Jesse here, commenting both on behalf of Variant and separately, in my role as a delegate.
I and Variant decided to vote “no” on the Uniswap v3 deployment to BSC.
Earlier I commented that it’s suboptimal for communities to be voting on singular bridges, and that the v3 BSC discussion has highlighted a valuable opportunity to rethink deployment design from first principles. Luckily, the ball is already rolling:
- The Uniswap Foundation set up a new cross-chain bridge assessment process for evaluating modular deployment designs
- Martin proposed an additive security model that integrates multiple bridges
- Hyperlane detailed their design for a modular multi-bridging architecture that wouldn’t require building out exhaustive infrastructure, and is easy to change. Their protocol provides a standard format for incorporating any existing bridge into individual “security legos”, which are composable and thus can be combined in any way to create a validating condition before an interchain message is accepted.
- For example, you can build out these “legos” for Wormhole, L0, and Celer, and then use Hyperlane to combine these to create the condition “message X is valid IFF all of these 3 ISMs have confirmed the message”.
- Zefram built a universal bridge that can wrap multiple bridges simultaneously, to reduce security assumptions tied to any single bridge
Uniswap is a leader within the broader ecosystem, and the process here has the potential to set a standard for new chain deployments. There’s good progress in the works for building more modular deployment solutions, and I think it’s better for the protocol to fully explore those – and other emerging solutions – before finalizing a v3 BSC deployment.
Disclaimer: Variant is an investor in UNI tokens and Hyperlane. Disclosures – Variant