In the new world of Ethereum 2.0, there are no more stacks of GPU rigs and meshes of wires burning costly coal and oil. Proof of work still is the engine behind bitcoin, but Ethereum, which also promises to be the global computer of the world, is now cleaner, quieter and smarter. What makes POS a better choice, other than energy consumption, and is it really a smarter option than POW? Here is how POS works.

If sending 1 Ether is basically just sending some lines of code, how about copying and sending it multiple times to different addresses, without having the money in the wallet at all!

That’s the double spend problem that Satoshi Nakamoto solved with POW – which creates a consensus of many different computers or network of nodes. The consensus ensures that bad blocks or dubious transactions are not added to the longest, “official” chain being formed. Finders of valid blocks get “rewarded” with newly minted bitcoins, and thus there is no incentive for those who are trying to add a wrong block or transaction. They will simply end up wasting energy and thus money.

POS goes about it in a different way. It tries to “punish” or “slash” those who are trying to add bad blocks or transactions or trying to corrupt the blockchain.

To make POS work, some basic steps are undertaken:

POS works by creating “checkpoint blocks” that 2/3^rd of validators agree to as part of ongoing work of adding new blocks. Agreement is secured with voting.

So from one end of chain, several branches can spring up in various directions, but after some time, the finalized checkpoint blocks occur which create the official Ethereum blockchain. Those who voted twice for different blocks at the same height get punished with a stake burn. Same for those who vote for blocks in different directions.

Some basic things to know about POS:

• To become a validator, a user must send ETH for staking to a special address. This list allows for slashing or voting actions.
• The joining fee is high – 32 ETH. (Validator pools can however let users join with small amounts)
• Total owned Ethereum coins stand around 120 million. About 10% of these are staked.
• There are a handful of validator pools – Unknown stakers, Lido, Kraken, Binance, Staked.us, and a few more.
• Withdrawing staked Ether is not possible until now, but will be added as part of the ongoing series of upgrades.
• If all validators decide to leave, there are restrictions on how many can quit at what rate.
• The actual algorithm to carry out these operations is divided into features called Gasper, Casper, and Ghost.

Despite these features, POS leaves some ambiguity on which chain to vote for… this is the element of “subjective truth”. In case of POW, there is no such doubt, as the longest chain has done the most work and is to be trusted.

Here is how the founder of Ethereum, Vitalik Buterin, describes the element of subjective truth:

Proof of Stake: How I Learned to Love Weak Subjectivity | Ethereum Foundation Blog

Some of the POS features that are different from POW:

• No burning power to secure the blockchain – Validators or nodes of Ethereum POS don’t do any work. But they must agree all the time with other nodes – creating consensus – and not add any discrepancy or corrupt data – this will simply burn their ether stake and they end up losing money.
• Lesser issue of new coins – So Ethereum POS does not have to issue too many new Ether coins as it was doing until now on POW. On the opposite, due to burning, the net supply can actually go down.

Validators as a whole will earn ~1,690 ETH per day, before fees, on PoS.

Miners as a whole were earning ~13,100 ETH per day, before fees, on PoW.

— eric.eth (@econoar) September 15, 2022

• No extra mileage for centralized players with high computing power – Staking is linear, and more ether on stake will earn only a proportionately higher amount, but not extra money just because one has a faster rig.
• Subjective truth
• No rules
• Permission needed to join validators pool
• Stakers with money become validators and carry power

Ref: A Bitcoiner’s Guide to Proof of Stake – by Scott Sullivan (substack.com)