Elastic block gas

gistfile1.txt

assumptions 

	• miners and users desire stability in systems
	• block rewards will go down,  fee market must compensate 

	• humans setting the block gas limit is inefficient and can lead to perverse incentives 

	• setting arbitrarily large gas limits can bloat nodes

	• the optimal solution should be driven by the network utilization

	• Hard edges create targets for exploitation, mess was an exponential scale for this reason

	• 1559 style basefee + tip stabalizes fees

	• ETC will never burn base fees or tips due to hardcapped supply of coins

	• the community should have some control of growth

	• the target gas limit should optimize for some community defined percentage of block space utilization 

	• there is a hard floor and soft ceiling 

	• the community defines the goldilocks zone

	• as block rewards decrease, more fixed gas fee transactions are needed to offset the miner revenue

	• unpredictable gas fees are undesirable. fixed gas limits create unpredictable gas fees since higher fee transactions get included.

	

proposal

	• create an 'elastic gas limit' by using block reward incentives 

	• the ideal block gas limit should be set based on the exponential moving average of the blockchain 

	• if the current block utlizes gas that is outside the ema +- some % (defined by the community) the block reward is reduced exponentially. 

	• EMA ensures growth/shrink limit between blocks should be proportional to the trajectory of the network 

	• If a developer would like to include a very large transaction, a miner could be incintivized to accept the block reward being reduced to zero if the tip is high enough to offset the loss. all nodes should accept this as a valid block.

	• if a miner produces empty or low value blocks, they will forfit their block rewards. This encourages creating useful blocks.

	• The current block reward should be viewed as offsetting  the fee market until adaquate transactions volume exists. glide slope to tx market

goals

	• security - miners long-term business planning should target the fee market completely replacing the current block reward system. The number of transactions and gas usage overtime is knowable, more transactions more fees. 

	• stability - fixed base fees and predictable block sizes allow for long term infrastructure planning, development/deployment cost planning, and user transaction cost modeling

	• decentralization - letting the network usage define the block limit prevents collusion of actors

	

	

	

	

	

	

	

	

	

	

If EMA would be introduce, wouldn't that lead to gas limit instability in a situation when there are fluctuations in transaction volume? If "someone" manipulates transaction volumes to influence the gas limit, wouldn't that lead to network congestion or under utilization?

Could a threshold strategy for the gas limit, similar to bandwidth allocation in networking (bandwidth throttling - prioritization of certain types of traffic) help with transaction spikes by allowing temporary increases when needed, then bringing the limit back to a more sustainable level? That might create more predictability and mitigate bad behavior, also adding efficiency while still allowing flexibility, right?