Notes from the Evolving Web3 Infrastructure panel at AraCon
Although I decided to take a break from traveling this month, and for that reason pass on attending AraCon (but the FOMO is real), I've enjoyed watching the livestream from the comfort of my couch.
Yesterday (day 1), I watched the Evolving Web3 Infrastructure panel. It was moderated by Evan van Ness and featured the panelists Jutta Steiner (Parity Technologies), Lane Rettig (Ethereum Foundation), Martin Köppelmann (Gnosis), Adan Sanchez de Pedro (Witnet, Stampery).
Here are a few takeaways and highlights from the panel.
What belongs on-chain and what doesn't?
- Lane: The jury is still out but the answer lies in the trade off between putting as much on-chain as possible while disincentivizing people to use Ethereum as their personal hard drive. At the moment, we lack the economics which would allow the market to sort this out. If one takes a minimalist view, blockchains can be reduced to nothing more than a set of Merkle roots. Every thing else can live on Layer-2 or beyond. Technologies like zkSNARKs will pull us towards this direction.
- Martin: Hopefully sharding will bring about a more competitive fee market. He uses the analogy of living in a major city versus the countryside. New York City has high living costs but offers the benefit of high population density and more opportunities (high value transaction). On the other hand, one may chose to live on their own land in the Miwdest where cost of living is low (low value transactions).
- Adan: For privacy reasons, not every transaction belongs on-chain. Layer 2 solutions are valuable for this reason.
Referring to Dani Grant and Nick Grossman's post The Myth of The Infrastructure Phase, Evan posed the questions about which should come first to drive adoption, apps or infrastructure.
Overwhelmingly, the panelists expressed the need for better infrastructure.
- Jutta: Mentioned the importance of proper dev tools which make it easy for developers to build dapps.
- Martin: Made the excellent point that Ethereum is running at capacity. Any attempt to drive massive adoption on-chain at this point is futile as rising fees would cause many dapps to be priced out of the network.
- Lane: Ethereum remains a science experiment and being heads down in an infrastructure phase is ok.
What is most needed in terms of dev tooling?
- Martin: We need simple proof-of-data-availability solutions. The ability to verify if the data corresponding to a blockchain hash is or was available at a certain point in time would make many Layer 2 solutions much more feasible.
- Lane: No compelling storage at this point. We also lack a functional messaging layer (although teams like Status and the Ethereum Foundation are making progress).
- Adan: Proof-of-availability. Narrowing the gap between the undeterministic web and the deterministic nature of smart contracts.
Where does governance fit in the infrastructure of Web3?
- Lane: There are many schools of thought. Tezos is doing exciting work in the area of Layer-1 governance. Polkadot takes a different approach with partial on-chain governance and partial upgradability. Most of Ethereum takes the position that governance should happen off-chain.
- Adan: There is value in having projects which have no governance, like Bitcoin (although some would likely disagree with that statement). There is also value in protocols which can be upgraded by their users. It comes down to what makes sense for each individual use case.
- Martin: Ethereum needs significant improvements like proof-of-stake and sharding. He hopes that after this transition, Ethereum can emulate Bitcoin, becoming a stable foundation with slow and calculated improvements. This foundation would allow better Layer 2 solutions to be built.
Watch the full panel here.