Interview with Danny Ryan — Eth2 Researcher at the Ethereum Foundation
Last week, long-anticipated Medalla multi-client testnet went live! It represented the final point before the launch of the Ethereum 2.0 network, which is tentatively expected by the end of the year.
In light of these developments, we were thrilled to have an opportunity to interview Danny Ryan — Core Researcher at the Ethereum Foundation. He gave insights of Eth2 progress, talked about PoS and Casper sharding, shared his thoughts on the Ethereum future, the benefits of Eth2 for the community and challenges involved in bringing Ethereum 2.0 to life.
Tell us about your background. When did you first hear about crypto and Ethereum in particular? How did you become part of the project?
I heard about bitcoin in college and wasted a little bit of time day trading bitcoin and losing money. Although I understood bitcoin, it didn’t quite click and capture my interest. As for Ethereum, I learned about it from a friend before launch. It seemed interesting, but I still didn’t give it much more than a cursory look. Another friend texted me when The DAO made national news as it raised 10s of millions of dollars. I dug a little deeper and became quite captivated by the possibilities. I haven’t been able to pull myself out of the rabbit hole since. I started small, contributing code to random repos, but soon I got rid of all my clients (was a freelance dev) and decided to figure out how to make this my work. From there, I started a local New Orleans Ethereum meetup and did a deep-dive on proof of stake research. Eventually after a number of contributions, the EF hired me on to continue my work on a more formal basis.
Ethereum has all these phases coming. In brief, what should we actually expect from Ethereum 2.0 during the phases of the upgrade? And when do you expect Eth2 to go live?
If you haven’t had a chance to read my recent blog post, The State of Eth2, I highly recommend it. I try to give context on the project, discuss the trade-offs, the timelines, and the benefits over time. Specifically, check out the “Benefits of eth2 to the community over time” section. In short, Phase 0 bootstraps the system, Phase 1 provides a highly scalable data-layer that, through the use of eth2 light clients in Ethereum, can be leveraged by layer 2 constructions for scalability, Phase 1.5 is the unification of the Ethereum chain that we know and love as a shard under the upgraded eth2 consensus, and Phase 2+ is expanding the functionality of shards over time.
So there are a lot of the benefits of Eth2 to the community.
Better security properties through PoS, more layer 1 scale through sharding, and the elimination of energy intensive PoW which is also naturally centralizing due to asymmetric access to physical hardware.
The multi-client paradigm has many significant advantages (and also difficulties). Which ones?
Again, check out my recent post, specifically the “So many clients” section for my views on this. In short, it provides a wider exploration of the design and end-user space, as well as providing redundancies in the event of failures and attacks. The complexities come in additional coordination overhead, but we have many stellar client teams so that hasn’t been too much of a drag.
Could you please share with us the current running state of Eth2 clients and testnets?
We just launched the Medalla testnet. This testnet is composed of 20k+ validators on over 600 nodes distributed across the world and is greater than 80% run by the community. Client teams continue to harden their systems and work on UX and documentation, but the stabilities we are seeing on this production-grade testnet are signalling for a clear path to mainnet.
What are your thoughts on Proof of Stake leading to a wider and wider wealth gap over time?
PoS has a more pure mechanism to seek a return on capital no matter the size of the participating asset, whereas PoW has distinct economies of scale that allow those with wealth to compound it at a higher rate. Crypto-economic protocols do not seek to remove the notion of being able to make a return on capital, but proof of stake can make access to that return more equitable and widely distributed.
Casper Sharding is a combined effort between the migration to Proof of Stake and Sharding for Ethereum. Tell us about this approach, its difficulties, and why it’s worth it.
Both of these mechanisms are designed to enhance the system while retaining strong properties of decentralization. You can easily scale up a block chain through simpler mechanisms than sharding, but these mechanisms tend to reduce the ability for users to follow and participate in consensus with consumer hardware. Proof of Stake and Sharding, though, allow for a wide set of participants to contribute to the construction of a highly scalable protocol even with standard consumer hardware. There are plenty of difficulties in this path. Both PoS and sharding require sophisticated in-protocol mechanisms. It is tough to design and build these systems. In addition to that, the p2p network has to be “sharded” in a sense. If all nodes have to relay all gossip’d data about all shards, then we lose scale with respect to bandwidth. Fortunately, libp2p has a mechanism called “gossipsub” that we’ve been able to leverage to meet most our needs for this problem. Still, managing such a network is much more complex than a single chain paradigm.
In your point of view, what are the biggest challenges involved in bringing Ethereum 2.0 to life?
For a long time, turning abstract research into complex specs was a big challenge. At this point, it’s primarily engineering. As discussed above, the mechanisms here are complex, and due to the high value of the Ethereum network, we have to get them right out the gate.
As we know, the fundamental design decision that makes what you’re building so hard is that, in Ethereum, you choose to gain scalability without compromising on decentralization or availability. Tell us a bit more about that.
As discussed in the past few answers, both PoS and sharding require a sizable amount of consensus complexity but do achieve our goal of scale without sacrificing decentralization. This complexity ripples throughout the entire process — research, analysis, specs, testing, engineering, etc.
DeFi “yield farming” is one of the hottest topics in the crypto world so far. What is your opinion on this field? Do you believe in them?
I don’t have time to follow the application space too closely. I’m aware of what’s going on, but don’t have too much opinion here. I’m excited to see experimentation and innovation, but won’t make a claim on sustainability of this particular mechanism or anything.
Do you view other developing blockchain technologies like Polkadot or Cosmos as competitors to Eth2? Or as collaborators?
It’s a bit of a mix. I think that the blockchain ecosystem is going to continue to grow massively in the next 10 years, but I also am unsure of how many layer 1’s the market (and thus the security of these systems) can support. As seen with the repeated attacks of ETC recently, low value chains aren’t going to be able to thrive in the long run so I do expect some natural coalescence around only a small number of high value chains.
Besides Ethereum, what projects do you consider promising today?
My garden. I just harvested some tomatoes and peppers a couple of days ago and have some more veggies queued up.
How will the world be different once Ethereum 2.0 and crypto win?
More open access and transparency in the digital systems we interact with. This will lead to better designs, less chance for predatory practices, and more power for the user. Ideally we can disrupt the status quo and the current trajectory of the siloed, centralized internet in which we are constantly subject to manipulation and are in fact the product. I am deeply concerned about technology and society, and hopefully decentralized technologies become a tool for us to find a better equilibrium.
What Ethereum 2.0 means to you personally?
Fun, stress, community, years of my life.
One last question for prosperity. What is your favorite crypto meme for now?
I’m so out of the loop…
Follow Danny on Twitter, GitHub.
Subscribe to Paradigm!
