One comment about delegation if we end up going down the MFS route: delegators might want the ability to stake to multiple nodes.
Only mentioning this because although “staking to multiple nodes” feels like a detail unrelated to the core protocol mechanics, it’s a significant change under the hood (with the corresponding implementation complexity), so I wanted to highlight the reasons for doing so, from a delegator experience standpoint.
Under MFS, delegators will need to be active and shift stake around as the fee distribution changes. Since the equilibrium is to have stake in proportion to fees, delegators will probably want the ability to stake to multiple nodes at once, otherwise it becomes something of a collective bin-packing game to get the most out of a given chunk of stake. If stake needs to be shifted around multiple wallets to optimize a delegator’s rewards, each of which needs to be filled with gas, etc then the delegator UX becomes quite difficult.
To make delegation effective in securing network value under MFS, the underlying implementation needs to be structured to facilitate that activity. Allowing a single delegator to stake multiple nodes seems like the way to go here.
It’s probably much cheaper and less risky to consider multi-staking within the scope of the 2.0 changes from the outset, than to tack this on after launch if we’ve realized that single-staking is a bottleneck in getting delegators to act in the way that we want them to.