Don’t be Ethereum, Centralisation then decentralisation, Avoid one-coin-one vote
A market crash focuses organisations to revisit their mission. The point of a blockchain is to offer an open platform that is always-on, always secure, and that one person (or small number of people) cannot control. As I have previously argued, many blockchains have drifted far from this premise – particularly by compromising to provide low transaction costs.
Here follow three reflections for L1s, from learnings during the last crypto bull cycle:
Don’t be like Ethereum
I subscribe to Cosmos founder Ethan Buchman’s philosophy that it is healthy to have multiple strong chains instead of an Ethereum empire or Bitcoin maximalism. However, each strong chain should be differentiated, otherwise it does not add to the overall robustness of the ecosystem.
I think Bitcoin and Ethereum are fallible as designs, but I like them both – and my crypto holdings are primarily BTC and ETH. Both are good platforms to build on. For those looking to be independent, I think L1s should seek to strongly differentiate from Ethereum in design (i.e. don’t use the EVM and, ideally, think of alternatives to PoS). Side-note: if an L1 is differentiating on other factors than EVM and PoS, then why not just roll-up onto Ethereum?
Some notes on non-EVM chains:
Cosmos is not EVM. Since Terra was built on Cosmos architecture, that had some collateral marketing and liquidity damange. I have also read criticism that Cosmos validators are challenging to run. Still, I need to read more about Cosmos to give it a fair chance.
Solana is not EVM. Solana has a good number and geographical split of nodes, but 70%+ are hosted on AWS. Solana also crashes a lot and the transaction history is not readily available.
There don’t seem to be strong alternatives to Ethereum that are differentiated. There aren’t even good Ethereum clones (they have far inferior security, although one would expect this from winner-take-all effects).
Centralisation then Decentralisation
Blockchain organisations start with highly motivated and centralised visionaries or groups (e.g. Satoshi or Vitalik) that put in place a lot of the core structure. Once the organisation decentralises (if it does at all), then progress slows and evolves more slowly. A worthwhile example is Liquity USD, where a centralised team had a vision and then deployed that vision immutably (with no governance) to Ethereum.
L1s expect to get too much done via decentralisation. That isn’t how these systems work. Big ideas need to be firmly in place in a centralised way, and then the system can be decentralised so that others can build on top of it. L1s that are not Ethereum are stuck in a place where they are trying to decentralise without having a firm set of differentiated primitives settled in the first place (or don’t have a realistic roadmap for decentralisation at all).
In short, I think one should only decentralise when the offering has been fully (or at least nearly fully) figured out. (Although I leave open the possibility that we find better ways to dynamically evolve decentralised offerings.)
Avoid one-coin-one vote
Bitcoin and Ethereum don’t have one-coin-one-vote. Their rules only change if there is adoption by a majority of nodes. Bitcoin goes even a step further and makes adoption largely backwards compatible (which comes with pros and cons).
Organisations that do have one-coin-one-vote end up – maybe not legally – but structurally, being like corporations. Our empirical experience with corporations over the last centuries is that they operate well with a CEO and with boards. That isn’t to say some other model can’t work under one-coin-one-vote, but it pays to be realistic on what history says.
I think there is a place for one-coin-one-vote (which is another way of saying I think there is a way for corporation-type structures), but I’m unsure this approach has a place in underpinning open censorship-resistant structures.
In short, what I see is working so far in crypto are:
a) projects that have a strong vision and put it out there immutably (or slowly mutable).
b) projects (often private companies) that build on those immutable primitives – often by providing a friendly UX layer.
Other crypto projects operate as headless corporations.