Instant Bitcoin confirmation time (IDEA)

[madmadmax]

Assuming Bob wants to purchase a loaf of bread physically from Alice, Alice works in a supermarket yet Alice wouldn't use Bitcoins due to the lines getting extremely long because people are waiting for confirmations as well as having to physically keep her eyes on the customers so no one bails with the physical merchandise (or double spends), what solutions could we implement to allow for instant transactions?

My model is the following:
Bob has a total of 30BTC of transactions sent to his address (his "banking account"), Bob then purchases a loaf of bread for 1BTC (an extremely tasty loaf of bread) and walks out of the store, if the network detects that Bob has double spent the same inputs that he used to pay Alice with all the BTC that was sent to his address (his "banking account") is divided equally between all the double spent transactions to the outputs by the network. Ultimately leaving Alice with 15BTC because Bob tried to scam her.

If Bob tries to pay for a loaf of bread with less than 2BTC in his address, then Alice has a message that pops up so she has to ask Bob to wait for confirmations before walking out of the store.

Using this system we could both promote people to accept Bitcoin and not sacrifice anonymity or force Bitcoiners into bondage. Future improvements could allow for parents to sign their kids addresses so if their kids double spend the penalty BTC is taken from the parents account.

[1714649894]

1714649894

[1714649894]

1714649894

[1714649894]

1714649894

[1714649894]

1714649894

[1714649894]

1714649894

[DannyHamilton]

Assuming Bob wants to purchase a loaf of bread physically from Alice, Alice works in a supermarket yet Alice wouldn't use Bitcoins due to the lines getting extremely long because people are waiting for confirmations as well as having to physically keep her eyes on the customers so no one bails with the physical merchandise (or double spends), what solutions could we implement to allow for instant transactions?

My model is the following:
Bob has a total of 30BTC of transactions sent to his address (his "banking account"), Bob then purchases a loaf of bread for 1BTC (an extremely tasty loaf of bread) and walks out of the store, if the network detects that Bob has double spent the same inputs that he used to pay Alice with all the BTC that was sent to his address (his "banking account") is divided equally between all the double spent transactions to the outputs by the network. Ultimately leaving Alice with 15BTC because Bob tried to scam her.

If Bob tries to pay for a loaf of bread with less than 2BTC in his address, then Alice has a message that pops up so she has to ask Bob to wait for confirmations before walking out of the store.

Using this system we could both promote people to accept Bitcoin and not sacrifice anonymity or force Bitcoiners into bondage. Future improvements could allow for parents to sign their kids addresses so if their kids double spend the penalty BTC is taken from the parents account.

I don't understand how you've been active here long enough to be a "Sr. Member" and yet have so little understanding of anything about bitcoin.  You've recently come up with several ideas and concerns that have been discussed 100s of times and that any "Sr. Member" should understand by now.

[bg002h]

Yeah. Or, you could just insist on having his contact info. Give up anonymity and gain instantaneous transactions. Low tech, but easy.

[madmadmax]

Assuming Bob wants to purchase a loaf of bread physically from Alice, Alice works in a supermarket yet Alice wouldn't use Bitcoins due to the lines getting extremely long because people are waiting for confirmations as well as having to physically keep her eyes on the customers so no one bails with the physical merchandise (or double spends), what solutions could we implement to allow for instant transactions?

My model is the following:
Bob has a total of 30BTC of transactions sent to his address (his "banking account"), Bob then purchases a loaf of bread for 1BTC (an extremely tasty loaf of bread) and walks out of the store, if the network detects that Bob has double spent the same inputs that he used to pay Alice with all the BTC that was sent to his address (his "banking account") is divided equally between all the double spent transactions to the outputs by the network. Ultimately leaving Alice with 15BTC because Bob tried to scam her.

If Bob tries to pay for a loaf of bread with less than 2BTC in his address, then Alice has a message that pops up so she has to ask Bob to wait for confirmations before walking out of the store.

Using this system we could both promote people to accept Bitcoin and not sacrifice anonymity or force Bitcoiners into bondage. Future improvements could allow for parents to sign their kids addresses so if their kids double spend the penalty BTC is taken from the parents account.

I don't understand how you've been active here long enough to be a "Sr. Member" and yet have so little understanding of anything about bitcoin.  You've recently come up with several ideas and concerns that have been discussed 100s of times and that any "Sr. Member" should understand by now.

Please enlighten me as to why the idea is bad?

Yeah. Or, you could just insist on having his contact info. Give up anonymity and gain instantaneous transactions. Low tech, but easy.

Or you could just move him to the corrupt banking system and make it faster

[eb3full]

The idea is bad. Bob creates a ton of double-spends and when the fidelity bond is distributed among them he retains most of what he used to defraud Alice. Also, who exactly is going to 'distribute' this reward? Who decides what double-spends to consider or how many?

A similar idea already exists and was proposed somewhere a month or two ago. Double-spends could be deterred by allowing the merchant to 'replace' the transaction with a transaction that gives miners 100% via fees. It then could not be replaced again, and Bob ultimately loses money (and maybe even reputation) as a result of the double-spend.

[madmadmax]

The idea is bad. Bob creates a ton of double-spends and when the fidelity bond is distributed among them he retains most of what he used to defraud Alice. Also, who exactly is going to 'distribute' this reward? Who decides what double-spends to consider or how many?

A similar idea already exists and was proposed somewhere a month or two ago. Double-spends could be deterred by allowing the merchant to 'replace' the transaction with a transaction that gives miners 100% via fees. It then could not be replaced again, and Bob ultimately loses money (and maybe even reputation) as a result of the double-spend.

The network as a whole would decide to "distribute" who else? We could distribute between a fixed amount of first transactions, e.g. if there are double spends at 6:24, 6:25, 6:26, 6:27 and 6:28 then it would only be distributed between the first three since the probability of Alice accepting the fourth transaction before the double spend is detected would be very small.

[drawingthesun]

If the transaction has a fee, the chance of a double spend working is much smaller. Very few double spends have ever worked (I think S.Dice dealt with a few)

Also the longer you wait (in seconds) the more miners pick up the transaction with the fee, a second transaction with a even with a larger fee would not be included because the first has already been propagated. This all happens before the first confirmation mind you, once that happens no double spend will work (would take a lot of coordination to double spend a loaf of bread (When waiting for you Bitcoin house purchase please wait for at least 3 confirmations))

But about your idea and why its bad. You're plan is to put some of the customers balance at risk every time they purchase something, and for a third party to mediate this. The alternative is far easier on everyone involved. Use a service like Bitpay, they take on the risk of a double spend (very unlikely, remember bitpay is connected to many more nodes than you are) and can cover the odd double spend with their profits.

[peonminer]

Assuming Bob wants to purchase a loaf of bread physically from Alice, Alice works in a supermarket yet Alice wouldn't use Bitcoins due to the lines getting extremely long because people are waiting for confirmations as well as having to physically keep her eyes on the customers so no one bails with the physical merchandise (or double spends), what solutions could we implement to allow for instant transactions?

My model is the following:
Bob has a total of 30BTC of transactions sent to his address (his "banking account"), Bob then purchases a loaf of bread for 1BTC (an extremely tasty loaf of bread) and walks out of the store, if the network detects that Bob has double spent the same inputs that he used to pay Alice with all the BTC that was sent to his address (his "banking account") is divided equally between all the double spent transactions to the outputs by the network. Ultimately leaving Alice with 15BTC because Bob tried to scam her.

If Bob tries to pay for a loaf of bread with less than 2BTC in his address, then Alice has a message that pops up so she has to ask Bob to wait for confirmations before walking out of the store.

Using this system we could both promote people to accept Bitcoin and not sacrifice anonymity or force Bitcoiners into bondage. Future improvements could allow for parents to sign their kids addresses so if their kids double spend the penalty BTC is taken from the parents account.

I don't understand how you've been active here long enough to be a "Sr. Member" and yet have so little understanding of anything about bitcoin.  You've recently come up with several ideas and concerns that have been discussed 100s of times and that any "Sr. Member" should understand by now.

Cut him some slack.. his name is madmadmax after all XD

[madmadmax]

If the transaction has a fee, the chance of a double spend working is much smaller. Very few double spends have ever worked (I think S.Dice dealt with a few)

Also the longer you wait (in seconds) the more miners pick up the transaction with the fee, a second transaction with a even with a larger fee would not be included because the first has already been propagated. This all happens before the first confirmation mind you, once that happens no double spend will work (would take a lot of coordination to double spend a loaf of bread (When waiting for you Bitcoin house purchase please wait for at least 3 confirmations))

But about your idea and why its bad. You're plan is to put some of the customers balance at risk every time they purchase something, and for a third party to mediate this. The alternative is far easier on everyone involved. Use a service like Bitpay, they take on the risk of a double spend (very unlikely, remember bitpay is connected to many more nodes than you are) and can cover the odd double spend with their profits.

The whole point of decentralization is being independent which is what we're trying to achieve so please plug your bitpay shill messages somewhere else.

As for risking the whole balance "for a third party", it will be decided upon by the whole network collectively, how is it any less secure than having Bitcoins in the first place? 

[DannyHamilton]

Please don't feed the trolls.

[drawingthesun]

The whole point of decentralization is being independent which is what we're trying to achieve so please plug your bitpay shill messages somewhere else.

No need for the agression. There must be a third party in your idea or how else will the network determine the double spend? To the network it will eventually just look like 1 transaction with no guarantee enough nodes saw 2 transactions. So there is no way to automatically flawlessly detect a double spend, a third party is essential to settle disputes. Also your idea with no third party requires code to send coins from addresses without needing the private key which has far too many security implications.

Also if the standard is that people without double the money they are spending in store having to wait 10 minutes, 95% of people will be waiting 10 minutes. Not sure where you are from but most people I know live paycheck to paycheck and a grocery shop after rent and fuel does not leave much.

Also I do not work, have not invested and have never even used bitpay. I have heard how it works and see bitpay as a possible future for point of sale, especially if the merchant does not want to hold Bitcoins at all.

[madmadmax]

The whole point of decentralization is being independent which is what we're trying to achieve so please plug your bitpay shill messages somewhere else.

No need for the agression. There must be a third party in your idea or how else will the network determine the double spend? To the network it will eventually just look like 1 transaction with no guarantee enough nodes saw 2 transactions. So there is no way to automatically flawlessly detect a double spend, a third party is essential to settle disputes. Also your idea with no third party requires code to send coins from addresses without needing the private key which has far too many security implications.

Also if the standard is that people without double the money they are spending in store having to wait 10 minutes, 95% of people will be waiting 10 minutes. Not sure where you are from but most people I know live paycheck to paycheck and a grocery shop after rent and fuel does not leave much.

Also I do not work, have not invested and have never even used bitpay. I have heard how it works and see bitpay as a possible future for point of sale, especially if the merchant does not want to hold Bitcoins at all.

Currently 95% of people will be waiting 10 minutes, but it will open a door for the average Joe from the street to accept BTC as well, ultimately rewarding anyone who holds BTC currently.

Also what makes you think that if we make Bitcoin clients relay double spends that it won't propagate as fast through the network (assuming the double spend is valid), there is no reason to suggest it wouldn't be as fast as the transactions themselves.

[drawingthesun]

Currently 95% of people will be waiting 10 minutes, but it will open a door for the average Joe from the street to accept BTC as well, ultimately rewarding anyone who holds BTC currently.

I would love for people to start using Bitcoin on the high street, my life would be so much better in so many ways. But waiting even 10 seconds for a transaction (EFTPOS, VISA, etc....) is unacceptable. We either use 0 confirmations with at least the minimum fee and let the customer walk straight away or use a layer on top such as Bitpay. (If this takes off eventually VISA and MasterCard may support Bitcoin too)

The added benefit of a layer on top of Bitcoin is that we can pay in Bitcoin and the merchant can choose to have the transaction instantly converted into local fiat or remain as Bitcoin. (This will increase adoption)

If people are forced to wait 10 minutes for Bitcoin transactions ever (on the high street) this will forever destroy the reputation of Bitcoin in the minds of normal non Bitcoin people.

Also what makes you think that if we make Bitcoin clients relay double spends that it won't propagate as fast through the network (assuming the double spend is valid), there is no reason to suggest it wouldn't be as fast as the transactions themselves.

If the first transaction is 0 fee it might take some time to be included into blocks, the double spend with a fee then propagates faster before the first confirmation and eventually >50% of the network will see only 1 transaction. This might not happen all the time but it makes automated detection of double spends difficult. Your idea relies on the Bitcoin network always being able to detect the double spends correctly.

Also how do you move the coins without having the private key? This might be the most difficult part of your idea.

[madmadmax]

Currently 95% of people will be waiting 10 minutes, but it will open a door for the average Joe from the street to accept BTC as well, ultimately rewarding anyone who holds BTC currently.

I would love for people to start using Bitcoin on the high street, my life would be so much better in so many ways. But waiting even 10 seconds for a transaction (EFTPOS, VISA, etc....) is unacceptable. We either use 0 confirmations with at least the minimum fee and let the customer walk straight away or use a layer on top such as Bitpay. (If this takes off eventually VISA and MasterCard may support Bitcoin too)

The added benefit of a layer on top of Bitcoin is that we can pay in Bitcoin and the merchant can choose to have the transaction instantly converted into local fiat or remain as Bitcoin. (This will increase adoption)

If people are forced to wait 10 minutes for Bitcoin transactions ever (on the high street) this will forever destroy the reputation of Bitcoin in the minds of normal non Bitcoin people.

Also what makes you think that if we make Bitcoin clients relay double spends that it won't propagate as fast through the network (assuming the double spend is valid), there is no reason to suggest it wouldn't be as fast as the transactions themselves.

If the first transaction is 0 fee it might take some time to be included into blocks, the double spend with a fee then propagates faster before the first confirmation and eventually >50% of the network will see only 1 transaction. This might not happen all the time but it makes automated detection of double spends difficult. Your idea relies on the Bitcoin network always being able to detect the double spends correctly.

Also how do you move the coins without having the private key? This might be the most difficult part of your idea.

Well you could make Bitcoin clients relay transaction with or without fees, you wouldn't need any private key at all because the 51% would decide who the coins rightfully belong to and include it in a block. We could also give the miners an additional fee for the dispute in order to reward them and make propagation more rapid.

[🏰 TradeFortress 🏰]

Well you could make Bitcoin clients relay transaction with or without fees, you wouldn't need any private key at all because the 51% would decide who the coins rightfully belong to and include it in a block. We could also give the miners an additional fee for the dispute in order to reward them and make propagation more rapid.

Computer code cannot detect which is the "legitimate" double spend or which is the "attacker" double spend. They only know which transaction they themselves heard first.

[madmadmax]

Well you could make Bitcoin clients relay transaction with or without fees, you wouldn't need any private key at all because the 51% would decide who the coins rightfully belong to and include it in a block. We could also give the miners an additional fee for the dispute in order to reward them and make propagation more rapid.

Computer code cannot detect which is the "legitimate" double spend or which is the "attacker" double spend. They only know which transaction they themselves heard first.

That's why I propose to divide all the funds equally between every double spend, surely there haven't been 30 double spends.

[🏰 TradeFortress 🏰]

Well you could make Bitcoin clients relay transaction with or without fees, you wouldn't need any private key at all because the 51% would decide who the coins rightfully belong to and include it in a block. We could also give the miners an additional fee for the dispute in order to reward them and make propagation more rapid.

Computer code cannot detect which is the "legitimate" double spend or which is the "attacker" double spend. They only know which transaction they themselves heard first.

That's why I propose to divide all the funds equally between every double spend, surely there haven't been 30 double spends.

It doesn't cost anything to make a transaction. An attacker can create 1000 double spends.

[madmadmax]

Well you could make Bitcoin clients relay transaction with or without fees, you wouldn't need any private key at all because the 51% would decide who the coins rightfully belong to and include it in a block. We could also give the miners an additional fee for the dispute in order to reward them and make propagation more rapid.

Computer code cannot detect which is the "legitimate" double spend or which is the "attacker" double spend. They only know which transaction they themselves heard first.

That's why I propose to divide all the funds equally between every double spend, surely there haven't been 30 double spends.

It doesn't cost anything to make a transaction. An attacker can create 1000 double spends.

But we'd know that the first ones would propagate rapidly through the network, so we could divide the funds between the first 6 for example, since the probability of the 7th one being the legitimate double spend is low.

[AliceWonder]

The solution is green address.

Actually the solution is to not worry about it, it is really just another form of shoplifting. People who want to steal will find a way, and going through the register line is a bad way to do it because most places you will be caught on camera and eventually caught by law enforcement.

But double spends will be rarely attempted and I suspect the losses will likely be less than what is currently lost in debit/credit card transaction fees and chargebacks.

[madmadmax]

The solution is green address.

Actually the solution is to not worry about it, it is really just another form of shoplifting. People who want to steal will find a way, and going through the register line is a bad way to do it because most places you will be caught on camera and eventually caught by law enforcement.

But double spends will be rarely attempted and I suspect the losses will likely be less than what is currently lost in debit/credit card transaction fees and chargebacks.

Who would wait 10 minutes for confirmation while in line? Or do you propose 0 conf?