6 ms·

Bitcoin is not slow. It's fast, if you are prepared to pay for it. This is where the disagreement lies. Should the price of on chain transactions be reduced an

by hashmp 9y ago

Bitcoin is not slow. It's fast, if you are prepared to pay for it. This is where the disagreement lies. Should the price of on chain transactions be reduced and remain low.

nathanvanfleet 9y ago
The article mentions this
rebuilder 9y ago
That's one way of putting it. Another way to look at the blocksize issue is: should the supply of transaction space be limited by free market mechanisms alone, or by committee decision/voting/negotiations?
- hashmp 9y ago
  There is an argument that block propagation and orphan rates acts as natural barrier to the blocksize. With a very large blocksize limit, miners who produce blocks which are very big, run the risk of a smaller block from a competing miner winning the race, as it is faster to validate and propagate over the network. This limiting factor should mean the blocksize converges to a size that the network can comfortably handle.
  - dgacmu 9y ago
    This isn't fundamental. The large-block miner can pre-stage copies of the block (before the nonce has been found) at a few replicas. If they successfully find a nonce, they blast the nonce to all of their replicas, which immediately disseminate the complete block.
grey-area 9y ago
Compared to existing payment processing networks (responses typically in < 1 sec), it is slow as molasses.
- mrb 9y ago
  Apples to oranges. A 0-confirmation Bitcoin transaction similarly needs only < 1 sec to be broadcasted to the network. And a 0-conf tx offers as much security as existing payment networks (which are reversible). When I sell computer gear on Craigslist in person and someone pays me in Bitcoin, I only wait ~1 second for the tx to reach my phone's wallet before letting the buyer walk away. A merchant needs to wait 1-2 business days before he can spend the money he received through these payments networks. With Bitcoin he only has to wait 1 block (~10 minutes), reducing pressure on his cash flow. A merchant must wait typically 60 days for a CC transaction to be considered irreversible. With Bitcoin he only has to wait 6 blocks (~1 hour). So, comparing apples to apples, Bitcoin is either as fast as, or MUCH faster than legacy payment networks.
  - vesak 9y ago
    What is less clear, however, is the capability of the traditional payment networks to improve. Before bitcoin et al, they might have had no reason to. I don't find it very likely that the 60 days transaction speed, for instance, is unfixable. Some people just needs to get off their asses and dive deep into some horrific legacy code and/or processes and fix or rewrite them. If they are pushed hard enough, they'll do it.
  - grey-area 9y ago
    A 0-confirmation Bitcoin transaction similarly needs only < 1 sec to be broadcasted to the network. No, payment networks verify that the user can spend immediately. That is full confirmation in a few seconds. Bitcoin fails dismally here on very limited transaction volume by taking 10 minutes at best for something approaching confirmation and has no chance with the current architecture of POW of ever realistically replacing those transaction networks - both because of energy use and time. A merchant needs to wait 1-2 business days before he can spend the money he received through these payments networks. Yes, Bitcoin has the advantage here - you'll get your money faster, though there is no real technical barrier to existing networks paying out faster, the banks are just incompetent and greedy. This does depend how the payment is done - using FPS it's available instantly, using cc it's not (often more like 7 days, not 2). A merchant must wait typically 60 days for a CC transaction to be considered irreversible. That is a matter of policy, not tech. If Bitcoin were a serious transaction protocol, customers would demand the same features, or a guarantee in law that they could return shoddy goods - either way, the protocol is not the problem here. When I sell computer gear on Craigslist in person and someone pays me in Bitcoin, I only wait ~1 second for the tx to reach my phone's wallet before letting the buyer walk away. In that case you have no idea if that transaction is actually confirmed and not a double spend (typically Bitcoin requires at least minutes to confirm), that's not equivalent to them paying you by bank transfer in seconds. Comparing a zero confirmation transaction from bitcoin to a full confirmation transaction from a bank is disingenuous at best. That's not even touching on all the other flaws of bitcoin as a method of payment - irreversibility (yes people really don't want that!), pseudo anonymity (just enough for criminals, not enough to stop states), complexity meaning average users hand their entire savings over to someone else, lack of insurance or regulation (yes people really do want that!), lack of backing as a currency making it extremely vulnerable should it become popular to being targeted and cornered by states or money markets, hidden centralisation through miner cartels running the transaction network and a small group of devs/companies producing the complex software required, energy use of POW which means it could never handle enough volume.... Using merkle trees to store transaction ledgers is really interesting, but with the current architecture of bitcoin is not a rival to current payment networks mostly because it appears to be designed for something else entirely (to replace fiat cash it seems, which is on the way out anyway).

by hashmp 9y ago

Bitcoin is not slow. It's fast, if you are prepared to pay for it. This is where the disagreement lies. Should the price of on chain transactions be reduced and remain low.

nathanvanfleet 9y ago

The article mentions this

rebuilder 9y ago

That's one way of putting it. Another way to look at the blocksize issue is: should the supply of transaction space be limited by free market mechanisms alone, or by committee decision/voting/negotiations?

hashmp 9y ago
There is an argument that block propagation and orphan rates acts as natural barrier to the blocksize. With a very large blocksize limit, miners who produce blocks which are very big, run the risk of a smaller block from a competing miner winning the race, as it is faster to validate and propagate over the network. This limiting factor should mean the blocksize converges to a size that the network can comfortably handle.
- dgacmu 9y ago
  This isn't fundamental. The large-block miner can pre-stage copies of the block (before the nonce has been found) at a few replicas. If they successfully find a nonce, they blast the nonce to all of their replicas, which immediately disseminate the complete block.

grey-area 9y ago

Compared to existing payment processing networks (responses typically in < 1 sec), it is slow as molasses.

mrb 9y ago
Apples to oranges. A 0-confirmation Bitcoin transaction similarly needs only < 1 sec to be broadcasted to the network. And a 0-conf tx offers as much security as existing payment networks (which are reversible). When I sell computer gear on Craigslist in person and someone pays me in Bitcoin, I only wait ~1 second for the tx to reach my phone's wallet before letting the buyer walk away. A merchant needs to wait 1-2 business days before he can spend the money he received through these payments networks. With Bitcoin he only has to wait 1 block (~10 minutes), reducing pressure on his cash flow. A merchant must wait typically 60 days for a CC transaction to be considered irreversible. With Bitcoin he only has to wait 6 blocks (~1 hour). So, comparing apples to apples, Bitcoin is either as fast as, or MUCH faster than legacy payment networks.
- vesak 9y ago
  What is less clear, however, is the capability of the traditional payment networks to improve. Before bitcoin et al, they might have had no reason to. I don't find it very likely that the 60 days transaction speed, for instance, is unfixable. Some people just needs to get off their asses and dive deep into some horrific legacy code and/or processes and fix or rewrite them. If they are pushed hard enough, they'll do it.
- grey-area 9y ago
  A 0-confirmation Bitcoin transaction similarly needs only < 1 sec to be broadcasted to the network. No, payment networks verify that the user can spend immediately. That is full confirmation in a few seconds. Bitcoin fails dismally here on very limited transaction volume by taking 10 minutes at best for something approaching confirmation and has no chance with the current architecture of POW of ever realistically replacing those transaction networks - both because of energy use and time. A merchant needs to wait 1-2 business days before he can spend the money he received through these payments networks. Yes, Bitcoin has the advantage here - you'll get your money faster, though there is no real technical barrier to existing networks paying out faster, the banks are just incompetent and greedy. This does depend how the payment is done - using FPS it's available instantly, using cc it's not (often more like 7 days, not 2). A merchant must wait typically 60 days for a CC transaction to be considered irreversible. That is a matter of policy, not tech. If Bitcoin were a serious transaction protocol, customers would demand the same features, or a guarantee in law that they could return shoddy goods - either way, the protocol is not the problem here. When I sell computer gear on Craigslist in person and someone pays me in Bitcoin, I only wait ~1 second for the tx to reach my phone's wallet before letting the buyer walk away. In that case you have no idea if that transaction is actually confirmed and not a double spend (typically Bitcoin requires at least minutes to confirm), that's not equivalent to them paying you by bank transfer in seconds. Comparing a zero confirmation transaction from bitcoin to a full confirmation transaction from a bank is disingenuous at best. That's not even touching on all the other flaws of bitcoin as a method of payment - irreversibility (yes people really don't want that!), pseudo anonymity (just enough for criminals, not enough to stop states), complexity meaning average users hand their entire savings over to someone else, lack of insurance or regulation (yes people really do want that!), lack of backing as a currency making it extremely vulnerable should it become popular to being targeted and cornered by states or money markets, hidden centralisation through miner cartels running the transaction network and a small group of devs/companies producing the complex software required, energy use of POW which means it could never handle enough volume.... Using merkle trees to store transaction ledgers is really interesting, but with the current architecture of bitcoin is not a rival to current payment networks mostly because it appears to be designed for something else entirely (to replace fiat cash it seems, which is on the way out anyway).