--- tags: newineth2 description: The latest update on Ethereum 2.0 development image: https://benjaminion.xyz/f/favicon-96x96.png --- <style> a {text-decoration: underline;} a {color: #0000ee;} a:visited {color: #551a8b;} </style> # What's New in Eth2 - 13 February 2021 ![My avatar](https://benjaminion.xyz/f/ms-icon-144x144.png =32x32) Ben Edgington (Eth2 at [ConsenSys](https://consensys.net/) — all views expressed are my own) Edition 62 at [eth2.news](https://eth2.news/) ## Top picks No top pick this week. Read my editorial instead :slightly_smiling_face: ## Editorial By and large, I don't think people have realised quite how much the Ethereum roadmap has evolved over the last four months. In particular, very few have noticed that the current plan for Ethereum 2.0 _does not bring any useful scalability on its own_. In earlier plans, the full weight of bringing much needed scalability to Ethereum was borne by the transition to Ethereum&nbsp;2.0. Eth2 was to deliver scalability by sharding data/blocks (Phase&nbsp;1), by sharding execution (Phase&nbsp;2), and by sharding state (also Phase&nbsp;2), or maybe going completely stateless. With these three all parallelised at the base-layer, the Ethereum protocol would become genuinely scalable. The current Eth2 plan is to shard only data (blocks). On it's own, this does not improve things at all. Instead, we are relying on Layer&nbsp;2 rollups---outside the core protocol---to deliver the equivalent of state and execution sharding. This change came about almost overnight after Vitalik published his [rollup-centric ethereum](https://ethereum-magicians.org/t/a-rollup-centric-ethereum-roadmap/4698) concept. Rollups are like a turbo-charger for Ethereum's engine, the EVM. The EVM is underpowered because it is starved of the data that is its fuel. In part this is deliberate to avoid massive state bloat. Rollups act like a turbo, compressing the fuel--air mix, the data, and forcing it under pressure into the EVM. They also take care of the state, outside the protocol. But rollups are completely independent of Eth2. They are in development and running on Eth1 today, and already beginning to deliver on their scalability promise. So, then, what role for Eth2 in scalability? Our current Eth2 plans _assume the success of rollups_. Without rollups, there is no extra scalability. Once a functioning rollup ecosystem is in place, however, Eth2 will deliver orders of magnitude in performance boost. The rollup turbo-charger is effective on Eth1, but is still limited by the capacity of the fuel line. Data-sharding in Eth2 lays down 64 new fuel lines, each with hugely more capacity than the current Eth1 blockchain. We will be turbocharging the turbocharger. Some quick advantages of the new approach: - Rollups allow us to decouple the move to proof of stake (the merge) from sharding, and we can deliver the merge quickly in the form of the [executable Beacon Chain](https://ethresear.ch/t/executable-beacon-chain/8271?u=benjaminion). This design allows the move to PoS to be minimally disruptive to current dapps, tooling, and users. - We have a choice as to whether to deliver sharding first or the merge first, which reduces risk in case we find an issue with one of them. The overall time to deliver is shorter (but we're delivering less). - The layer&nbsp;1 protocol is greatly simplified. No more worrying about cross-shard transactions and all that jazz. They are now somebody else's problem. This is generally [a good thing](https://vitalik.ca/general/2018/08/26/layer_1.html). Some things that make me nervous: - Rollup technology is in its infancy, and there's no guarantee that it will work out well. In the worst case, we would probably need to bring back executable shards into Eth2. - A rollup-centric world---a world in which lots of protocol type stuff is kicked out of the protocol---will be much more chaotic in the short term. However, not being part of the base layer will enable it to iterate rapidly, and market forces should eventually lead to the adoption of standards for interoperability, tooling, tracking, and so on. But in the mean time, it's going to be a bumpy ride. - Cheap layer&nbsp;1 transactions are probably gone for good, and Eth2 as currently planned is not going to change that. This will be surprising and disappointing news to many people. Anyway, those were some things I thought y'all ought to know. Please support your favourite rollup developers! ## The Beacon Chain All remains quiet on the Beacon Chain front. Just chugging along with 87,725 active validators, and 8,000 in the activation queue. The queue is only 9 days long now, having reached around 21 days a while back, so the pace of new deposits has slowed. Nonetheless, there's over 3 million ETH staked, worth some $5 billion. There has been one curiosity: Justin Drake noticed that six validators have performed [voluntary exits](https://beaconcha.in/dashboard?validators=5650,23210,24388,35051,62802,62807) on Mainnet. This is a little bit of a surprise. They haven't been slashed, but a voluntary exit message has been submitted for each of them, so they are no longer active as validators. But the staked ETH remains locked until the Eth1--Eth2 merge has been executed. Afri has published another set of [syncing performance benchmarks](https://dev.to/q9/ethereum-2-0-mainnet-clients-3and) for the clients. It's a nice piece of work, but unfortunately focuses on totally the wrong metric. We don't optimise for sync speed at all in Teku as, in a weakly-subjective proof of stake world, very few people will be syncing from genesis. In fact, it will be dangerous to do so. See Meredith Baxter's [excellent video explainer](https://www.youtube.com/watch?v=Oz3flfj50Ig) of long-range attacks for insights into this. Instead, we've implemented an instant snapshot sync that gets you fully up and running in under a minute. This is much, much better. But unfortunately doesn't show up in the benchmarks :man-shrugging: ### Slashings The only real excitement has been a spate of slashings. Staked.&#x200D;us had 75 validators slashed on the 3rd of February, and wrote [a postmortem](https://blog.staked.us/blog/eth2-post-mortem). This elicited [some](https://twitter.com/superphiz/status/1357025776181391363) [sympathetic](https://twitter.com/preston_vanloon/status/1357027569783234563) [tweets](https://twitter.com/matterhorn_eth/status/1357062960510881792). I have little sympathy. Two days later they had another 17 validators slashed, so far unacknowledged as far as I can see. Also recall that Staked had 16 validators slashed back in December, also not publicly acknowledged. I was going to misquote Oscar Wilde at this point, but, honestly, this is beyond a joke[^fn0]. [^fn0]: We offered last year to help Staked design their Eth2 staking set-up. This is how the CEO refused our offer, "Having done this for three years, maybe we should be charging you?" Seems like it would have been money well-spent :man-shrugging: By contrast, here's an [insight from Attestant](https://www.attestant.io/posts/upgrading-attestants-infrastructure-without-missing-a-beat/) into how a well-architected staking service can maintain the highest possible uptime with zero risk of slashing. If you'd like to learn more about slashing, and how we are working to protect clients against it, then take a look at the Ethereum Cat Herders' [Peep an EIP #22: EIP-3076](https://www.youtube.com/watch?v=-z6OQNTAF7Y) session. This is a great walk through of the standardised slashing protection interchange format that client teams are adopting for user convenience and safety (currently supported in Lighthouse and Teku; soon to be supported in Nimbus and Prysm). There's a bonus last 10 minutes on the how the [Eth1--Eth2 merge](https://twitter.com/poojaranjan19/status/1357735802403500032) might be managed. ### Upgrade HF1 There's an open issue to come up with a [naming convention](https://github.com/ethereum/eth2.0-pm/issues/202) for future Beacon Chain upgrades. Pitch in if you have ideas![^fn2] (There are only two difficult things in computer science: cache invalidation, naming things, and off-by-one errors.) [^fn2]: My vote is for [British synth-pop band names](https://en.wikipedia.org/wiki/Category:British_synth-pop_new_wave_groups)[^fn3] [^fn3]: I was disturbed to discover this week that the band I grew up knowing as Yazoo was known as "Yaz" in the US :exploding_head: The [proposed updates](https://notes.ethereum.org/@vbuterin/HF1_proposal) for HF1 were discussed and agreed on the recent devs' call. Most of it is straightforward, but there were two bonus items initially proposed: (1) accounting reform tier 2, and (2) a proposal to put a hard cap on the number of active validators at any one time. We agreed to adopt the first, but not the second (it's basically moot for now, since it will take a minimum of a year to onboard enough validators to reach the proposed cap). The [document](https://notes.ethereum.org/@vbuterin/HF1_proposal) has been updated to reflect this. We also agreed to adopt two changes to the fork choice rule to mitigate different theoretical attacks. Adrian of the Teku team wrote [a nice analysis](https://hackmd.io/@ajsutton/balancing_attack) of what implementing the second change (defense against the "balance attack") would look like in a client. ### Tooling The Ethstaker team put up a 24/7 [client live stream](https://www.twitch.tv/ethstaker/) on Twitch. This is intended help new Ethereum 2.0 users see whether their clients are in sync or not by comparing the client logs. And it's simply cool to watch. Unfortunately, the stream looks to be down right now. But, when it's working, it cycles through the consoles of the four beacon chain clients, and Geth as well. Infura's [Eth2 API](https://twitter.com/Consensys/status/1356634209016774657) is now in public Beta. Also available [for Pyrmont](https://twitter.com/infura_io/status/1359614162154921985). The Nimbus team has [extended the deadline](https://twitter.com/ethnimbus/status/1360273972944920582) for submissions to their [dashboard challenge](https://our.status.im/nimbus-grafana-challenge/). You now have until February 28th, and up to $5000 of prizes are available for the best submissions. Why not take a run at it? Grafana is fun! Jim McDonald's `chaind` utility for extracting Eth2 data into a SQL database is now at [version 0.2.0](https://github.com/wealdtech/chaind/releases/tag/v0.2.0). Jacek has a [weekend project](https://twitter.com/jcksie/status/1359043872861650948) called [ngui](https://github.com/status-im/nimbus-eth2/pull/2304) for exploring Nimbus's RPC API. Bitfly [announced](https://twitter.com/etherchain_org/status/1357337769756880901) that their truly excellent Beaconcha.in mobile app is now [open source](https://github.com/gobitfly/eth2-beaconchain-explorer-app). The [Blst library audit](https://research.nccgroup.com/2021/01/20/public-report-blst-cryptographic-implementation-review/) is complete. No high severity or critical issues found. The full report is a very nice read, if you enjoy that kind of thing. Finally, polynomial commitments are hot. Alex Stokes is working on a KZG commitments library [in Rust](https://github.com/ralexstokes/kzg). Meanwhile, I am working on the same thing [in C](https://github.com/benjaminion/c-kzg). (Because I'm hardcore. And I really enjoy debugging [segmentation violations](https://www.youtube.com/watch?v=Fm5Ust7vEhk)[^fn1].) We're both building on Blst. [^fn1]: I am indebted to Mamy for sharing the Program in C song with me :joy: ## Articles and Media A very quick, [eth2 quick update no. 22](https://blog.ethereum.org/2021/02/11/eth2-quick-update-no-22/) from Danny. The awesome, intelligent, and good looking Adrian Sutton from the Teku team has some [Hard Truths for ETH Stakers](https://www.symphonious.net/2021/02/13/hard-truths-for-eth-stakers/). This is a thoughtful counterbalance to all the hype. Being Australian, Adrian is not shy about telling it like it is. A reminder to subscribe to Superphiz's [State of the Stake]() updates: [playlist here](https://www.youtube.com/watch?v=VYvuUd2JktA&list=PLOB9GGXGcc31_rKi1PlG0kGYf35e6l1wy). Phiz is always worth listening to, and the [Ethstaker](https://ethstaker.cc/) community remains wonderful. Viktor Bunin of Bison Trails reviewed the slashing incidents mentioned above, among other Eth2 topics, in his latest excellent [eth2 updates](https://bisontrails.co/eth2/010/) article. As mentioned last time, my former nemesis, Coindesk, is now producing some quality Eth2 material. The Valid Points newsletter is first rate ([3 Feb](https://www.coindesk.com/valid-points-coindesks-ethereum-2-0-validator-staked), [10 Feb](https://www.coindesk.com/valid-points-ethereum-2-0-validators-earn-record-1-2m-price-rallies)) - now that the professionals have entered the room I expect to be out of a job shortly. In addition, Christine Kim and Will Foxley have started a new podcast [Mapping Out Eth 2.0](https://www.coindesk.com/podcasts/mapping-out-eth-2-0). In [Episode 1](https://www.coindesk.com/podcasts/mapping-out-eth-2-0/eth-staking-challenges-podcast) they talk to Spencer Beggs, CoinDesk Director of Engineering, about setting up their own [Eth2 validator](https://beaconcha.in/validator/90969) (!). [Episode 2](https://www.coindesk.com/podcasts/mapping-out-eth-2-0/eth-2-0-development-roadmap) is with Danny Ryan. I'm sorry to say that ETHDenver more or less passed me by. There are [a lot of videos](https://www.youtube.com/c/ETHDenver/videos?view=0&sort=dd&flow=grid) of the talks. The Eth2 talk that caught my eye is Daniel Hwang of Stake Fish on [Beacon Chain Decentralization & Transparency](https://www.youtube.com/watch?v=k10pTecu_j0). I've chatted with Daniel about this previously, and he makes a lot of sense. ## Research The Ethereum Foundation [Research Workshop](https://hackmd.io/@hww/workshop_feb_2021) took place on February the 2nd. Here are direct links into the video for the various presentations: - [Intro](https://www.youtube.com/watch?t=520&v=uGeIDNEwHjs) - Dankrad on [KZG polynomial commitments](https://youtu.be/uGeIDNEwHjs?t=746) - Vitalik with [Eth2 Sharding Design](https://youtu.be/uGeIDNEwHjs?t=3959) - Danny, Mikhail, and Guillaume on [The Eth1--Eth2 merge](https://youtu.be/uGeIDNEwHjs?t=7450) On reflection, reversing the order of the presentations might have worked better. Very few of us need to know how KZG commitments work, any more than we need to know the innards of hash functions: they are just a black box for the most part. In any case, the whole event was excellent, and well worth your time. ## Implementers' call Call #57 took place on the 11th of February. * [Agenda](https://github.com/ethereum/eth2.0-pm/issues/203) * [Video](https://youtu.be/z3Gj6TXgcb0?t=208) * My [quick notes](https://hackmd.io/@benjaminion/H1p6GhG-d) and [live tweets](https://twitter.com/ralexstokes/status/1359935655803899905) from Alex Stokes. ## In other news The final Beacon Chain launch POAPs are [now available](https://twitter.com/superphiz/status/1358150517130747906). Claim them [here](https://poap.delivery/) if you are [eligible](https://beaconcha.in/poap). The Ethereum Foundation has [announced the grantees](https://blog.ethereum.org/2021/02/09/esp-staking-community-grantee-announcement/) in its recent [Eth2 Staking Community Grants](https://ethereum.org/en/eth2/get-involved/staking-community-grants/) round: over $1 million awarded to 25 projects. ## Upcoming events - February 20th, maybe 8pm UTC: I think I've agreed to appear as the guest on an Ethstaker call. I'll tweet out the details when they're nailed down. - February 28th: new deadline for submissions for the Nimbus [dashboard challenge](https://our.status.im/nimbus-grafana-challenge/). * * * [![[Twitter]](https://benjaminion.xyz/newineth2/img/twitter.svg =40x40)](https://twitter.com/benjaminion_xyz) Follow me on [Twitter](https://twitter.com/benjaminion_xyz) to hear when the next edition is out 🙌. [![[RSS]](https://benjaminion.xyz/newineth2/img/rss.svg =32x32)](https://benjaminion.xyz/newineth2/rss_feed.xml) We also have an [RSS feed](https://benjaminion.xyz/newineth2/rss_feed.xml). [Advertising](https://hackmd.io/@benjaminion/eth2_news/https%3A%2F%2Fhackmd.io%2F%40benjaminion%2Fadvertising) on this newsletter. [Give Feedback](https://docs.google.com/forms/d/e/1FAIpQLSfkESc4CmNfRGHHjWfNeF3ceLwrXDvynetda4sKfJFJ71Oabw/viewform).