Ethereum is going POS

One of the most anticipated events in the Ethereum ecosystem is the transition from Proof of Work to Proof of Stake. Both of these systems are used to reach consensus and maintain blockchain security.

Proof of Work uses mining, requiring a large amount of electricity and computational power. Proof of Stake intends to replace a system that requires so much spending of resources in order to obtain a secure consensus through the network's participants, who will have to make a stake with their own coins Virtual (in this case Ether) in return for a proportional prize (the amount wagered against all the coins available throughout the network) for their virtual mining.

No PoW-style mining occurs in the PoS; The deposit is made with the aim of encouraging honest actions. The validator's deposit is burned (automatically lumped and destroyed) if it performs a network damage move, as defined by the slashing minimum conditions.

This step will be implemented in sequential steps, passing from an intermediate phase during which there will be a Hybrid PoW and PoS system before switching completely to the PoS. Enter the code for the contract Casper is part of the first step to get to the PoS. According to a recent presentation by Buterin at the Ethereum Meetup held in Taipei, the first phase has almost been completed.

By making a highly transparent move, Buterin has published the code that will implement the hybrid system for his Casper contract. While Casper is the protocol set up to govern the Ethereum PoS system, it will be implemented as a Ethereum contract in order to lessen the possibility of having problems reaching consensus among clients.

If Casper is implemented on the client side, it will be more difficult to ensure that everyone uses the same slashing conditions, which is important because they are the economic penalties that keep the system together. Without assured customer consent about these penalties, there are more chances of networking attacks. In addition, in a contract, Casper could easily be updated by swapping the contract address with an updated version for future hard fork.

While Buterin was ready to share the code of the Casper contract, the latter was still ready for only three quarters. The fourth incomplete part of the code was thus described by Vitalik on Reddit:

It remains to be implemented:

- A sequence message sequence mechanism (necessary to fully test the slashing conditions);

- Switch ECDSA signature checks to arbitrary validation control

Apart from that, it's almost complete. Also, it should be noted that the Casper contract is only one of three parts. The other two treat the logic of the demon validator and the rule in the case of hard fork. These are both being written, but they are not in such advanced phase as the contract.

So once the contract code is ready, only the daemon and the fork will remain to be done. The demon validator will be a program that will run off-chain (not on the blockchain but on the user's device) that interacts with the Casper contract so that a node will know when to send messages to validate the block.

The fork rule of choice is a piece of code that will be contained in each node in order to determine to which exact conditions the canonical chain (the longest) will have to undergo. This serves to ensure that a winning knot on the winning chain. If a node bets on a false chain, that node (the validator) loses its deposit. This is really a very important aspect for the security of the blockchain, because it is the time when consensus is determined with regard to the choice of the canonical blockchain.

The question everyone put up is: how many ETHs should I bet to offset gas costs (ie transaction costs) to send messages and make a profit? Buterin spoke of the minimum amount for validators' deposits, saying:

This is a thorny issue. Here is a minimum implicit as you have to pay for the gas to prepare and send the message, so it is not profitable unless the deposits are at least 1000-4000 ETH. One possibility that we are thinking of in the later stages of hard forking is a specific discount for the performance associated with this contract, but it is not yet close to being finalized. 

It's a lot of money. Quite similar to the PoW mining, unless you are rich you need to join a PoS virtual mining pool to see some gain. Proof-of-Stake should possibly be enabled to support mining pools.

Another useful feature of Casper is the ability to send a deposit from a contract instead of just from your wallet. A person may withdraw their rewards as a validator from a contract that releases funds only if some other condition is met. This would allow a "duplicate use of your Casper deposit as a security deposit in some other application." Thus, if a validator needed to lock off a fair amount of money, it could use it as a guarantee while it is still virtually in Extraction and is getting the rewards of a block.

When the proof of the stake is to be implemented, Ether will be subject to some volatility. You are ready? Do you know how to deal with it?