4 ms·

The author seems to argue that the Bitcoin hard cap (i.e. the 21M limit) is as inalienable as a law of nature. But it's not. Even if it's in the code itself, it

The author seems to argue that the Bitcoin hard cap (i.e. the 21M limit) is as inalienable as a law of nature. But it's not. Even if it's in the code itself, it's made by humans who can build consensus around a change in the future. The decentralized nature of Bitcoin might help prevent or delay this, but if there are sufficient incentives for a majority of miners/contributors, changes to rules could occur.

asah 4y ago
Sure in theory. In practice, the kinds of people attracted to Bitcoin strongly correlate with this message, and attempts to dilute BTC would likely result in a fork instead. Besides, what's the point? Satoshis (sats) are 100 millionth of a bitcoin, stop just do business in sats if BTC is too large for your transaction...
- lucozade 4y ago
  It may well fork under those circumstances but we've seen, multiple times, that a fork isn't terminal. You just usually end up with a dominant fork and the others tend to wither. So the real question is, would the dominant fork be the one that ends mining or the one that continues it? And that would likely be answered by which the the power brokers in Bitcoin-land will support ie the mining pools. So, for your contention to be correct, we would have to believe that the mining pools would willingly give up their source of power because of Satoshi's vision. It's not impossible for that to happen, of course. But the whole history of just about everything suggest that it's not the most likely outcome.
  - cesarb 4y ago
    > So the real question is, would the dominant fork be the one that ends mining or the one that continues it? And that would likely be answered by which the the power brokers in Bitcoin-land will support ie the mining pools. That's a common misconception, but the ones who matter to decide which will be the dominant fork are the so-called "economic nodes", not miners. Nowadays, the most important "economic nodes" seem to be the exchanges (after all, Bitcoin is useless unless you can convert it to something else, be it goods, services, or fiat money, and the later seems to be its most important use nowadays).
- rspeele 4y ago
  The hard limit on BTC requires mining to eventually be funded entirely by transaction fees. It's unclear whether this will work. Is there any successful PoW currency that runs on transaction fees alone now? How sure are we that this model pays enough to miners to keep them outspending a hypothetical attacker? For PoW to remain secure, mining honestly must always be more profitable than mining dishonestly. In other words, the total amount paid to miners over a given period (say, 2 hours) must be greater than the potential profit from a 51% attack over that same period. At present, through block rewards and fees miners are getting about $1.5 million every 2 hours. Due the the competitive nature of mining they are probably spending nearly that much in aggregate, on rent, hardware, and electricity. That's a fairly significant amount of spending to match, if you wanted to out-hash them and perform a 51% attack. However, only 1-2% of that is from fees. Almost all of it is newly minted coins. If the current level of transaction fees was all that was securing Bitcoin, we would be talking about only $30,000 of hashing spend per 2 hours and a 51% attack would look pretty darn feasible. I've heard it said that users will pay more in fees as the coinbase reward shrinks to zero, but there doesn't seem to be any reason for users to increase their fee payment to support the security of the network. Each user will spend just enough to get their transaction included in a block. As long as there is not much competition for block space, there is no need to pay more for fees, security be damned. So far in the history of BTC halvings we have seen fees rise very slightly, but not nearly enough to replace the lost block reward. Mining revenue (in $) dropped sharply following the last halving, and what restored it was not users picking up the slack in fees, but the price of BTC increasing, making the 6.25 BTC minted reward worth more. Obviously that same effect doesn't help you when the minted reward goes to zero. ---- Another way of thinking about it is that with a minted block reward, HODLers are "taxed" (in the form of dilution) to keep supporting the security of the network. In a 21M coins, no-more-minting, fees-only model HODLers expect to get their security for free, fully supported by those suckers actually transacting in Bitcoin. I'm not convinced it will work out.
boltzmann-brain 4y ago
> it's made by humans who can build consensus around a change in the future Good luck herding all of those kittens into one basket. There's a smooth 30 million bitcoin wallets out there. This is such a big issue that there exist cryptocurrencies with on-chain governance that replace hard forks with voting, such as Tezos (which I worked on, disclosure, yadda yadda).
- super256 4y ago
  > Good luck herding all of those kittens into one basket. There's a smooth 30 million bitcoin wallets out there. Irrelevant for consensus. You need miners and network nodes to update their rules, not "wallets". The wallet software itself isn't concerned about the coin emitting rules.
  - boltzmann-brain 4y ago
    i didn't say /wallet software/
    - super256 4y ago
      > There's a smooth 30 million bitcoin wallets out there. What else do you mean? Those 30m bitcoin wallets are irrelevant consensus.
olalonde 4y ago
It's happened tons of time already (e.g. most shitcoins). It's just not Bitcoin.

asah 4y ago

Sure in theory. In practice, the kinds of people attracted to Bitcoin strongly correlate with this message, and attempts to dilute BTC would likely result in a fork instead. Besides, what's the point? Satoshis (sats) are 100 millionth of a bitcoin, stop just do business in sats if BTC is too large for your transaction...

lucozade 4y ago
It may well fork under those circumstances but we've seen, multiple times, that a fork isn't terminal. You just usually end up with a dominant fork and the others tend to wither. So the real question is, would the dominant fork be the one that ends mining or the one that continues it? And that would likely be answered by which the the power brokers in Bitcoin-land will support ie the mining pools. So, for your contention to be correct, we would have to believe that the mining pools would willingly give up their source of power because of Satoshi's vision. It's not impossible for that to happen, of course. But the whole history of just about everything suggest that it's not the most likely outcome.
- cesarb 4y ago
  > So the real question is, would the dominant fork be the one that ends mining or the one that continues it? And that would likely be answered by which the the power brokers in Bitcoin-land will support ie the mining pools. That's a common misconception, but the ones who matter to decide which will be the dominant fork are the so-called "economic nodes", not miners. Nowadays, the most important "economic nodes" seem to be the exchanges (after all, Bitcoin is useless unless you can convert it to something else, be it goods, services, or fiat money, and the later seems to be its most important use nowadays).
rspeele 4y ago
The hard limit on BTC requires mining to eventually be funded entirely by transaction fees. It's unclear whether this will work. Is there any successful PoW currency that runs on transaction fees alone now? How sure are we that this model pays enough to miners to keep them outspending a hypothetical attacker? For PoW to remain secure, mining honestly must always be more profitable than mining dishonestly. In other words, the total amount paid to miners over a given period (say, 2 hours) must be greater than the potential profit from a 51% attack over that same period. At present, through block rewards and fees miners are getting about $1.5 million every 2 hours. Due the the competitive nature of mining they are probably spending nearly that much in aggregate, on rent, hardware, and electricity. That's a fairly significant amount of spending to match, if you wanted to out-hash them and perform a 51% attack. However, only 1-2% of that is from fees. Almost all of it is newly minted coins. If the current level of transaction fees was all that was securing Bitcoin, we would be talking about only $30,000 of hashing spend per 2 hours and a 51% attack would look pretty darn feasible. I've heard it said that users will pay more in fees as the coinbase reward shrinks to zero, but there doesn't seem to be any reason for users to increase their fee payment to support the security of the network. Each user will spend just enough to get their transaction included in a block. As long as there is not much competition for block space, there is no need to pay more for fees, security be damned. So far in the history of BTC halvings we have seen fees rise very slightly, but not nearly enough to replace the lost block reward. Mining revenue (in $) dropped sharply following the last halving, and what restored it was not users picking up the slack in fees, but the price of BTC increasing, making the 6.25 BTC minted reward worth more. Obviously that same effect doesn't help you when the minted reward goes to zero. ---- Another way of thinking about it is that with a minted block reward, HODLers are "taxed" (in the form of dilution) to keep supporting the security of the network. In a 21M coins, no-more-minting, fees-only model HODLers expect to get their security for free, fully supported by those suckers actually transacting in Bitcoin. I'm not convinced it will work out.

boltzmann-brain 4y ago

> it's made by humans who can build consensus around a change in the future Good luck herding all of those kittens into one basket. There's a smooth 30 million bitcoin wallets out there. This is such a big issue that there exist cryptocurrencies with on-chain governance that replace hard forks with voting, such as Tezos (which I worked on, disclosure, yadda yadda).

super256 4y ago
> Good luck herding all of those kittens into one basket. There's a smooth 30 million bitcoin wallets out there. Irrelevant for consensus. You need miners and network nodes to update their rules, not "wallets". The wallet software itself isn't concerned about the coin emitting rules.
- boltzmann-brain 4y ago
  i didn't say /wallet software/
  - super256 4y ago
    > There's a smooth 30 million bitcoin wallets out there. What else do you mean? Those 30m bitcoin wallets are irrelevant consensus.

olalonde 4y ago

It's happened tons of time already (e.g. most shitcoins). It's just not Bitcoin.