Greetings to the initial eth2 brief update of 2020! This promises to be an exhilarating year.
tldr;
- Launch of v0.10.0 specification as a stable benchmark for multi-client testnets and security assessments
- @paulhauner and @sigp_io crew diligently working on developing Lighthouse
- Reintroduction of Prysm testnet, now featuring aggregators and mainnet configuration
- A new proposition for a quicker merger of eth1+eth2 (also known as Phase 1.5)
Launch of v0.10.0 for security evaluations and multi-client testnets
v0.10.0 — 404 Not Found was unveiled last week. Browse the release notes for the technical specifics (integration of IETF BLS, simplified eth1 caching, etc), but what does it truly signify for eth2?
v0.10.0 represents the culmination of modifications made to the Phase 0 specification after the restart post deVcon, along with the adoption of the new IETF BLS standards. For a period, these updates caused disruptions in the steady development cycle of the eth2 client teams and delayed the arrival of testnets that were anticipated. With the launch of v0.10.0, we are re-entering a consistent rhythm.
New and refreshed resources
With the introduction of v0.10.0, we once more have a remarkably stable target for both client testnets and third-party evaluations and security reviews.
To facilitate this, we have recently revised several onboarding resources (and created a few new ones). If you strive to gain a better understanding of the eth2 Phase 0 specifications, these are definitely worth checking out!
Audits in progress
Least Authority commenced their security evaluation this Monday and is currently delving deeply. We have high hopes and are excited to collaborate with a team of such caliber!
Another intriguing piece of news involves an audit and formalization of the Phase 0 cryptoeconomics led by a new EF group — Robust Incentives Group (RIG).
RIG has developed a cadCAD environment to model the eth2 economic frameworks and investigate numerous forms of attacks: including timing assaults, the potential consequences of various sized cartels, and more!
Explore the essence of their work presented as an easily digestible python notebook.
What’s on the horizon?
While I will only offer a sneak peek here, I promise to delve into further details in forthcoming updates.
In tandem with ongoing audits, clients will develop towards v0.10.0, assimilate new BLS standards, and establish steady testnets. Once initial stability in v0.10.0 is accomplished, multi-client activities will commence.
Concerning this activity, I anticipate a limited number of multi-client assessments against primarily single-client testnets. Specifically, I foresee some clients joining previously single-client testnets — as minority nodes — to evaluate initial interoperability.
Once we achieve success in this area, we will orchestrate a shared genesis with two or more clients from the outset. I predict some degree of disruption and a significant amount of learning during this phase. Ambiguities in the specification might surface with minor clarifying changes incorporated. We may even discover that something was faulty that we initially believed was functioning correctly — large testnets serve as their own form of audit.
As audit results start to come in throughout February, we expect to implement a wave of modifications to the Phase 0 specification. The scale of these modifications and their potential impacts on client development and timelines are still to be assessed. Nonetheless, a post-audit release of v0.11.0 is slated to be available near the beginning of March.
If the adaptations are minor, clients will implement the changes, patch testnets, and proceed. If more significant, the integrations may require additional time, extra testing, and complete restarts for existing networks. Following adequate time at a stable v0.11.x, a definitive v1.0.0 will be prepared for mainnet deployment: the precise duration will hinge on the extent of the changes.
This offers a broad overview of the upcoming months. I will continue to update you and provide more specifics as we progress.
Lighthouse on an optimization streak
Although Sigma Prime has remained relatively quiet since they disabled their initial public testnet, the team has been diligently focused! Lighthouse lead, Paul Hauner, recently shared a thrilling tweetstorm packed with enticing details regarding recent optimizations and advancements ontheir eth2 client.
It seems they have been conducting a little-known public testnet over the last 4 weeks, achieving remarkable improvements across the board, from BLS to block processing to database reads/writes. Download it and give it a shot!
Additionally, if you have some frontend skills and wish to participate, Sigma Prime has recently issued an RfP for a Lighthouse User Interface. Such initiatives are crucial for ensuring that validating is easily available to all user types.
Prysmatic’s testnet reboots with mainnet configuration and aggregators
Prysmatic Labs has recently reinitiated their public testnet featuring the mainnet configuration along with the attestation aggregation strategy! As mentioned in previous articles, the mainnet setup includes larger caches, extended epochs, and is generally more robust than the initially employed minimal setup. Running a stable testnet with this configuration and tens of thousands of validators is a significant benchmark that Prysmatic is currently pursuing.
If you’re eager to get involved, become a validator or join the discussion on Discord.
And by the way, they’re hiring! Collaborating remotely with Prysm offers an excellent opportunity for seasoned developers. Help shape the future!
Phase 1.5
During the holiday season, Vitalik shared a fresh proposal on how we might accelerate the eth1+eth2 merger and start reaping the advantages of the new eth2 framework sooner. This proposal advocates for merging eth1 into a shard of eth2 post the Phase 1 infrastructure (shard data chains) implementation while preceding a complete Phase 2 (featuring extensible EEs and all the cross-shard functionalities) — thus Phase 1.5!
A phase 1.5 offers numerous potential advantages to the Ethereum protocol, its developers, and users.
- Eth1 residing within eth2 permits native access to the scalable sharded data layer. Some of the most thrilling projects being developed on Ethereum today are layer 2 protocols that scale alongside the data volume accessible to layer 1. These will pair excellently with eth2, even with just a single chain featuring native computation (i.e., eth1 as a shard). Collectively, these projects are termed “rollups”. They come in various forms and I anticipate this overall design space will continue to grow and thrive.
- Eth1 transitioning to a shard of eth2 eliminates PoW from the protocol, significantly lowering issuance and putting an end to energy-intensive mining on Ethereum for good.
- Lastly, integrating eth1 into eth2 on an expedited timeline reduces the complexity — unifying the system, the community, and the development of the core protocol. Although the eth2 framework is being developed in parallel to the existing Ethereum chain, an earlier integration of eth1 into eth2 (beyond the technical advantages) ensures that the community of protocol developers, application developers, casual contributors, and end-users remains cohesive within a single, unified Ethereum.
Based on initial discussions and reactions, developers and community members are enthusiastic about this proposition. Phase 1.5 as it stands presently primarily depends on the success of two independent facets — Phase 1 of eth2 and Stateless Ethereum on eth1. The relative timelines of these elements will guide how and when this proposal might materialize. In the upcoming months, we will diligently examine better specifications and comprehend the technical challenges at hand so that we are ready to advance on this if and when the moment is right 🚀.
