3 ms·

Fat finger. Or a background story that would make an awsome movie.

by jansan 3y ago

unsupp0rted 3y ago
How easy is it to fat finger a million dollar transaction like this? Let’s say somebody had Satoshi’s wallet address on their clipboard, tried to copy their own, failed, Satoshi’s was still on the clipboard, then they proceeded with the transaction. Wouldn’t they at least look at the first few and last few chars of the address before hitting submit? Also this transaction seems to have been sent out at 9am GMT. Institutional sender during business hours?
- nullc 3y ago
  > Wouldn’t they at least look at the first few and last few chars of the address before hitting submit? The addresses 'look' like line noise, and usually one looks a lot like another. Should they have compared carefully, especially considering the value? Sure. But errors happen. A good workflow for high value transactions lets you preform the transaction and pass it around for validation before signing it (or after signing it but before sending it)-- but virtually no wallets enable such a workflow. So, sadly, many high value transactions are handled without meaningful review. Other ways it could happen due to error: Perhaps someone intended to send 26.9 BTC to "Genesis Trading" and they tasked the most unfortunate of interns to carry it out. They googled "Genesis wallet address", and the google blurb for that query gives _this address_ (I checked). Or... A party intended to send a test transaction to that address with a tiny amount (there are many such payments). But they are using some awful tool that makes them manually manage the payment, change, and fees amounts and they manage to pay the amount that should have been change to the destination and send the amount they intended to pay as fees. (The transaction has no change) Or, Imagine a backend system is pulling addresses out of a database which has them in the same order they're in the blockchain. Unlike a bitcoin node, this system indexes the genesis block so the very first address in it is this address. In that environment multiple failure modes could result in selecting this address. I don't think this one is that credible because indexing addresses by observation in the chain doesn't make a lot of sense for sending.
  - jansan 3y ago
    > A party intended to send a test transaction to that address with a tiny amount (there are many such payments). But they are using some awful tool that makes them manually manage the payment, change, and fees amounts and they manage to pay the amount that should have been change to the destination and send the amount they intended to pay as fees. (The transaction has no change) Hehe, reminds me of the Fat finger trade when a stock trader wanted to buy 100,000 units, but there was a built in multiplier (factor 1000) in his trading tool that resulted in a 100,000,000 units order worth a whopping $10.8bn. But that error only created a damage of $300,000 for the bank. https://www.fnlondon.com/articles/the-fat-finger-points-to-trouble-for-traders-20070312 https://www.fnlondon.com/articles/the-fat-finger-points-to-t...
- kijin 3y ago
  Perhaps somebody hardcoded Satoshi's wallet address as the default value in a testing script, and it somehow made it into production?
  - Ekaros 3y ago
    Having it as default value makes some sense. But on other hand blacklist of known bad addresses might make lot more sense in actual logic.
- knorker 3y ago
  It's easy to not make a mistake. But bitcoin handles a whopping 7 transactions a second. It's basically impossible to not make a mistake EVER. That's why in all sane money systems we can always just reverse the tr... oh... right.

Fat finger. Or a background story that would make an awsome movie.

by jansan 3y ago

Fat finger. Or a background story that would make an awsome movie.

How easy is it to fat finger a million dollar transaction like this? Let’s say somebody had Satoshi’s wallet address on their clipboard, tried to copy their own, failed, Satoshi’s was still on the clipboard, then they proceeded with the transaction. Wouldn’t they at least look at the first few and last few chars of the address before hitting submit? Also this transaction seems to have been sent out at 9am GMT. Institutional sender during business hours?

nullc 3y ago
> Wouldn’t they at least look at the first few and last few chars of the address before hitting submit? The addresses 'look' like line noise, and usually one looks a lot like another. Should they have compared carefully, especially considering the value? Sure. But errors happen. A good workflow for high value transactions lets you preform the transaction and pass it around for validation before signing it (or after signing it but before sending it)-- but virtually no wallets enable such a workflow. So, sadly, many high value transactions are handled without meaningful review. Other ways it could happen due to error: Perhaps someone intended to send 26.9 BTC to "Genesis Trading" and they tasked the most unfortunate of interns to carry it out. They googled "Genesis wallet address", and the google blurb for that query gives _this address_ (I checked). Or... A party intended to send a test transaction to that address with a tiny amount (there are many such payments). But they are using some awful tool that makes them manually manage the payment, change, and fees amounts and they manage to pay the amount that should have been change to the destination and send the amount they intended to pay as fees. (The transaction has no change) Or, Imagine a backend system is pulling addresses out of a database which has them in the same order they're in the blockchain. Unlike a bitcoin node, this system indexes the genesis block so the very first address in it is this address. In that environment multiple failure modes could result in selecting this address. I don't think this one is that credible because indexing addresses by observation in the chain doesn't make a lot of sense for sending.
- jansan 3y ago
  > A party intended to send a test transaction to that address with a tiny amount (there are many such payments). But they are using some awful tool that makes them manually manage the payment, change, and fees amounts and they manage to pay the amount that should have been change to the destination and send the amount they intended to pay as fees. (The transaction has no change) Hehe, reminds me of the Fat finger trade when a stock trader wanted to buy 100,000 units, but there was a built in multiplier (factor 1000) in his trading tool that resulted in a 100,000,000 units order worth a whopping $10.8bn. But that error only created a damage of $300,000 for the bank. https://www.fnlondon.com/articles/the-fat-finger-points-to-trouble-for-traders-20070312 https://www.fnlondon.com/articles/the-fat-finger-points-to-t...
kijin 3y ago
Perhaps somebody hardcoded Satoshi's wallet address as the default value in a testing script, and it somehow made it into production?
- Ekaros 3y ago
  Having it as default value makes some sense. But on other hand blacklist of known bad addresses might make lot more sense in actual logic.
knorker 3y ago
It's easy to not make a mistake. But bitcoin handles a whopping 7 transactions a second. It's basically impossible to not make a mistake EVER. That's why in all sane money systems we can always just reverse the tr... oh... right.