Oops - 20BTC fee paid on .05 transaction?

[DeathAndTaxes]

Well grandma wouldn't try to be stupid enough to manually create a tx by hand.  She would use a wallet either a wallet program or an eWallet and neither make these kind of mistakes because they are written by people who actually know what they are doing.

If you want to create tx by hand you are working at the functional layer, like programming in machine language, or manually constructing binary TCP/IP packets.  If you program in machine machine language you can cause all kinds of problems too.  Based on that your conclusion would likely be that grandma can never use facebook because machine language is dangerous.

[Bugpowder]

Some people are just not up to the challenge of handling their money responsibly.  

Funny how we are split between sympathy and .... whatever the opposite of sympathy is.

People need to learn how to treat their bitcoins with respect. Clicking is a lot easier than handling real cash. Its not like anyone ever went into a restaurant, paid for a $10 meal, and left a $3000 tip.

I believe the word you are looking for is schadenfreude.

It sucks for him, but the reality is you need to be educated, responsible and careful with your coins.  Otherwise you will eventually be separated from them.  Its a shame for him to lose his whole stack, but given the amount of care used here it was probably only a matter of time before they would have been lost to an investment scam or gambled away.  Different recipients same result.

A excessive fee warning in the client would be nice too.

[SgtSpike]

It sounds like he created the transaction using brainwallet.org from a brain wallet address to the payee address without a change address. So there is 20.199BTC input and 0.05271705 BTC output.

https://en.bitcoin.it/wiki/Transactions#Output

Any input bitcoins not redeemed in an output is considered a transaction fee; whoever generates the block will get it.

This website should be changed to default to sending the change back to the original address.  There is no reason to default to sending the change as a tx fee.

[kireinaha]

The good news is, his transaction must have been confirmed by the network really fast! The bad news is, bitcoin is still a long ways from becoming an acceptable currency for the mainstream, given that horrific errors like this can occur.

This posts sums it up nicely:

Can't wait to see grandma start using bitcoins. Oops, there goes 500 grand.

Yes, let's blame the entire Bitcoin infrastructure on one person's stupid mistake with raw transactions.

People have been losing their lives, daily, to automobile accidents since the automobile was invented. Not money, their lives!

Yet, we don't go around saying "automobiles are not an acceptable means of transportation for the mainstream, given that horrific errors like this can occur," do we?

"Can't wait to see grandma start driving a car. Oops, there goes grandma's life along with the family of 5 she crashed into."

Humans are going to make huge, sometimes life threatening, errors regardless of the tool you give them. It's up to the individual to educate and protect themselves. Even so, accidents will occur.

I've never heard of anyone making such a fee mistake unless they were messing with raw transactions to being with.

It's simple to call it an idiot mistake from atop your ivory tower, but if a bitcoin "enthusiast" (we are still in the early adopter phase, afterall) can manage a colossal fuck up on the scale of blowing $20,000 USD worth of bitcoin, then there is a problem with the bitcoin ecosystem. There is no equivalent to this in real life. You would never accidentally leave a $20k tip at a restaurant, or accidentally overpay $20k for your Starbucks frappacino.

Your car analogy is flawed. This is why: we have been using cash and credit cards for years, each have provided great utility to consumers. Like I said above, there is essentially no room for massive fuck ups, and even if your money is stolen somehow, you typically have recourse.

It would be like if supercharged motorbikes from the present were suddenly introduced to denizens of the world of a distant future Earth, where all cars traveled at 200 mph and were automated so that accidents were a thing of the distant past.

Now you give them these bikes which can be crashed, and the people, without adequate knowledge or proper concern for their personal safety and that of others are cut loose. Of course they'll have accidents, and for what point, ultimately?

If you guys want bitcoin to succeed, it will need to be idiot proofed. Nobody is going to convert their cash to a protocol that allows them and their friends to potentially lose their life savings to some momentary lapse of judgement!

[empoweoqwj]

It sounds like he created the transaction using brainwallet.org from a brain wallet address to the payee address without a change address. So there is 20.199BTC input and 0.05271705 BTC output.

https://en.bitcoin.it/wiki/Transactions#Output

Any input bitcoins not redeemed in an output is considered a transaction fee; whoever generates the block will get it.

this is infact what happened

[–]aliens_exist_1 11 points 3 hours ago
Nope, I was using brainwallet, and I could've sworn that in the box next to source address I had the my total balance, and in the box next to destination I had the ~0.05 I intended to send

[–]ummhaha 37 points 4 hours ago
I'm sure you did, because that's exactly how the transaction was structured.
When you spend from a brainwallet, you must spend ALL your funds. The difference goes into the fee.

talk about an expensive lesson 

Sums it up. He couldn't be arsed to read what was on his screen.

Hence the need for "protection from ourselves and our grandmas". Several confirmations, even complete rejection of transactions where the transaction fee is disproportionately large. Makes sense.

[BitThink]

It has nothing to do with the bitcoin protocol, but the problem of the client software. Any reasonable client should at least warns the user about this error, or better forbids the tx fee to be higher than a certain value.

The brainwallet.org transactions are not supposed to be used by ordinary users at all, and they should put a big warning there. It's used to generate something like 'raw transactions', not used for daily BTC transactions.

[empoweoqwj]

The good news is, his transaction must have been confirmed by the network really fast! The bad news is, bitcoin is still a long ways from becoming an acceptable currency for the mainstream, given that horrific errors like this can occur.

This posts sums it up nicely:

Can't wait to see grandma start using bitcoins. Oops, there goes 500 grand.

Yes, let's blame the entire Bitcoin infrastructure on one person's stupid mistake with raw transactions.

People have been losing their lives, daily, to automobile accidents since the automobile was invented. Not money, their lives!

Yet, we don't go around saying "automobiles are not an acceptable means of transportation for the mainstream, given that horrific errors like this can occur," do we?

"Can't wait to see grandma start driving a car. Oops, there goes grandma's life along with the family of 5 she crashed into."

Humans are going to make huge, sometimes life threatening, errors regardless of the tool you give them. It's up to the individual to educate and protect themselves. Even so, accidents will occur.

I've never heard of anyone making such a fee mistake unless they were messing with raw transactions to being with.

It's simple to call it an idiot mistake from atop your ivory tower, but if a bitcoin "enthusiast" (we are still in the early adopter phase, afterall) can manage a colossal fuck up on the scale of blowing $20,000 USD worth of bitcoin, then there is a problem with the bitcoin ecosystem. There is no equivalent to this in real life. You would never accidentally leave a $20k tip at a restaurant, or accidentally overpay $20k for your Starbucks frappacino.

Your car analogy is flawed. This is why: we have been using cash and credit cards for years, each have provided great utility to consumers. Like I said above, there is essentially no room for massive fuck ups, and even if your money is stolen somehow, you typically have recourse.

It would be like if supercharged motorbikes from the present were suddenly introduced to denizens of the world of a distant future Earth, where all cars traveled at 200 mph and were automated so that accidents were a thing of the distant past.

Now you give them these bikes which can be crashed, and the people, without adequate knowledge or proper concern for their personal safety and that of others are cut loose. Of course they'll have accidents, and for what point, ultimately?

If you guys want bitcoin to succeed, it will need to be idiot proofed. Nobody is going to convert their cash to a protocol that allows them and their friends to potentially lose their life savings to some momentary lapse of judgement!

Yeah I love credit cards as a merchant. People buy stuff off you then charge back. Nearly put me out of business two times. You aren't getting any sympathy from me with that angle.

Sure, the software needs to get better. But the guy has to take some responsibility for his own actions. He didn't have a gun to his head as far as I know.

I hope the pool pays the transaction fees back though. Nothing like the world thinking of bitcoiners as a bunch of immoral scammers to kill the concept.

[StarfishPrime]

It sounds like he created the transaction using brainwallet.org from a brain wallet address to the payee address without a change address. So there is 20.199BTC input and 0.05271705 BTC output.

https://en.bitcoin.it/wiki/Transactions#Output

Any input bitcoins not redeemed in an output is considered a transaction fee; whoever generates the block will get it.

This website should be changed to default to sending the change back to the original address.  There is no reason to default to sending the change as a tx fee.

Yes, that default behavior is just ridiculous. Wait until more people start redeeming paper- and brain wallets. Expect to see a lot more stories like this.

This is not good PR for bitcoin at all - Some type of 'maximum fee' principle by nodes might be more reasonable, with obviously erroneous fees returned to the input instead?

[Abdussamad]

This website should be changed to default to sending the change back to the original address.  There is no reason to default to sending the change as a tx fee.

It does send change back to the source address.

Actually what appears to have happened is that the user got caught by a UI bug. Assuming he used this page - http://brainwallet.org/#tx - if you change the figure next to the destination address the json transaction updates and the appropriate change is sent back to the source address. But if you change the source amount the transaction does not update meaning you end up paying more as fees.

[BitThink]

This website should be changed to default to sending the change back to the original address.  There is no reason to default to sending the change as a tx fee.

It does send change back to the source address.

Actually what appears to have happened is that the user got caught by a UI bug. Assuming he used this page - http://brainwallet.org/#tx - if you change the figure next to the destination address the json transaction updates and the appropriate change is sent back to the source address. But if you change the source amount the transaction does not update meaning you end up paying more as fees.

So it's the bug of brainwallet.org, and nothing to do with the bitcoin.

[empoweoqwj]

This website should be changed to default to sending the change back to the original address.  There is no reason to default to sending the change as a tx fee.

It does send change back to the source address.

Actually what appears to have happened is that the user got caught by a UI bug. Assuming he used this page - http://brainwallet.org/#tx - if you change the figure next to the destination address the json transaction updates and the appropriate change is sent back to the source address. But if you change the source amount the transaction does not update meaning you end up paying more as fees.

In other words, the software is crap, and should be called out as such. We don't need such amateur software ruining the reputation of bitcoin

[StarfishPrime]

I have to say it again. This has nothing to do with bitcoin, and it's just the issue of brainwallet.org. It should make it very clear that the transaction page is just used to manually create a raw transaction that move all BTC from one offline address to another offline address. It should not be used for other usages at all.

That's correct, but it's not unthinkable that sometime in the future, nodes could invalidate obviously erroneous transactions (like 20btc fees or unspendable outputs for example).

Extensive error controls exist in all other financial transaction systems. No properly architected commercial system would ever pass through things like this.

[BitThink]

I have to say it again. This has nothing to do with bitcoin, and it's just the issue of brainwallet.org. It should make it very clear that the transaction page is just used to manually create a raw transaction that move all BTC from one offline address to another offline address. It should not be used for other usages at all.

That's correct, but it's not unthinkable that sometime in the future, nodes could invalidate obviously erroneous transactions (like 20btc fees or unspendable outputs for example).

Extensive error controls exist in all other financial transaction systems. No properly architected commercial system would ever pass through things like this.

20 BTC fees maybe, but unspendable outputs no. How can the mining nodes know which output is unspendable?

Even for the 20 BTC fee, I still think the restriction should be done in the client side. The mining node has no reason to forbid people to pay 20BTC fees, and maybe the user just want to ensure the transaction to be included in a block as soon as possible.

It's like if we leave $1000 in the table of a restaurant, any reasonable waiter will warn us, but he will accept as the tip it if you insist. There's no reason to establish a law to forbid any tip larger than $100.

[empoweoqwj]

I have to say it again. This has nothing to do with bitcoin, and it's just the issue of brainwallet.org. It should make it very clear that the transaction page is just used to manually create a raw transaction that move all BTC from one offline address to another offline address. It should not be used for other usages at all.

That's correct, but it's not unthinkable that sometime in the future, nodes could invalidate obviously erroneous transactions (like 20btc fees or unspendable outputs for example).

Extensive error controls exist in all other financial transaction systems. No properly architected commercial system would ever pass through things like this.

20 BTC fees maybe, but unspendable outputs no. How can the mining nodes know which output is unspendable?

Even for the 20 BTC fee, I still think the restriction should be done in the client side. The mining node has no reason to forbid people to pay 20BTC fees, and maybe the user just want to ensure the transaction to be included in a block as soon as possible.

It's like if we leave $1000 in the table of a restaurant, any reasonable waiter will warn us, but he will accept as the tip it if you insist. There's no reason to establish a law to forbid any tip larger than $100.

Sure we don't need a law. but if someone mistakenly tipped you $1000 at a restaurant, would you give it back? Maybe not ........ but then a restaurant isn't risking the whole reputation of what we hope is a major world currency.

[BitThink]

I have to say it again. This has nothing to do with bitcoin, and it's just the issue of brainwallet.org. It should make it very clear that the transaction page is just used to manually create a raw transaction that move all BTC from one offline address to another offline address. It should not be used for other usages at all.

That's correct, but it's not unthinkable that sometime in the future, nodes could invalidate obviously erroneous transactions (like 20btc fees or unspendable outputs for example).

Extensive error controls exist in all other financial transaction systems. No properly architected commercial system would ever pass through things like this.

20 BTC fees maybe, but unspendable outputs no. How can the mining nodes know which output is unspendable?

Even for the 20 BTC fee, I still think the restriction should be done in the client side. The mining node has no reason to forbid people to pay 20BTC fees, and maybe the user just want to ensure the transaction to be included in a block as soon as possible.

It's like if we leave $1000 in the table of a restaurant, any reasonable waiter will warn us, but he will accept as the tip it if you insist. There's no reason to establish a law to forbid any tip larger than $100.

Sure we don't need a law. but if someone mistakenly tipped you $1000 at a restaurant, would you give it back? Maybe not ........ but then a restaurant isn't risking the whole reputation of what we hope is a major world currency.

Actually last time Asicminer did return the 200 BTC tx fee. I think most reputable miners will return the fee, but this time unfortunately the block is mined by the P2PPool and this makes things much more complicated.

[kjj]

These websites really should not be exposing the technical innards of bitcoin to random people.  Also, "brainwallets" should die in a fire.

And as a lesson to everyone watching, always run your raw transactions through a decoder before you sign it.  Hopefully one smart enough to look up the inputs and warn you if you are about to make a huge mistake.

[cczarek123]

?
Only 20?
i this its a mistake

[nate008]

This will give some Bitcoin haters enough material to start ranting about how imperfect the system this.
But hopefully it will also educate some people on how to use it , and to double check what they do if unsure of it.

[satosh007]

This will give some Bitcoin haters enough material to start ranting about how imperfect the system this.
But hopefully it will also educate some people on how to use it , and to double check what they do if unsure of it.

It's not the system but his choice of wallet.
Ido feel sorry for him and his lost.
20 coins to may 100s be on their way to him.
and to me too  

[bryant.coleman]

When you spend from a brainwallet, you must spend ALL your funds. The difference goes into the fee.

Too bad for him.... 20 BTC is 18,000 USD at today's rate.