Vitalik Buterin, Chief Scientist at the Ethereum Foundations, has given a long term and short term roadmap at the opening of Devcon in Prague this morning.
After going through a brief history of Casper and Sharding starting in 2014, Buterin said the fourth phase of ethereum – after Frontier, Homestead and Metropolis – was to begin soon. Buterin said:
“Serenity is a realization of all of these different strands of research that we have been spending all of our time on for the last four years.
This includes Casper and not just Hybrid Casper: 100% organic, genuine, pure Casper.” It also includes Sharding, eWASM and a long list of other improvements.
“Serenity is a new blockchain in the sense of being a data structure,” Buterin said, “but has a link to the Proof of Work (PoW) chain. The Proof of Stake (PoS) chain would be aware of the block hashes of the PoW chain, you’d be able to move eth from the PoW chain to the PoS chain.
So it’s a new system, but it is a connected system and the long, long term goal is that once this new system is stable enough, then basically all of the applications on the existing blockchain can be sort of folded into a contract on one shard of the new system that would be an EVM interpreter written in eWASM.”
“Beacon Chain Proof of Stake is kind of… the blockchain is not holding information. It’s kind of like a dummy chain. All you have is validators, and these validators are executing and running the PoS algorithm.
This is kind of like a half-way between a testnet and a mainnet. It’s not quite a testnet because you’d be able to actually stake real eth and earn real rewards on it, but it’s also not quite a mainnet because it does not have applications and so if it breaks people are hopefully not going to cry too badly.”
Phase 1 is where the sharding part turns on. It’s a simplified version that doesn’t do sharding of state, it does sharding of data.
“If you want to do a decentralized Twitter on the blockchain, you’ll now have the scalability to do this, but you won’t have all the state execution capability to build smart contract applications and all the fancy, complex stuff.”
That would be available during the second phase when eth can be moved across shards, enabling state transition, the virtual machine, and “all these cool stuff.”
One confirmation in the Beacon Chain involves confirmation from thousands of validators so you’d be able to treat one confirmation as close to final.
“Hopefully 1,000x higher scalability,” Buterin said. That would be circa 14,000 transactions a second from the current 14 or so.
The spec has been moving fairly quickly, Buterin said, but it will stabilize fairly soon. There’s something like eight implementations of Ethereum 2.0 happening right now, according to Buterin.
There’s suggestions the testnet might launch during the first quarter of next year. It would be nice to see a testnet working between two implementations, or even just one implementation, Buterin said.
Buterin said it took 19 months to go from the whitepaper to the launch of ethereum 1.0, and part of the reason why it took so long was because they would work on cross compatibility way before the spec even finished. So there would be changes to the spec, then cross compatibly tests, more changes to the spec, more cross compatibility tests and so on about five times.
Now they don’t focus on cross-client compatibility until there is a release candidate of the spec. “I don’t think we’re actually that far from a release candidate of the spec at least for limited portions that don’t include state execution.”
In other words, Proof of Stake should hopefully be out sometime next year. Then data only, or simple sharding, launches perhaps sometime in 2020. Buterin clarified what this means for dapps, but not what it means for transactions.
Once eWASM launches, then the most important part of sharding is completed. At that point we can be fairly sure ethereum will be able to handle a lot more transactions, but it may be that the 1,000x scalability starts with simple sharding.
The Beacon Chain is full Proof of Stake. Once that launches, the role of PoW miners will decrease considerably, as will presumably their reward. Eventually, then, the PoW chain is effectively fully discarded, but according to Buterin that will be in the very, very long term. So miners might hang around for quite some time.
That’s, at least, as far as current estimates are concerned. Things may change depending on how fast they move or indeed how slow.
Now, however, there’s a very clear road to Serenity. Just as there is a far better understanding as to why they made the decision they did in discarding Hybrid Casper. It looks like they effectively had no choice if they were to meet the 2020 target for massive scalability.
Hybrid Casper would have slowed down considerably and would have complicated considerably any work towards sharding. Here, instead, they are effectively starting anew.
The Beacon Chain is a brand new blockchain. Fully Proof of Stake. Presumably optimized for sharding. Having no meaningful effect on the PoW chain, but still linked to it during a transition period. Meaning, presumably, they can move a lot faster on the Beacon Chain.
Soon now this is all to begin, with a lot of activity to be expected as 2019 nears with a clear vision for a massively scalable PoS blockchain.