Why people afraid so much about the confirmation's duration?

[BlackHatCoiner]

When someone sends X BTC to an address the transaction becomes unconfirmed and gets spread over the world. No one can stop that.

I've just seen many people here discussing that bitcoin is not instant payment for shops. Since no one can stop the transaction and the satoshi per byte has a limit on how low you can set it, why people afraid that miners won't mine their transactions? Memory pool keeps them for 2 whole weeks.

[o_e_l_e_o]

There are a variety of different ways that an unconfirmed transaction can end up not being confirmed at all.

It can be double spent, most commonly via a race attack where the sender of the transaction spends the same inputs in a second transaction with a higher fee, and that second transaction is mined first, rendering the first transaction invalid. If replace-by-fee is enabled, then it is very easy to replace a transaction and essentially "cancel" it. Transactions can end up in stale blocks, either through a malicious attack or just because two blocks were mined in rapid succession, even after they have received a confirmation. Transactions can time out and be dropped from the mempool entirely, although in this case there is nothing stopping the receiving party from rebroadcasting it.

Have a read of this page for more info: https://en.bitcoin.it/wiki/Irreversible_Transactions#Attack_vectors

[hosseinimr93]

There is no guarantee that an unconfirmed transaction will be finally confirmed.

Unconfirmed transaction may be replaced by another one if the attacker tries to double-spend the fund.  
However nodes will reject the second transaction from a same input, there is no guarantee that it will always happen. A miner may include the second transaction if it pays a higher fee.

You shouldn't even trust transactions with 1-2 confirmations let alone unconfirmed transactions.

Assume that someone spends some bitcoins to addresses A and B simultaneously. Some nodes see the transaction to address A first and some nods see the transaction to address B first. A miner includes the transaction A into a block and another miner includes the transaction B into a different block at a same time. Now each of transactions have 1 confirmation and we have two chains.
Once the next block is mined, one of transactions receives the second confirmation and the other one becomes invalid.

Also, RBF (Replaceable by fee) transactions can be abused. RBF transactions can be easily replaced by a new transaction.

Not only RBF transactions can be replaced, but also they can be abused to remove a transaction.
Assume that a fund is sent from address A to address B and then from B to C. The transaction from address A to address B is RBF, the fee is bumped and the transaction is replaced by a new transaction. The transaction from address B to address C will disappear.

[Mbitr]

I’m not sure if you’ve used BTC as a payment OP , but I use BTC fairly regularly for buying and selling etc . No matter what you buy/ sell etc ,personally  there is always a bit of stress envolved until you get that first confirmation! I know it’s going to confirm , but when it doesn’t on the odd occasion - after 10 mins or 1 hour or (please god no ) 12 hours  - it is stressful .
I believe fully in BTC and wouldn’t want anything changed about the confirmation time , it’s just takes a while to get used to  

[coupable]

I don't like to call it an issue, but until getting the first confirmation, we can't deny that this is a little bit stressful. Everything is ok with bitcoin and how it's designed, but we all remember what happened in 2017 when a simple transaction of a small amount costs in fees more than the value of the transaction itself and some transactions had to wait more than a week to get confirmed. Bitcoin cash and segwit implementation were some of the solutions made by the bitcoin community to overpass the block issue which results in transactions not being confirmed in a reasonable time.
I totally agree that users still get frustrated when the receive a transaction with very low fee. I sell btc in local and really fed up of people asking me to set a high fee without noting that am using segwit for less fees and that a small fee transaction can get confirmed using RBF or CPFP .

[20kevin20]

Besides the technical stuff o_e_l_e_o has said above, I'll get in this thread with my own, personal concerns:

• When I have to move funds fast to an exchange, I never really know how much it's going to take exactly until the tx is confirmed. Sometimes had to wait +10hrs for a tx that I wanted to be confirmed within dozens of minutes at last.
• Trying to make a face-to-face tx and having to wait hours until it's confirmed. Most of the times, the other party will not have enough time to sit next to you and wait until it's confirmed. If they have an Electrum somewhere with a "replace by fee" enabled on it, you could be screwed.

The list must've been longer, but at the moment I can't think of more realistic scenarios you could encounter. The number of times I've had the issue with exchange deposits is larger than I wish it'd be.

[yazher]

I think one of the reasons is they are paranoid whether the transaction will confirm or not, there are some scenarios where the transaction is failed, especially when the network is congested. sometimes it will took a day to confirmed a single transaction when you used the lowest fee available. that's the reason why lots of people will only feel relieved when they see the green icon of confirmed and most of the time these people are taking care of their time there are lots of things to do you know, work and other stuff when you still have in your mind about your unconfirmed information it really annoys you, it just really hassles you can't do your work properly. Let's just hope with their ongoing development they will find some necessary updates to fixed this problem.

[Stedsm]

Not adding reasonable fee to our transactions is what we do wrong sometimes because then, expecting our transaction to get in immediately in the next block is worthless as miners adjust fee-structure according to the slabs they decide based on the minimum they want to mine. Nobody is afraid here because there are many websites as well as the wallets also recommend the fee based on the least-to-max fee in last transaction that got accepted in a block.

Visit https://jochen-hoenicke.de/queue/#0,2h
Scroll down and see "Mempool Size in MB"
Hover on the image and check which slab has its size less than 1 MB, you can freely use that slab in order to get your transaction confirmed.

[stompix]

Since no one can stop the transaction and the satoshi per byte has a limit on how low you can set it, why people afraid that miners won't mine their transactions?

Because as o_e_l_e_o it's not true and because when you try to purchase something, you usually want your tx to get confirmed, not dropped  , beyond those technicalities there is also the practical side.

It's not the case now anymore, but imagine paying for a thing with 2sat/b and then...waiting not for the shop with the order, not for FedEx, but for the fee to confirm. What if it's a monthly subscription and it gets canceled before your confirmation? When you send coins to an exchange you send them because you want to sell them now, at this price, not in two years' time when the exchange is already hacked, seized, and taken offline.
You find a great deal, a promo for24h and you send your order with a low fee to make sure you miss the promotion? 

Yeah, in some cases it doesn't matter: when you send coins to your cold storage, if you transfer between wallets, if the guy you're sending them is patients, but imagine paying back a debt and the guy who gave you the loan now has to wait an extra week, that's not the best choice where to cut expenses.

[LTU_btc]

I can't say that I'm afraid, but yeah, after I send BTC to someone, I always check it's status on blockchain and I reload website multiple times until I see that it got confirmations. As said by users above, there is multiple things when something just will get wrong and transaction will not confirmations after all. Only when I see that it got confirmation I can feel 100% relaxed.
But it's just about Bitcoin. I have similar feeling when make bank transfer. When I make it, I usually have doubts, did I made everything as I had to do. And only when I know that my recipient got money, I can keep calm.

[Little Mouse]

Because with zero confirmation tx, it can be double spent. For example, if I send you $100 BTC for buying ETH and with zero confirmation, you accepted the payment and sent me ETH. Since BTC is in still zero confirmation, I can now send the same input to another address. You will be scammed. Usually, this does not happen in a normal time. When the fee increase, chances of double spent increases too. I have seen some guy to be scammed last time too.

[ice18]

If someone is new to digital currency and no idea how blockchain works and try to buy btc for the very first time of his life and send to his btc wallet and happens that after 3 days transaction is still in receiving status because of very low fees input the initial reaction will be of course in a panic mode its a normal reaction from new users of bitcoin especially if we are talking a large sum of btc.

[hd49728]

People worry about that because depends on each platform, their deposit transaction will be required from 1 to 3 confirmations to be credited to their accounts' balance. Without credited amount in deposits, they will not be able to use their funds to trade. People tend to feel in hurry after they make deposits and they often have demands for as fastest confirmed transactions as possible.

For personal trade deal, depends on total value of deals, people can make personal confirmations from 1 to 3 or even bigger than 3 confirmations. If they make trades with their trade partners for a total values of $millions, they should expand the number of confirmations bigger than 3.

[pooya87]

Because with zero confirmation tx, it can be double spent.

this has nothing to do with "duration" which is what OP asked. unconfirmed transactions are always unsafe on all PoW-based cryptocurrencies. the same thing you explained could happen in reverse meaning you could send someone BTC and it confirms while the ETH transaction remains unconfirmed and is double spent. in fact it is a lot easier to double spend ETH transactions since nobody runs a full node. all the attacker has to do is to intercept your communication with the centralized node you depend on and reply with fake transaction state.

[Pmalek]

The uncertainty is what makes people worry. We have gotten used to instant payment methods. We know once we swipe that credit card, that's it. With Bitcoin there is always that what if it doesn't confirm? What if I didn't use a high enough fee? What if that bonus offer expires before my transaction confirms? What if the other party cancels our deal because it takes too long?

Especially for users who are new to cryptocurrencies there are plenty of uncertainties here. Using the RBF feature should be standard practice nowadays.

[o_e_l_e_o]

there are some scenarios where the transaction is failed, especially when the network is congested.

The transaction does not fail just because the mempool is congested. Provided it has been signed correctly and broadcast above the minimum relay fee, the worst that can happen is that it is delayed. Even if it disappears from the mempool, the sender, recipient, or anyone else for that matter, can choose to rebroadcast it at any time. The only time it would fail is if a competing double spend transaction was mined first.

They're too used with convenience offered by digital fiat wallet (e.g. instant confirmation)

We know once we swipe that credit card, that's it.

I think this is a mischaracterization of what is happening. I agree the people think that it is an instant confirmation, but it isn't. The credit card transaction shows up immediately, just like a bitcoin transaction is broadcast immediately, but it takes weeks or even months for the credit card transaction to "confirm" in the same way a bitcoin transaction would after ~10 minutes. It takes several days for the money to actually be transferred to the recipient, and it is significantly easier to cancel a credit card transaction than it is a bitcoin transaction. I can phone up my credit card company up to 120 days (or sometimes even more) after a transaction and say my card was lost, stolen, cloned, duplicated, hacked, I was overcharged, didn't buy the product, etc., and cancel a transaction.

[bob123]

The problem is that the possibility of a double spending attack exists.

Either a simple one by replacing the transaction which has the RBF-flag set or by more complicated ones which need some technical knowledge or the cooperation with a miner.
For small purchases (like a coffee) it doesn't really matter. If you don't accept RBF transactions, you are free to accept 0-conf transactions.

However, for larger purchases (or medium purchases where the buyer is kind of anonymous) you actually should wait for at least one confirmation.

[BIT-BENDER]

I’m not sure if you’ve used BTC as a payment OP , but I use BTC fairly regularly for buying and selling etc . No matter what you buy/ sell etc ,personally  there is always a bit of stress envolved until you get that first confirmation! I know it’s going to confirm , but when it doesn’t on the odd occasion - after 10 mins or 1 hour or (please god no ) 12 hours  - it is stressful .
I believe fully in BTC and wouldn’t want anything changed about the confirmation time , it’s just takes a while to get used to  

Payment, transaction, purchase that you have the luxury of time then I don't mind when the confirmation time happens, but for something urgent and time is not friendly then a faster confirmation would be fulfilling, I don't know if the time for confirmation can be engineered to become faster or this is something we just have to deal with.

[McArthurRex]

When someone sends X BTC to an address the transaction becomes unconfirmed and gets spread over the world. No one can stop that.

I've just seen many people here discussing that bitcoin is not instant payment for shops. Since no one can stop the transaction and the satoshi per byte has a limit on how low you can set it, why people afraid that miners won't mine their transactions? Memory pool keeps them for 2 whole weeks.

It depends on the situation IMO. Some people need to transfer their BTC urgently and slow transaction may cause opportunity loss for them. Other than that, some people don't know how POW fully works misleading them to assume that their BTC might never be confirmed.

[o_e_l_e_o]

I don't know if the time for confirmation can be engineered to become faster or this is something we just have to deal with.

Not at the base layer without changing the target block time of 10 minutes, which is highly unlikely to happen. But this is exactly why Lightning Network (and potentially other second layer solutions) are being developed.

I spend bitcoin frequently online, and usually just pay with a normal transaction. If I'm waiting a few days for something to be shipped, then an extra 10-20 minutes for a confirmation is irrelevant. Paying in person is a different story though - nobody has time to stand around for 10-20 minutes after paying for a coffee or something small waiting for a transaction to confirm. The options available to the merchant are either to accept zero confirmation transactions, which is probably an acceptable risk to take for small value transactions such as a coffee, particularly if you are a frequent customer and build up some trust/rapport with the vendor or merchant, or to use the Lightning Network for instant confirmation.