Monero will never adopt proof of stake. There are very good reasons for this that we could get into if you're interested, otherwise I'll just focus on finality time because that's the core topic you're talking about.
Proof of stake would not improve finality or block times at all. In Monero new unspent outputs have a 10 block lock time to prevent them from disappearing in a reorg. Proof of stake would not change this at all. That's the main time constraint on being able to spend coins you recently received, it is a pain but there are good reasons for it and research is constantly happening to try to eliminate this requirement, there are always discussions on changes to the cryptography and signature scheme to eliminate it but they so far have all had drawbacks and the community hasn't come to a consensus on a trade off. Eliminating this requirement though is a very popular thing, people want it but if it reduces security it's not a good idea.
Block time is a constraint designed to prevent orphaned blocks, nodes need to validate the new chain tip before the next block is mined, this is a constraint imposed by network latency and not by mining.
I think I understand why the 10 blocks lock time exists but there are other consensus mechanisms such as avalanche that allow for fast finality and some blockchains that were 100% POW have evolved to an hybrid mechanism (#eCash, #Zano) which resulted in faster finality. There are probably technical challenges in implementing those on a private #blockchain that I don’t understand. However, it doesn’t seem infeasible as Zano did it. Now, there may be trade-off I’m not aware of. Anyway, if #Monero can keep POW and have faster finality that’s all good but I don’t know of any POW blockchains that have really fast finality (sub ten seconds). Again, I’m looking at this issue from the perspective of being the best medium of exchange and blockchains with long finality time can’t be best MoE for in-person transaction (not so much of an issue for online transactions).
BCH has "instant" finality in that there's no replace by fee, so transactions can be trusted with 0 confirmations. There are trade offs here and I'm not exactly a fan of it but it is one way to do it.
The issue of finality is a very complex one. Technically no blockchain has 100% finality ever. Practically speaking, finality is a bell curve and depending on network security practical finality is possible after a handful of blocks. In XMR that's deemed to be 10 blocks as of right now. This is true regardless of proof of work or proof of stake and is a fundamental problem of achieving consensus in a decentralized way. Proof of stake could actually reduce the security of finality, because it rewards holders and not external expenditure, and so trends towards oligarchic centralization enabling a small number of nodes to potentially collude to break finality to their benefit. If staking nodes decide to do this there's no way to fork away from them, as they will have the same number of coins on the fork and so will be able to continue the attack, as opposed to PoW where you can bit flip the hash algorithm and thus boot the attackers permanently.
I understand this is a complexe topic and I’m not the best person to discuss this in depth but I understand that the avalanche consensus has some unique properties versus BFT consensus. Also, I agree about the risk of large stakers potentially colluding over time. I think that the future of #ProofOfStake blockchains will be to integrate a #ProofOfPersonhood element like #Idena for which one human = one node. This would enable quadratic staking distributions which decrease the share held by large stakers over time (the greater your stake, the smaller your APY).
Thread collapsed
Thread collapsed
Thread collapsed
Also, for #eCash the avalanche consensus is implemented post and pre Nakamoto consensus (I don’t think the pre-consensus is deployed yet). Post-consensus allows for one block finality and pre-consensus will allow for sub three seconds finality.
Thread collapsed
