Topic: Should we promote acceptance of "zero-conf transactions"? (Read 265 times)

the obvious solution is an off-chain one. find an exchange (like bakkt or gemini) who can instantly deliver bitcoins through a custodial account on the back end. this is how the SPEDN app is set up: https://cointelegraph.com/news/flexa-launches-app-where-shoppers-can-spend-crypto-at-15-major-us-retailers

that seems like a more practical and safe solution than trying to gauge how many customers you might gain/lose by risking (or not risking) zero-confirmation bitcoin payments.

i think trusted off-chain intermediaries like that are the direction things are headed until decentralized and instant point-of-sale solutions built on lightning (or other layer 2 protocols) are viable at scale.

I think it all depends on the overall outcome. If with accepting zero-confirmation transactions you lose more to thieves than what you gain from additional customers, then you don't do that. But it's possible that your losses will be negligible compared to additional profit. In economic terms, I would compare accepting zero-conf transactions to installment selling. Yes, it's risky, but it turns out it's worth it.

There are different types of Bitcoin adoption. Those bemoaning the dearth of merchant adoption should know how reluctant people who already hold it are to spend it. And then when you have a world which is largely ignorant about Bitcoin at all it's no surprise it's not exactly a booming area. It's possible it will remain a relative niche even if BTC becomes a fixture worldwide for other reasons like saving. Why would you want spend it if you have depreciating alternatives?

The best choice for Bitcoin payments is to use the Lightning Network, not zero-conf transactions. Because the risk of
zero-conf transactions is quite large, I doubt if there are merchants who agree with zero-conf transactions, with the risk of
double spends. The problem is that there are still many merchants that don't use the Lightning Network. Makes the development
of Bitcoin adoption is not going well until now.

Bitcoin's core developers won't give a shit about this. They have more important things to ponder. Gmaxwell celebrated the moment fees approached the block reward briefly in 2017, more for the sign of future sustainability than rinsing faucet users I presume.

Nothing I've seen about LNs has indicated that it's anything other than an interesting experiment, but most of us are completely unused to watching open source development happening in real time. When we arrived at Bitcoin all the hard work had already been done. LNs may get there eventually.

The chances of this getting abuse is high so it won't work period. This sounds like you wanting the bank to trust you for your words, that you'll send them money you intend withdraw but first they should give you cash since there's lot of queue and you're not having the patience to wait. You might think the bank are at disadvantage since they're interested in satisfying their customers to bring about more customers (bitcoin adoption) they should bypass the procedure and take the words of their customers but that's an impossible request same with waiting zero confirmation to become a thing.

We just have to trust the developers working on improving the bitcoin blockchain as it's just a matter of time before they come out with something that'll do the trick. They introduced lighting network but I think it's too complicated to do the job.

it's probably less of a question of how many people are willing, and more a question of how much damage those few people can do. it boils down to loss prevention, and i am confident that not accepting zero-confirmation transactions is one of the best loss prevention decisions a company can make---especially if crypto payments become more popular.


that might apply to large scale thefts, but i don't think it applies to low value transactions. no retailer is going to the police over a $4 cup of coffee. that's a waste of time and money. it's more important for them to figure out how to prevent the $4 losses.

I presume it would fall under the same remit as a cheque bouncing or cancelling the cheque before it clears. You're obtaining by deception and knowingly buying things with money you don't have. Conventional law isn't going to be very impressed with that. No matter what you're paying with you're exchanging value and intentionally withholding your value.

I doubt any merchant would be willing to risk with zero-confirmation transaction. I mean someone might send a really low fee during the time when the network gets congested. Chances are the transaction will get dropped, or the user will double spend it! You will find some merchants where when you send above certain fee, your order gets processed (the fee depends on the network. If you pay the current average fee, they will process your transaction with zero conf).
I doubt any casino does that. What if someone makes a huge deposit, quickly does an all in x2. If he wins, he waits for the confirmation. If he loses he tries to double spend it. Too risky for a casino.

The advantage of doing this, will be immediately be cancelled when merchants starts losing money or when scammers start exploiting this. We want a flawless payment system, where people get what is paid and where you do not have to be worried about this.

If I have to pay $2 to $6 to transfer any amount of bitcoins to someone in another country, then I will settle for a few hours waiting for that transaction to be successful.

We are boasting about Bitcoin not having the same problems as credit cards, where transactions can be reversed, also called chargebacks. So let's settle for a more reliable and trusted payment option and sacrifice some speed and higher fees for that privilege. 

It will only do more damage to the merchants due to scam and for the bitcoin itself, so zero confirmation transactions should never be encouraged in the decentralized payment system.As other said lightning network can be useful for payment of small amount or better go with the zero transaction fee altcoins for such small amount of transfers.

Err. Good point. Idk, while there are no laws specifically concerning double spending, it's still the case that the buyer attempted to buy a product/service by taking back his payment and without ending up with the seller having the payment. So I guess it's technically theft? I don't know crap about law.

Fuck. Probably the same way how chargebacks work with a joint credit card account?

Even though the halving of the block reward had occurred recently, miners are still getting much more from the block reward than from all the fees combined. Look at some latest blocks, for example:

Block 643661


Block 643660


Block 643659


(Note that sometimes the Fee Reward exceeds 1 BTC, but this happens very rare.)

Besides, all fees will never be small. As it was proposed earlier in this thread by @kryptqnick, zero-conf transactions can be accepted for payments below $50, but people should pay higher fees when purchasing more expensive goods/services, or just transferring big amounts from one place to another, when they want it fast. So, the verifiers(miners) will be still incentivised. In the future, with more halvings, this situation can change, but actually it depends on the USD price of 1 BTC. It's possible that it will be so high that even 0.78125 BTC, the block reward we'll have after 2036, will be enough to incentivise hundreds thousands of miners.


Theoretically, yes, but in reality not many people want to engage in criminal activity. It is possible to buy a counterfeit 100 Dollar bill of very good quality for $40-$60, and use it in places where it can't be detected(there are plenty of them), but not many people are willing to do that.


Oh, c'mon, man! They confiscate your phone, computer, or whatever, and prove the guilt.

this is something i've always been curious about, and that we don't have legal precedent for. are we sure that double spending constitutes criminal fraud or a similar crime? this could make for a fascinating legal case.

the courts haven't established whether an unconfirmed bitcoin payment constitutes money received, in the same way cashing a check or taking possession of cash would. i know i don't personally consider money received until a transaction is confirmed on the blockchain.

on top of that, there is also plausible deniability re who sent the double spend transaction. since multiple people can have access to the same private keys, can it be proven beyond a reasonable doubt that you double spent?


they at least support withdrawals to bech32 addresses now and if you look at their hot wallet, they've started using them internally. indeed, they do not support segwit address generation though.

I don't think it is a welcome solution to a temporary problem that is only happening when there are massive spikes in the mempool. Several initiatives have already been put into place to address this issue that we are facing every time a massive price rise is happening. Though not everyone is into it, or like the ideas that were proposed, still one can't deny the fact that those remain useful if utilized and developed the right way, such as the lightning network. Zero-conf transactions being accepted puts us at a greater risk of countless double spends and merchants not receiving payments, instead of fixing the 'problem.' The intention is good, though, just that the solution proposed is not really effective and will do more harm than good.

Well I think the Lightning Network was built to solve that problem. The high bitcoin transaction fees is quite expected in future considering that the price will increase too much. The Lightning Network is the only possible solution to solve this problem as of now and I think it will.
Though I am not sure if the LN is stable right now, I think as people start using it regularly it will get more and more stable.

I agree that supporting zero confirmation transactions is a good solution. I've never seen a transaction not getting confirmed, and average people probably would not bother finding out how to abuse the system (especially in case of it being a crime and people being actually punished for committing it). Besides, it's not hard for merchants to minimize the risk by putting a limit. I mean, say, zero-conf transactions for payments up to $50 and 1 confirmation for more expensive purchases.
This all might not work, but it's worth some research to determine how often people would abuse it on average.
As for the LN, there are groups of people who'll never use it (and they might be the majority given the popularity of the LN), and I don't think a controversial thing like that could go mainstream.

I definitely don't disagree; but the incentive of people attempting a double-spend would greatly drop(in my opinion) if the transaction is done in person(hence I chose the coffee example). Attempting to double spend to get back $4-5 while risking jail time due to your identity being exposed(assuming the shop has a CCTV camera) isn't simply worth it in my opinion. But idk, there are a lot of a-holes out there so I definitely could be wrong. 🤷‍♂️

True, and it's definitely frustrating. Binance, one of the biggest exchanges, doesn't even have SegWit.

And a few other services, like I've seen ad packages and minor buys, they are accepted and good enough for bookings but delivery only on confirmation. Same as gambling sites, free to use your account balance that updates as soon as txs are detected but you won't be able to withdraw anything until input confirms.

That said, also seen more sites require 1 confirmation because of double spend headaches.

On the other hand, sites requiring what used to be the standard 6 are now lowering to 2 or even 1.

I always accept unconfirmed when from a repeat user I know and trust. Maybe sites can implement a sort of trust setting that increased with time and deposit frequency and volume. The more trusted you are the closer to 0 confirmation or even higher percentages of "unlocked" balance from 0 confirms.

Or people will be pushed to go LN (that somewhat takes care of trust at 2nd layer?).

imagine being able to easily double spend every low value transaction back to yourself. you don't think anyone would take advantage?

if it can be exploited, it will be.


lightning adoption is sadly moving at a glacial pace. the liquidity just isn't there, and won't be for quite a while.

what i'd really like to see is boring stuff---transaction optimization, fee optimization, segwit support, batching. it's not glamorous, but if exchanges would just stop unnecessarily clogging the network, i think it would buy us considerable time re scaling.