshold since an attacker does not need to obtain 100% hashpower. During the same thread, Erik Aronesty suggests a "burned coin + VDF system" as an alternative to PoW, arguing that it may be more secure in the long run. Meanwhile, Zac Greenwood suggests using VDFs solely for making the time between blocks more constant. Proof of SquareSpace (Cardano, Polkdadot) is a UTXO-based PoS coin based on Ouroboros Praos with an inbuilt on-chain delegation system. Coin holders who do not want to run their node with their hot keys delegate to a "stake pool." Every UTXO must indicate which staking account it belongs to, ruining one of the main privacy advantages of the UTXO model, and growing the size of the blockchain significantly.
Updated on: 2023-06-14T21:21:48.918352+00:00