Ethereum 2.0's Phase 0 may not go live until 2021, project researcher says

The initial stage of Ethereum 2.0 — dubbed Phase 0 — may not go live until the start of 2021, according to a researcher involved in the project.

Ethereum Foundation's Justin Drake was asked during a Reddit Ask-Me-Anything session on Friday when the deployment of Phase 0 would begin. As previously reported, one of the main factors behind past launch delays was the multi-client approach being spearheaded by those involved in Phase 0's development. 

Drake cited "a public testnet with 3+ clients running smoothly for 2-3 months" and "a bug bounty program similar to bounty.ethereum.org running for 2-3 months" (among other items) as "some of the things I want to see before genesis" in his response to the question before noting:

"All the above cannot happen in Q3 2020. With Thanksgiving on November 26 and the December holidays I'd say the latest practical opportunity for genesis in 2020 is mid-November, 4 months from now. As such, I'm now inclined to say that the earliest practical date for genesis is something like January 3, 2021 (Bitcoin's 12th anniversary)."

"As mentioned in this tweet we've made Eth2 hard for ourselves (for good reasons)," he continued, highlighting the work on production validator clients and other projects in the run-up to Phase 0.

Ethereum co-founder Vitalik Buterin previously said that a Phase 0 launch would be preceded by the successful operation of multi-client test networks. Project lead Danny Ryan said during a session of the Consensus 2020 conference that having multiple clients operating on the network is a crucial security element.

"If there's a critical bug in a single client and it goes down, the network can continue moving forward because the majority of nodes may not be running on that client," he said at the time.

In a follow-up message to Drake's, Buterin wrote that: "FWIW I personally quite disagree with this and I would favor launching phase 0 significantly before that date regardless of level of readiness :D," continuing:

"Eth1 took 4 months from the first multi-client testnet to launch (~end of March 2015 Olympic to end of May 2015 for eth1 launch), and I'd argue the four-month clock started ticking for us at the beginning of July when Altona launched. Eth2 phase 0 is in some ways simpler than eth1 and in some ways more complex: more complex PoS, but no complicated GPU-oriented PoW; more optimization required, but no complicated VM, etc etc. I'm inclined to say eth2 phase 0 is a little simpler on-net. Also, eth2 is not going to have any critical applications depending on it until phase 1, so the practical risks of breakage are lower (though you could argue the ecosystem as a whole is bigger). So on the whole I see no reason to take more time for the eth2 phase 0 launch cycle than we did for the eth1 launch."

Editor's note: This piece has updated with additional comments from Vitalik Buterin