OpenGov and treasury is a good thing and we need to use it properly.
We as a kusama community want to see as much as possible use cases on parachains and as much as possible burning of kusama with core time usage.
The best way to turn the table and create good vibe around kusama is to incentivise parachains from treasury depends how many KSM coins are on sovereign accounts of that parachains. This is only one metric that can't be abused.
For examle there are 3 chains on kusama: A, B, C chain. (during period of 28 days)a
A contains on severign account 10K ksm
B contains on severign account 70K ksm
C contains on severign account 20K ksm
So in total 100K KSM bridged from Kusama relay chain to rollups. Each of them will be rewarded proportionally on the sovereign accounts.
10% to A, 70% to B, 20% to C.
KSM for insentivice is from treasury and could be any what KUSAMA open gov will decide. for example
— % from all KSM in treasury.
— % from KSM collected during month,
— % from KSM that suppose to be burned as not spended.
The distribution to parachain/rollups could/should be done automaticly depends how many KSM on sovereign account was each block during the some period.
It could be done via extra pallet that check balances on sovereigns accounts each block and distribute KSM incentives to that accounts proportionally for total KSM during spending period duration.
This is the best and the fastest way to drive development activity and real product on kusama.
As open gov we still can try to fund some pallet developments, marketing but at the end of final use case for relay chain native currency is bridging as much as possible KSM to parachain/rollups.
To not let abuse the system of insentivise and not drive up only LSD via parachain/sovereign account — could be good to just apply some formula to that balance of sovereign parachain account that pull down total share of such LSD parachain use cases.
Threshold
I don't think that's a good metric, different parachains will have very different business models that use capital in different ways, there could be parachains that are more beneficial to the KSM economy that don't care about this metric while others that don't have any utility to the KSM will benefit form this policy as they will be built to incentivize people to hold for the right amount of time. You might end up incentivizing the wrong kind of behavior! perhaps a way to distinguish user's KSM from the parachain's hard earned KSM could help.
e.g. What is better, a "MemeParachain" that collects fees in its own token that tells whales, I'll give you tons of meme-coin rewards if you lock your KSM for a month(potentially lots of KSM in the sovereign account with 0 benefit to the ecosystem)? or a parachain like Kreivo with KSM as native currency, that incentivizes its DAOs to keep their KSM in their own sovereign account(not the para's), this chain might end up collecting considerable amounts of fees in its treasury(and the para's account) from membership sales("prepaid gas") and goods&services payments as it takes a percentage of every payment but still might pale in comparison to the whale collecting meme coins example.
If we were to do something like what you propose I would like to distinguish between a para's own hard-earned KSM in its own treasury from their users' KSM, but it's quite challenging! we can't trust the balances of accounts inside the parachain. A good approximation could be how much of that free balance in the sovereign account the para is willing to lock for a long enough amount of time to get rewards, we already have a system for that, it's called staking 😜 and what you propose can probably be an improvement/extension to the staking system.
A parachain can already stake funds with its sovereign account, a few paras do this with liquid staking systems where their users know their funds will not be available for a while, if done arbitrarily users would be pissed if suddenly they can't move their funds. Liquid staking probably doesn't earn the parachain much so it's not great incentive for bringing adoption, to be more aligned with your proposal, I think we would need some kind of "super staking", a way to get better/extra rewards perhaps by locking funds for longer periods of time or under certain conditions that are not attractive to regular users but could make sense for a parachain's KSM treasury, the rewards can come as part of the inflation or from a dedicated pool(the treasury counts), we can also restrict it to specific origins like a parachain's origin, it's relatively easy to setup a 1 user parachain just to get better rewards but at least people would have to be paying coretime while making core more scarce.
As an experiment I don't hate it, I could implement it and propose it via the fellowship as an experimental "extra staking rewards pallet" usable only by parachain origins, should be an interesting experiment :) However in the long run might have undesirable side effects as it encourages parachains to collect and hold KSM in treasuries that wouldn't much different from burner addresses, instead of creating lively circular economies where money moves and is spent on regular basis.