Comment by wruza | Hacker News Reader

Comment by wruza | Hacker News Reader

Is this a motivation issue with crypto? Here is what I mean. Centralized systems work like a web of tunnels - you can’t send “nowhere” or make a huge mistake, but if your part of a tunnel gets shut for some reason, you’re locked. Decentralized ones are like an endless empty void with point-like islands of ownership in it. You can send anywhere, but make one mistake and it’s over. Now pepper this with our usual issues with code correctness and debugging.

The issue is that people do like tunnels because they feel safe, and empty voids are scary. This incentives to still build and use tunnels around this void, which makes “crypto”-ness just an implementation detail not worth using as is by the masses.

Afaiu, this happens for two reasons: 1, the usual, people tend to forget assertions. 2, the new one, crypto institutions who make contracts are incentivized to omit assertions because runtime now costs serious money. They are like factories where there is no rule of safety yet.

Is the above correct, or is it a true half a million dollars mistake? You can’t trust node_modules as it seems, how can you trust piles of contracts code when there is no support, no takedowns of bad versions and no easily accessible (and decentralized?) audit?

Also, could a type system help in this case?

>Centralized systems work like a web of tunnels - you can’t send “nowhere” or make a huge mistake

You can totally make a 1 digit typo while making a bank transfer that results in all the money being lost.

Perhaps you’ll recover it after years of litigation, but that’s really not guaranteed.

https://www.theguardian.com/money/2019/dec/07/i-lost-my-1930...

E: downvoted by upset bankers? :)

But at least in that case, it's possible to get the money back. In this case, isn't the money absolutely, totally, 100% gone forever?

Sort of? You could appeal to the Ethereum developers, they could reasonably recover these funds without upsetting anyone too much by just refunding all WETH transactions sent to that contract. (This is the root of the spiritual disagreement between Ethereum and Ethereum Classic)

Of course you’re more likely to succeed when it’s real money and the other party has to worry about getting sued.

But even in the case of a bank transfer, the other party can abscond with your money and it’ll be extremely hard to recover since law enforcement probably won’t help you and (especially international) litigation costs can quickly render such an endeavor far too expensive.

Americans do not understand money transfer, they are still using cheques and western union.

To everyone: If you wrongly send money to a mistyped IBAN it is gone.

But that's very unlikely because IBAN has a checksum precisely to prevent typos.

Plus banks in Europe have a transfer reversal grave period so you just call them and they'll most likely be able to undo the transaction.

I'm not sure about IBAN, but in the UK there are multiple layers that help you avoid sending to the wrong account. However, if you do ultimately send money into an actual incorrect person's account, the only thing the recipient's bank will do is to ask them to send it back. They will not give you contact details or anything else.

TIL - more on IBAN checksum:

> The IBAN check digit consists of two digits in positions 3 and 4 of the IBAN. It is calculated using the MOD97 algorithm and provides the primary integrity check for the IBAN standard.

https://www.iban.com/iban-checker

To everyone: the above is wrong.

> mistyped

Very unlikely, the checksum would prevent that.

And most banks (even though I’m unsure it is mandatory) will check the recipient, if totally mismatched a warning will be raised.

But if you send money to the wrong iban+recipient, the money is gone.

It is not gone. You may have to go through courts if the recipient does not return the money voluntarily though.

You will not have any luck enforcing that court judgment if the recipient does not want to play ball. All they need to do is transfer the money to another country.

It is very unlikely that law enforcement would help you in a situation like this, especially internationally.

I made a transfer with a mistyped IBAN once in the online interface of my German bank. The bank called me a day later and said that there’s something wrong with my IBAN, so I double-checked, realised my mistake and did the same transfer to the correct IBAN. I did not loose any money.

every online interface would check the IBAN checksum, this seems like a bad bank or a bad story aka lie.

There’s SWIFT, and they have a reasonably good dispute mechanism, at least for cross border payments.

> To everyone: If you wrongly send money to a mistyped IBAN it is gone.

You clearly have no idea how an IBAN works. It has a checksum exactly to avoid mistyping.

And even if you send the money to an valid, but wrong, IBAN then one of two things happens:

* The account doesn't exist -> transaction bounces * The account does exist -> ask for money back, worst case go to court to recover it

But the money is not gone. There are no black holes just swallowing everything up.

Bad example. IBAN (https://en.wikipedia.org/wiki/International_Bank_Account_Num... ) for example has a two digit checksum that prevents this and similar simple typos.

Besides the checksum, in some countries (e.g. in many banks in The Netherlands), the given name is also checked against the name associated with the target account [1].

[1] https://www.betaalvereniging.nl/betaalproducten-en-diensten/...

How is it a bad example? Not everyone uses IBANs.

And besides, big banks still have to deal with checksum-passing iban typos on a daily basis. IBANs only have two check digits.

> big banks still have to deal with checksum-passing iban typos on a daily basis

Citation needed.

This is elementary school level math, no citation needed. The modulo 97 checksum offers known guarantees.

I think there's a misunderstanding here. If the customer mistypes the IBAN account number, the bank (website, app etc) will automatically reject it as invalid immediately, he won't have to deal with the customer service.

Yes, but the misunderstanding is on your side.

This is overly simplified so not entirely accurate, but if 100 customers typo their destination IBAN the modulo 97 checksum will probably let 3 of them through.

Wikipedia says that "where used, IBANs have reduced trans-national money transfer errors to under 0.1% of total payments", although there's no source for the claim.

0.1% of total payments, i.e 1 in 1000 payments. That’s not 0.1% of typos but all payments.

> And besides, big banks still have to deal with checksum-passing iban typos on a daily basis. IBANs only have two check digits.

Actually the checksum was designed especially to deal with typos. As such, dealing with typo errors is usually not an issue at all.

What they do have to deal with is maliciously created IBANs though. However, if the account an IBAN should point at doesn't exist then the transaction usually just bounces.

> And besides, big banks still have to deal with checksum-passing iban typos on a daily basis.

That's the thing - they can deal with it. There are fallbacks.

In crypto, your money is just instantly destroyed.

You’re mistaken, they usually can’t deal with it.

They’ll ask the other bank if they’d like to return the money, and that bank will maybe ask the recipient if they’d like to return the money.

The recipient doesn’t want to return the money? You’re SOL. You can go to court, but they can trivially evade civil action by transferring the money overseas.

Those are all things you can try, though. In crypto, again, the money just instantly disappears.

You could still appeal to the developers, it has been done successfully in the past.

Once.

Which indeed proves that it can be done.

Ok, you are definitely not arguing this is good faith any longer. Get lost.

What do you mean? If you can convince a sufficiently large share of the community, you can in fact roll back transactions.

This is a massively disingenuous argument, and you know it.

Ethereum addresses also have a checksum. In both systems it's possible to enter the wrong destination address.

I have more than once typed ref-number into the amount field. Ofc a transmission of $9765775689758 won't succeed, but if it did, I'm told there's little to be done about it, finders keepers.

Yeah, if you do that and the other party runs off with the money you’re screwed.

You can certainly pursue the matter in civil courts, but law enforcement does not like to touch this kind of stuff at all. If the recipient transfers the money overseas, you will most likely never be able to recover it.

I don't think that's the case everywhere. In Switzerland at least, if I send you money by mistake, I can claim that money back and you would have to give it back to me, even if you have spent it already - unless you can reasonably claim that you were unaware that this money was not meant for you.

As a result, if I received a large amount of money that I coulnd't explain, I would contact the sender or the bank first. I can't just spend it immeditaly and claim that I had no idea.

Yeah, good luck getting that money back if the other party doesn’t want to play along. Enforcing court judgements isn’t easy in these situations.

Where would it be non-reversible? At least in Europe you totally do not own money that is transferred to your bank account by mistake. You'll have to pay it all back. There have been court cases after people spent such windfalls, and yes, they are still in the hook for full returns.

If the recipient of the windfall is not honest, they can just move the money to another country.

You can go to court and get a judgement against someone with no assets to collect. In the end you’ll just have lost even more money.

If you read the followup, that man ended up getting his money back. Additionally, it would be astronomically unlikely to make a mistake like that nowadays because confirmation of payee is used so widely (banks check that the name of your recipient matches the intended name before completing a transaction).

[0] https://www.psr.org.uk/publications/consultations/cp19-4-con...

But this costs us much much more than 500k in sum

This is a really good case against crypto.

In the crypto case of this thread, the OP made a mistake and there's no process to fix it. 100K's lost. There's no avenue for recovery.

In the banking case of your link, the man made a mistake and was able to correct it, despite the bank working against him. 100K's lost and recovered.

In the case of the bank, it sure wasn't perfect and definitely should be better. But ultimately, the bank not only paid to recover the money, but the banking system (according to the article) is at least promising a fix for the underlying issue that a simple mistake in writing digits can lead to missending funds.

Also: crypto is touts its trustless nature as a positive. But note that trust is the lever Mr. Teich used to get Barclays to pay the cost of recovering the funds. Barclays only paid after their reputation was threatened, and of course, trust is based on reputation.

> You can totally make a 1 digit typo while making a bank transfer that results in all the money being lost.

Difference being that it isn't "lost" -- not in the sense of vanished from the face of the Earth, gone, ceased to exist. It's only lost to the sender, but it still exists; someone still has it. Sure, the wrong person or entity, but still: It's not gone from existence. Only CryptoCrap does that.

What would really help is if the recipient had to prove they can spend the money they're about to receive before they receive it.

Which is in fact required with cryptocurrencies using the (pure) Mimblewimble protocol.

Reply on news.ycombinator.com