pull down to refresh
Having a protocol that allows anyone to run a coordinator doesn't automatically give you decentralization.
Exactly. For mostro, I was thinking that in theory we can push this concept a little on the client-side (on an alt client maybe) because I think that the coordinator can be abstracted away in the orderbook as an entrypoint, and instead make it a re-ranker dimension. Could even WoT that to some extent.
I need my stats tool to work though - and that's not done - because I want to have a better idea about what's really happening and am curious how many of those announced coordinators are actually serving trades - and how many are self-dealing (i.e. trading on their own coordinator, because that would be crap in terms of dispute resolution)
reply
Yes, Mostro is actually one of the experiments I had in mind.
And I think your concern about consolidation is the important one.
Having a protocol that allows anyone to run a coordinator doesn't automatically give you decentralization. If eventually most liquidity and users converge around a handful of well-known coordinators, you've recreated obvious pressure points — just at a different layer.
The interesting outcome would be exactly what you describe: hundreds or thousands of coordinators, where spinning up a new one is cheap, permissionless and normal, and losing one doesn't really matter to the network.
Whether users and liquidity naturally converge instead is probably the real experiment.
Maybe the hardest part isn't making P2P permissionless at the protocol level. It's preventing convenience, reputation and liquidity from centralizing it again at the social layer.