The First Version of Ethereum's Casper Upgrade Has Been Published

Yesterday, a new version of the code behind Casper, a planned change to the way the Ethereum network reaches consensus, has been published for wider scrutiny by auditors and client developers. For now, the first version of Casper has to be audited and once implemented it will alter Ethereum's software so that updating the blockchain involves a combination of proof-of-work and proof-of-stake. Since Casper FFG isn't compatible with earlier versions of Ethereum, the network will need to hard fork, however, there isn't an official date yet. David Ryan, the developer behind Casper FFG said:
"As these pieces of the puzzle are getting closer to being completed, I'll signal that it's time to start talking about fork block numbers."

New code written to change the way the ethereum network reaches consensus is now ready for review, developers said Friday.

Ethereum improvement proposal (EIP) 1011, known as Hybrid Casper FFG (short for "Friendly Finality Gadget"), would implement the first step in a long-planned shift away from the energy-intensive mining process and toward an allegedly greener method sometimes called "minting."

Ethereum's current consensus protocol - the way the network agrees to add a new block to the chain - is called proof-of-work and requires resources to be expended as its "proof."

Ethereum's creator Vitalik Buterin and other developers have discussed eventually moving to a proof-of-stake model, in which users lock ether up in special wallets and risk losing these "stakes" if they don't follow the consensus rules. That planned transition to proof-of-stake is known as Casper.

EIP 1011, if implemented, would be a first, partial step toward the full move to Casper, introducing a hybrid system that combines proof-of-work and proof-of-stake, an approach discussed in papers unveiled last year.

Casper, while long in the making, is still controversial in some quarters - for example, a security researcher at VMware called it "fundamentally vulnerable" last month.

Yet Danny Ryan, one of EIP 1011's authors, along with Chih-Cheng Liang, told fellow developers during a meeting Friday that the proposed code is "ready for review, community discussion, etc."

Ryan added that development work for ethereum clients could begin soon and that he was corresponding with the formal verification engineers.

"As these pieces of the puzzle are getting closer to being completed," he said, "I'll signal that it's time to start talking about fork block numbers."

As Ryan suggests, the change will not be compatible with existing ethereum software, meaning that the network will have to undergo a hard fork to be implemented. That said, there's still some way to go before that happens.

"In terms of testing ... I don't know when exactly that happens," Ryan continued, adding that he would "leave the EIP up for discussion a little bit longer before we start doing testing on that side."