pull down to refresh

Bitcoin Security Budget: Creating Asymmetries for 51% Attack Resistance
317 sats \ 8 comments \ @klk 18h bitcoin
Maybe there's no problem at all - transaction fees might naturally grow to sponsor a robust security budget as Bitcoin adoption increases. But if the security budget does become insufficient as block subsidies fade, I explored whether we could create asymmetries that make 51% attacks economically unfavorable without requiring hard forks.
The Failed Attempt
The idea was to implement delayed fee mechanisms where transactions pay fees across multiple blocks - part released immediately, part unlocked as subsequent blocks build upon it. The theory: create different incentives for different miners, making honest mining more profitable than attacking.
For example: instead of paying 10k sats upfront, you'd pay 4k + 4k + 2k across three blocks. The thinking was that later miners would have extra incentive to build on previous blocks rather than orphan them, since they'd only receive those delayed fees by mining honestly.
Why It Doesn't Work
If this were to become the norm, evil miner would end up collecting those extra fees from the previous blocks anyway.
The Real Question
Is it possible to create an additional asymmetry that allows for the same level of protection with a lower security budget?
write
preview
related posts
100 sats \ 7 replies \ @optimism 14h
The asymmetry is already there: bitcoin has extrinsically funded security because of proof of work, so miners have liabilities that do not need to be expressed in sats, but in something else (most often: fiat). This works in the miners' favor too: to convert an abundant resource (fiat, nuclear or solar energy) into a scarce asset, the liability is best to not be sats in the first place, because the sats aren't abundant.
Because of the asymmetry through the exchange rate between the liability and the reward - currently mostly BTC/<fiat>, but note that it could also become <energy>/BTC on a Bitcoin standard - there is an element inside the "security budget" that can be adapted to any reduction on the reward side. Since miners collectively control the liquidity of newly minted coins and fees, they have a large influence on these exchange rates.
So, with diminishing returns in absolute amounts of sats, the BTC/<fiat> miners need the exchange rate to go up and they have some influence over this, or in the case of an <energy>/BTC pair, the cost of energy has to go down.
If we assume that currently a majority of miners have fiat liabilities, the chart to watch is USD rewards per block:
reply
0 sats \ 6 replies \ @klk OP 14h
Pretty cool chart.
That being said, unfortunately the problem is something else. The asset that the miners protect also goes up in price/value in that process. And the existing ASICs go down in price as they become unprofitable.
So the security budget should be expressed as a percentage of whatever they are protecting. You can measure it in sats or fiat, since the unit is in both the numerator and denominator, it goes away.
If you have a safe full of gold, and your security guards get their salary paid in gold but constantly halved, your gold safe becomes less secure with time even if gold price goes up. Maybe never unsecure enough, but it's undeniable that the security budget decreases relative to the “jackpot” size.
#996005
reply
0 sats \ 5 replies \ @optimism 14h
Block work is cumulative (that's why we call it a block chain), so all this means is that you have to wait more confirmations before you consider something safe, and actually monitor getchaintips - which you should already be doing but realistically, there's a 99% chance that a reader of this comment doesn't even know what the hell that is.
You want to be keeping a real eye on reorgs the more sats you have at stake in recent blocks, and the lower the block reward, the more you want to keep an eye out. If this is what you mean, you are correct. You should also do it today though. If you don't, your complacency will come back to bite you when you get doublespent.
reply
0 sats \ 4 replies \ @klk OP 14h
Exactly.
That's why it would be great if there were a way to introduce an additional asymmetry so that we don't have to go through that. If it's not possible, other solutions will come up.
Or maybe it's not even needed because block fees alone become enough.
reply
0 sats \ 3 replies \ @optimism 14h
You already have to go through this. If you don't, you're still a slave.
reply
0 sats \ 2 replies \ @klk OP 14h
It's not binary but gradual. If double spend attacks or just casual forks happened all the time, you would have to rise your thresholds. That is annoying and bad for the network.
view all 2 replies