4 ms·

To be fair, there are times in bitcoin when two hashes are found consecutively. In this case, who decides which one is the correct hash? The answer is that the

by lojack 9y ago

To be fair, there are times in bitcoin when two hashes are found consecutively. In this case, who decides which one is the correct hash? The answer is that the network enters a contentious state and the miners compete over the correct hash until a new one is found ad infinitum. The chances of there multiple competing chains becomes virtually zero as time goes on. The same thing happens with IOTA, except the mining happens on the node sending the transaction (and confirming two others). Competing transactions can happen, and are propagated through the network, but eventually a consensus is reached. I’m only familiar with IOTA in passing, but the biggest flaw I see is of a malicious actor flooding the network with transactions to themselves. I believe the theory is that once the network is big enough this will economically not be feasible, but until then they have the master node. Whether or not this will end up working out, I’m not sure. Happy to be corrected for any of this. I’m not super familiar with IOTA, this is just my understanding of it.

csomar 9y ago
The contentious state in bitcoin is a network fork. Fork A and Fork B. Both have equal chances. It is the next winning miner that decides which fork will be elected. So the responsibility is just pushed to the next guy. That's not the same with IOTA. Who gets to decide which fork is better? Based on what? Number of txs? That wouldn't work and I think it is obvious why.
- lojack 9y ago
  The network fork is exactly what I described. With IOTA, it operates in a similar fashion. The winning chain is the one with the most successful hashes discovered. Just like bitcoin, except instead of a winning chain happening after ~2 hashes, it happens after hundreds(?) of hashes.
  - csomar 9y ago
    How do you reach a consensus and certainty though? What if there is low activity, few transactions; and then comes a big guy who was burning hash to generate txs and take over the chain?
    - lojack 9y ago
      > How do you reach a consensus and certainty though? I feel like we're going in circles. In both Bitcoin and Iota, we never reach an absolute certainty. In Bitcoin we reach a point where its infeasible for transactions to ever be overwritten. For high value transactions this is generally around 5 or 6 confirmations. I don't know the exact number people use with Iota, but I remember hearing it was some percentage of confirmations within the tangle. > What if there is low activity, few transactions; and then comes a big guy who was burning hash to generate txs and take over the chain? That's the big flaw I pointed out in my original comment, that I suspect may exist.
      - csomar 9y ago
        > In both Bitcoin and Iota, we never reach an absolute certainty. That's not correct. You are only uncertain in bitcoin when there are two equal (hash wise) blocks. This happens very rarely. That's not the case in IOTA since it does happen all the time and I'm guessing somebody could calculate txs to make it happen on purpose.
        lojack 9y ago
        > This happens very rarely Not true. Orphaned blocks happen all the time. https://blockchain.info/orphaned-blocks https://blockchain.info/orphaned-blocks > You are only uncertain in bitcoin when there are two equal (hash wise) blocks. Not true. Its possible to see the longest chain and not know about another chain that is of equal length. Nodes won't propagate chains that are of equal length to their current chain, so this is actually probably a somewhat common scenario. There has also been a number of orphaned blocks of length 2, 3, and 4. That means its totally possible (not likely, but not outside the realm of possibilities) for a node to be working on a chain and have never seen the correct previous block. https://bitcoin.stackexchange.com/questions/3343/what-is-the-longest-blockchain-fork-that-has-been-orphaned-to-date https://bitcoin.stackexchange.com/questions/3343/what-is-the... This is all exactly why many nodes don't accept transactions until there are 6 confirmations -- statistically, we shouldn't ever get to an orphan chain of length 6 even though its technically possible (outside of bugs). FWIW, I'm not arguing that IOTA is any more or less secure than Bitcoin.

by lojack 9y ago

csomar 9y ago

The contentious state in bitcoin is a network fork. Fork A and Fork B. Both have equal chances. It is the next winning miner that decides which fork will be elected. So the responsibility is just pushed to the next guy. That's not the same with IOTA. Who gets to decide which fork is better? Based on what? Number of txs? That wouldn't work and I think it is obvious why.

lojack 9y ago
The network fork is exactly what I described. With IOTA, it operates in a similar fashion. The winning chain is the one with the most successful hashes discovered. Just like bitcoin, except instead of a winning chain happening after ~2 hashes, it happens after hundreds(?) of hashes.
- csomar 9y ago
  How do you reach a consensus and certainty though? What if there is low activity, few transactions; and then comes a big guy who was burning hash to generate txs and take over the chain?
  - lojack 9y ago
    > How do you reach a consensus and certainty though? I feel like we're going in circles. In both Bitcoin and Iota, we never reach an absolute certainty. In Bitcoin we reach a point where its infeasible for transactions to ever be overwritten. For high value transactions this is generally around 5 or 6 confirmations. I don't know the exact number people use with Iota, but I remember hearing it was some percentage of confirmations within the tangle. > What if there is low activity, few transactions; and then comes a big guy who was burning hash to generate txs and take over the chain? That's the big flaw I pointed out in my original comment, that I suspect may exist.
    - csomar 9y ago
      > In both Bitcoin and Iota, we never reach an absolute certainty. That's not correct. You are only uncertain in bitcoin when there are two equal (hash wise) blocks. This happens very rarely. That's not the case in IOTA since it does happen all the time and I'm guessing somebody could calculate txs to make it happen on purpose.
      - lojack 9y ago
        > This happens very rarely Not true. Orphaned blocks happen all the time. https://blockchain.info/orphaned-blocks https://blockchain.info/orphaned-blocks > You are only uncertain in bitcoin when there are two equal (hash wise) blocks. Not true. Its possible to see the longest chain and not know about another chain that is of equal length. Nodes won't propagate chains that are of equal length to their current chain, so this is actually probably a somewhat common scenario. There has also been a number of orphaned blocks of length 2, 3, and 4. That means its totally possible (not likely, but not outside the realm of possibilities) for a node to be working on a chain and have never seen the correct previous block. https://bitcoin.stackexchange.com/questions/3343/what-is-the-longest-blockchain-fork-that-has-been-orphaned-to-date https://bitcoin.stackexchange.com/questions/3343/what-is-the... This is all exactly why many nodes don't accept transactions until there are 6 confirmations -- statistically, we shouldn't ever get to an orphan chain of length 6 even though its technically possible (outside of bugs). FWIW, I'm not arguing that IOTA is any more or less secure than Bitcoin.