6 ms·

Don't we track transaction security by chain length as well as new block generation speed? Those should let you determine the probability of your transaction be

Don't we track transaction security by chain length as well as new block generation speed? Those should let you determine the probability of your transaction being rejected no matter the mining incentive.

modeless 5y ago
The probability you care about is the probability that your transaction will be reversed by chain rewriting. The cost of chain rewriting is the amount of work it would take to make a longer chain starting from the block before your transaction. However, chain length is simply a proxy for the amount of work done, because each block has the same difficulty (except at a difficulty adjustment). The number of blocks or speed of blocks is not the relevant measure. The number that matters is the total accumulated work proved in the blocks. Dogecoin has faster blocks and more blocks than Bitcoin but it is not more secure. Its difficulty is much lower, so each block proves much less work, and rewriting its chain is therefore easier.
- omegalulw 5y ago
  I didn't mean to compare speeds for comparing difficulty for two different currencies, I agree that's largely meaningless. Rather, the key probabilities I would like to compute would the probability of chain rewriting (I don't like this term tbh; if someone forks off a new chain they haven't rewritten the one your block is in, rather they have "convinced" others to abandon it) within the next T days given an adversary with more compute than the Bitcoin network by a fraction f, for various T and f. Can't I approximate that with 1) the lead that my current chain has over the next competing chain, 2) the average time it's taking to mine a new block given the current compute? Ill confess I'm not super familiar with the intricate details of crypto, I may just be missing something obvious.
  - modeless 5y ago
    You measure the "lead" of a chain by the amount of work it proves, not by the number of blocks or speed of blocks. If you based the lead solely on block number then an attacker's chain could manipulate the difficulty adjustment to make the attack easier. > Can't I approximate that with 1) the lead that my current chain has over the next competing chain An attacker could keep their chain secret until the attack is sprung. You cannot know how far ahead or behind you are relative to a secret chain. What you know is the amount of work an attacker would have to do, by summing the work proved by all the blocks above and including yours. Once that value gets high enough, you can be pretty confident that nobody is going to rewrite your transaction. The probability that you would have to calculate is the probability that an attacker who has that much compute exists.
    - omegalulw 5y ago
      Thanks! This is good food for further reading, I had no idea miners could influence difficulty this way. I had thought, for Bitcoin, difficulty depends on how many coins mined till now so I (implicitly) assumed that small parts of the chain would have similar difficulty.

modeless 5y ago

The probability you care about is the probability that your transaction will be reversed by chain rewriting. The cost of chain rewriting is the amount of work it would take to make a longer chain starting from the block before your transaction. However, chain length is simply a proxy for the amount of work done, because each block has the same difficulty (except at a difficulty adjustment). The number of blocks or speed of blocks is not the relevant measure. The number that matters is the total accumulated work proved in the blocks. Dogecoin has faster blocks and more blocks than Bitcoin but it is not more secure. Its difficulty is much lower, so each block proves much less work, and rewriting its chain is therefore easier.

omegalulw 5y ago
I didn't mean to compare speeds for comparing difficulty for two different currencies, I agree that's largely meaningless. Rather, the key probabilities I would like to compute would the probability of chain rewriting (I don't like this term tbh; if someone forks off a new chain they haven't rewritten the one your block is in, rather they have "convinced" others to abandon it) within the next T days given an adversary with more compute than the Bitcoin network by a fraction f, for various T and f. Can't I approximate that with 1) the lead that my current chain has over the next competing chain, 2) the average time it's taking to mine a new block given the current compute? Ill confess I'm not super familiar with the intricate details of crypto, I may just be missing something obvious.
- modeless 5y ago
  You measure the "lead" of a chain by the amount of work it proves, not by the number of blocks or speed of blocks. If you based the lead solely on block number then an attacker's chain could manipulate the difficulty adjustment to make the attack easier. > Can't I approximate that with 1) the lead that my current chain has over the next competing chain An attacker could keep their chain secret until the attack is sprung. You cannot know how far ahead or behind you are relative to a secret chain. What you know is the amount of work an attacker would have to do, by summing the work proved by all the blocks above and including yours. Once that value gets high enough, you can be pretty confident that nobody is going to rewrite your transaction. The probability that you would have to calculate is the probability that an attacker who has that much compute exists.
  - omegalulw 5y ago
    Thanks! This is good food for further reading, I had no idea miners could influence difficulty this way. I had thought, for Bitcoin, difficulty depends on how many coins mined till now so I (implicitly) assumed that small parts of the chain would have similar difficulty.