Your Guide To Ethereum 2.0

Key Takeaways:

? 2020 will start the first phase of Ethereum 2.0 – dubbed Phase 0 – possibly as early as Q2 (95% confidence in July)

? Ethereum 2.0 is a Proof-of-Stake (PoS) version of the current Ethereum version and introduces sharding to increase the network’s capabilities

? Introducing Proof-of-Stake (PoS) will lead to over 10 million more ETH being locked into the protocol once version 1.0 and 2.0 are integrated

? There are a few mechanisms in place to discourage stake concentration, although the risk is still present.

Ethereum 2.0 is coming. For real this time. 

Preston van Loon, the CEO at Eth2 client implementation team Prysmatic Lab, has said that the initiative will soon see the debut of a major multi-client test network. Looking to the horizon, he said he expects a Phase 0 launch late in the second quarter of 2020.

While there have been several delays to get to this point, the project is reaching the end of its single-client testing period. Developers will soon implement a major multi-client test network, which will run for at least two months before the mainnet launch can take place.

“The latest state of phase 0 is that the spec is complete, all of the features that we’re expecting to be there are frozen, and our cryptography library has been standardized,” he said. “Testnets are now running on the most recent update.”

While much can change over the next few months, van Loon estimated that the mainnet launch will take place around late Q2.

There are currently a total of eight ETH2 clients, many of which have begun testing their features.

Prysmatic Labs’ Prysm, for instance, launched its testnet “Sapphire ” in June 2019. According to the latest update put out by Prysmatic Labs on Jan. 31, the testnet now features over 35,000 active validators. 

A few others testing Eth2 are Sigma Prime’s Lighthouse and Status’ Nimbus. Another interesting breakthrough came in December of 2019, when Parity’s Shasper connected with Prysm on the Sapphire testnet, making Sapphire the first public multi-client Eth2 test network. Van Loom has stated that they’re not pushing for this multi-client testnet until they’re better prepared, but they should be ready by the end of this month.

Multi-client testnets are important to see as they’re one of the requirements for the phase 0 mainnet launch. Two or more compatible clients need to run a multi-client testnet together for at least two months with everyone involved feeling confident about the network. 

Other checks are in place as well — client implementers need to signal their approval, auditors need to check the code for security purposes, and the Ethereum Foundation needs to give their consent. 

It’s a lot, and it’s deliberately a multi-tiered redundancy check in order to convince stakers that they won’t lose their money with the new system.

Vitalik Buterin, the co-founder of Ethereum, added that “development work with Eth2 phase 0 is going smoothly.”

The last big challenge left for developers to resolve before phase 0 can be rolled out, Buterin said, concerns peer-to-peer networking.

“The problem with peer to peer networking is that a lot of messages from all of the validators that are flying around add a number of bytes to the network,” he said. “It’s less than a couple megabytes a second, but we’re still trying to reduce that.”

“At this point, it’s just a matter of optimization and ensuring that the system is secure,” Buterin added.

Ethereum 2.0 simplified: proof-of-stake & sharding

Eth2 is a Proof-of-Stake (PoS) version of the current Ethereum mainnet and introduces the concept of sharding to increase the network’s throughput.

“I always wondered why Ethereum is not the biggest blockchain out there,” von Loon said. “The reasons, I think, are that it does have a lot of friction in terms of user experience and it doesn’t scale very well… I would guess that [ETH] is trading at a discount, not just due to Eth2 uncertainty, but also for these other reasons.”

The Eth2 updates, according to him, precisely seek to solve these problems.

There are a few ways to argue that PoS is theoretically superior to Proof-of-Work (PoW) in many ways. It is less energy-intensive, easier to reason about from a security point of view, cheaper to participate in, and necessary for sharding due to finality properties.

And the new shard structure aims to improve the ethereum blockchain’s scalability.

With the Eth1 system, van Loon explained in a blog post, every single node running on Ethereum has to process every single transaction that moves through the network. Though decentralized and secure, this method puts a limit on the network’s scalability and increases the risk that the whole system jams up.

Sharding provides a solution for the scalability problem without compromising the network’s security and decentralized nature, he said.

In Eth2, there will be a central blockchain, the Beacon Chain, that coordinates all 64 side chains called “shards.” Every shard will act as a full PoS system and contain an independent piece of state and transaction history. Instead of processing all network transactions, each node only processes transactions for a certain shard.

“Eth1 doesn’t scale very well. It doesn’t support any large use case to the point of global adoption,” van Loon said. “With Eth2, you will have a throughput increase exactly by the number of shards you have. We are launching with 64 shards so you’ll have 64 times more capacity than before.”

Currently, there are a few alternative solutions to Eth2 including Polkadot, NEAR, and Avalanche. NEAR, especially, took on a similar design approach as Eth2 – they both feature a Beacon Chain, shards, and shared blocks feeding input into the Beacon Chain.

According to Buterin, however, despite these similarities, the two protocols have very different focuses.

“NEAR is willing to have a protocol with more application-specific features, whereas the Ethereum approach is more similar to Bitcoin in the sense that Ethereum people value having a simple base layer and building other things on top of it using either programming languages or second layer,” he said.

Furthermore, Buterin believes that many of these alternative solutions have yet to understand that the longevity of a project depends more on the community than the underlying technology.

“They’re bringing a bunch of money and doing the entire project closed-source, assuming that once it launches, people will just flock to it,” he said. “They are taking a more corporate approach to launching a chain and I don’t think that’ll end up being effective.”

What comes after the Phase 0 mainnet launch?

Eth2 follows a phased rollout, as spelled out by an EthHub post:

? Phase 0 will feature two active Ethereum chains: the current PoW main chain and the Eth2 Beacon chain.

? Phase 1 will focus on the construction and deployment of the 64 shard chains, although the side chains will only act as data processing layers without shard state execution or account balances.

? Phase 2 will introduce structured chain states, smart contracts, a virtual machine based on EWASM, and other features that will start pulling the entire system together and enabling transaction processing on the side chains. Any plan beyond Phase 2 is still open to significant revision.

Beyond phase 0, Buterin has said, Eth2 teams are already researching and developing phase 1 and phase 2 specifications, with phase 1 work now almost complete.

“The core design of these phases has been nailed down for quite some time now,” he said. “There are just some details around optimizations.”

For example, developers are still optimizing the Proof-of-Custody game – a fairly new mechanism to ensure that blocks are cleaned up and verified before they get signed off, Buterin said.

Moreover, the data availability checking protocol – by which a client randomly checks for pieces of data in the shard chain – is still in the process of being optimized. Here, developers need to make sure that, on average, there is enough data available to construct an entire block.

“This is a construction that hasn’t been done before,” Buterin said. “We want to have one of the clients implement this at some point.”

Another crucial research question concerns the integration between the Eth1 and Eth2 chain.

With the current approach, the two chains will not be merged until phase 2, but Buterin pointed out that the community “wants Eth1 and Eth2 to work together instead of as separate ecosystems for a long time.”

On Dec. 19, Buterin proposed an accelerated schedule for early Eth1 and Eth2 merge. With the new process, the Eth1 chain will live on as a shard chain in the Eth2 system. There will be a new list of “Eth1-friendly validators” maintaining a node on both networks to verify Eth1 transactions and enter them into the new chain.

“The reason why I think a lot of people in the Ethereum community would want this is because they first of all care about scalability,” Buterin said. “This proposal would make it more feasible to verify individual Eth1 blocks without having to store the entire eth1 state, which is fairly big.”

The migration to come

In terms of user migration, the current design features a one-way bridge from eth1 to eth2.

But the incentive to cross over, according to van Loon, will depend on the type of people who are doing the migrating.

Stakers, for example, are incentivized to migrate early because rewards are higher when the active validate count is lower. Decentralized applications (dapps) built on Ethereum, on the other hand, will be drawn to Eth2’s higher capacity during phase 2 – when state execution and the smart contract platform have been implemented.

“But if dapps are not having a capacity problem, then they won’t be motivated to move because they want to be where the users are,” he said. “In the worst case, I imagine a great migration at some point where it stops being an opt-in migration. We’d just bring everyone over and separate the connection between the two.”

Although designing the migration process is a complicated technical matter, both van Loon and Butering envision user migration to a relatively smooth experience.

“On the back end, this is a very complicated process,” van Loon said. “But to the user, it is just a couple button clicks to use a new client or provider. It has to be easy to do or else people are not going to do it.”

Buterin agreed but pointed out that there might be a short period where the system sees no transactions.

“Challenges are concentrated in the client and application developers. If the migration is technically solved then it’s pretty intuitive,” he said. “But there is a possibility that the migration will lead to a period of maybe one to two hours where there are no transactions happening.”

More than 10 million ether locked in Eth2

A Bitcoinist report previously contended that, based on Buterin’s rough estimates, shifting to the PoS model might lock up approximately 10 million ether.

Buterin predicted that the amount of ether being staked will far surpass 10 million after the Eth1 and Eth2 systems merge.

“My guess is that in phase 0, we’re going to see a lower amount of ether being staked, likely to be the low millions,” he said. “But at phase one, especially past the merger, we’d see quite a lot more than 10 million.”

Meanwhile, the Ethereum Foundation is conducting economic modeling to get mathematically concrete numbers on validator costs and rewards, Buterin said, but much of the information will only become available after the phase 0 mainnet launch.

“We recognize that there’s a limit to how much information we can get just by analyzing,” he said. “And one of the reasons we want to do a phase 0 before phase 1 is to get an idea of exactly how many people will stake before there are a lot of activities depending on the chain.”

As of now, Eth2 does not have a fixed issuance schedule, which Buterin thinks works better given that the Ethereum network is meant not to store value but to build applications upon.

“I think people value the level of security more than the guarantee of a lot of issuance,” Buterin said. “If we say everything is just transaction fees, then if there is almost no transaction fee for some period of time, then the network would have no security for that period of time. Whereas in our model, you have this interest rate and it goes even higher when people are staking.”

Buterin added that the could be an alternative future for Ethereum where a large number of ether is being staked in Eth2 rather than being stored in other application protocols like MakerDAO, Uniswap etc.

“I could see a situation where ether being locked up in eth2 staking becomes competitive with ether being locked up with other protocols,” he said. “And that would look very different from the eth market as it exists today, where most if it’s just doing nothing and sitting in people’s cold wallets.”

The risk of stake concentration

Now that ethereum is set to transition to a PoS system, concerns linger regarding the possibility of stake concentration, which could pose a serious security risk to the protocol.

Buterin explained that there are several mechanisms in place to disincentivize stake concentration.

First, in the sharding system, the number of transactions one has to process is proportional to how much ether one is staking. As a result, even if two stakers merge, they still have to verify the same number of data; merging won’t save them any time or effort in this regard.

Second, Eth2.0 features an anti-correlation penalty system, in which the more validators that go offline at the same time, the heavier the penalty. It is, therefore, riskier to put a significant amount of stakes in one service or exchange.

Finally, developers are trying to improve the network design to make it easier and cheaper for users to run Eth2.0 nodes and stake by themselves.

However, according to Buterin, the risk is still present.

“Stake concentration has been one of our concerns right from the start,” he said. “We have put in different mechanisms in the protocol to deter it, but it’s definitely possible that even despite that, many people will just be lazy and we’ll see staking happening largely with centralized providers.”