The Ethereum research team working on the so-called Ethereum 2.0 (also called Serenity), a thoroughgoing update of the platform that will introduce the proof-of-stake consensus mechanism and the sharding approach to on-chain scalability, hosted an AMA session on the Ethereum subreddit that turned out to be hugely popular.
PoS involves validators getting the right to earn blocks in proportion to the ETH they staked instead of the computational resources they contributed to proof-of-work (PoW) mining. Sharding means splitting the network horizontally into several interconnected sub-networks (shards) to allow the platform to process more transactions at the same time.
When asked about the estimated date when a fully scalable version of Ethereum may arrive, core developer Justin Drake suggested the year 2021, although he noted that in 2020, some sharding functionality may well already become useful.
One of the most popular questions turned out to be about the competition between Ethereum and Polkadot (the latter being compatible with Ethereum smart contracts). Like the Ethereum’s proposed sharding-based architecture, Polkadot (as well as Cosmos, Dfinity and potentially others) is aiming at introducing a system of interconnected blockchains relying for security on a core chain to a certain extent. The Ethereum developers involved cast doubt on the idea that Polkadot will launch with full functionality if it launches by the end of this year. Justin Drake further noted that the key element of Ethereum’s sharded architecture – the beacon chain – may also have been launched by that time.
Another important query centered on how to allay the potential concerns of developers who may be hesitant to develop on Ethereum at this point, given that the network is supposed to undergo fundamental changes in the future. Ethereum developers’ responses were somewhat less reassuring because they essentially implied that the decentralized applications tailored to the current version of Ethereum are just prototypes and learning experiences, and would need to be reworked to be redeployed on Ethereum 2.0. However, in another thread, Buterin seemed to suggest that the redeployment may turn out to be not prohibitively difficult.
One of the most complex issues related to sharding appears to be the communications between shards. It may render the execution of transactions involving smart contracts from two or more different shards much slower than that involving only smart contracts from the same shard. Buterin acknowledged the issue and answered that there are potential mechanisms above the base layer that could help rectify it (discussed in more detail by Justin Drake here; warning: very technical).
To showcase the scale and scope of the effort undertaken by the Ethereum 2.0 team, one may note the fact that it is collaborating with multiple partners, including Synopsis and AWS, to develop specialized VDF hardware for the Serenity architecture.
Undoubtedly, if you are an Ethereum or blockchain developer or a technically-sophisticated blockchain community member, you will find a lot more to chew on there.