4 ms·

In particular, whether it can gain enough mining power to survive. Currently, SegWit2x (the fork discussed here) has 92% of miners signalling that they'll suppo

In particular, whether it can gain enough mining power to survive. Currently, SegWit2x (the fork discussed here) has 92% of miners signalling that they'll support it[1]. The question in this case is whether the original Bitcoin could survive after the fork if this support is maintained. [1] https://coin.dance/blocks https://coin.dance/blocks

metroidfan832 9y ago
No, the difficulty can be adjusted at fork time so it doesn't matter. Literally anyone can fork.
- stale2002 9y ago
  That would ALSO be a hard fork though. IE the legacy chain would be dead. Yes, the legacy chain people can create their own new fork, but then it has to deal with all the disadvantages of being a hard fork, and the old chain will be dead.
  - Nursie 9y ago
    Why would the legacy chain be dead? Bitcoin Cash is a hard fork of Bitcoin, both still run.
    - stale2002 9y ago
      Currently 93% of the hashpower is signaling that they will switch from the legacy chain to the new hard fork. If the main chain has a very low hashpower, then anybody and their mother can attack it/double spend/ ect very easily. The miners on the fork could even commit some hashpower on their own to attack the old chain, so as to make sure that the new one wins. Bitcoin cash never has the support of any significant amount of hashpower, so of course the main chain is still around.
      - Nursie 9y ago
        True, but event 7% of the BTC hashpower is quite significant, and we may find that the 'old' branch lives on as another "Classic" or something. Guess we'll see.
schoen 9y ago
> The question in this case is whether the original Bitcoin could survive after the fork if this support is maintained. When I read this kind of statistic, I'm a bit confused about why not, unless the 92% are planning to use some of their resources to try to actively disrupt the original chain (by censoring 100% of transactions or something).
- jeremyt 9y ago
  If the new chain gets, say, 1% of the hashing power, then blocks will be found on average 100 times less often, which extends the time to reaching a new adjustment period from two weeks to 200 weeks...enough to arguably kill it.
  - schoen 9y ago
    Thanks for that observation. It seems like this sort of "make the adjustment period take a horrifically long time to reach" effect only happens if the cutover is very abrupt and includes the overwhelming majority of hashpower, and if people are persuaded enough of the outcome of the fork that they don't speculatively bring new hashpower online to take advantage of the suddenly-easier-to-earn block rewards.
atemerev 9y ago
This is interesting. If this is correct, it amounts to a power grab, transferring decision power from Core maintainers to the miners consensus. However, algorithmically speaking, miners consensus _is_ Bitcoin. The only authority Core developers have is rooted in tradition. So far, it was enough to get miners to agree with their point of view... but it will be challenged eventually.
- d33 9y ago
  Correct me if I'm wrong, but they also control the software and could divert the users to a different fork if they wanted, right?
  - atemerev 9y ago
    Well, no. They can release the reference client, but most users rarely employ it for transfers (most Bitcoins are either in cold storage, or in online wallets like Coinbase). Anyway, after the hardfork, you'll get your balance in duplicated both chains (as all transactions before the hardfork are visible). The market forces will decide then how much each part costs.

metroidfan832 9y ago

No, the difficulty can be adjusted at fork time so it doesn't matter. Literally anyone can fork.

stale2002 9y ago
That would ALSO be a hard fork though. IE the legacy chain would be dead. Yes, the legacy chain people can create their own new fork, but then it has to deal with all the disadvantages of being a hard fork, and the old chain will be dead.
- Nursie 9y ago
  Why would the legacy chain be dead? Bitcoin Cash is a hard fork of Bitcoin, both still run.
  - stale2002 9y ago
    Currently 93% of the hashpower is signaling that they will switch from the legacy chain to the new hard fork. If the main chain has a very low hashpower, then anybody and their mother can attack it/double spend/ ect very easily. The miners on the fork could even commit some hashpower on their own to attack the old chain, so as to make sure that the new one wins. Bitcoin cash never has the support of any significant amount of hashpower, so of course the main chain is still around.
    - Nursie 9y ago
      True, but event 7% of the BTC hashpower is quite significant, and we may find that the 'old' branch lives on as another "Classic" or something. Guess we'll see.

schoen 9y ago

> The question in this case is whether the original Bitcoin could survive after the fork if this support is maintained. When I read this kind of statistic, I'm a bit confused about why not, unless the 92% are planning to use some of their resources to try to actively disrupt the original chain (by censoring 100% of transactions or something).

jeremyt 9y ago
If the new chain gets, say, 1% of the hashing power, then blocks will be found on average 100 times less often, which extends the time to reaching a new adjustment period from two weeks to 200 weeks...enough to arguably kill it.
- schoen 9y ago
  Thanks for that observation. It seems like this sort of "make the adjustment period take a horrifically long time to reach" effect only happens if the cutover is very abrupt and includes the overwhelming majority of hashpower, and if people are persuaded enough of the outcome of the fork that they don't speculatively bring new hashpower online to take advantage of the suddenly-easier-to-earn block rewards.

atemerev 9y ago

This is interesting. If this is correct, it amounts to a power grab, transferring decision power from Core maintainers to the miners consensus. However, algorithmically speaking, miners consensus _is_ Bitcoin. The only authority Core developers have is rooted in tradition. So far, it was enough to get miners to agree with their point of view... but it will be challenged eventually.

d33 9y ago
Correct me if I'm wrong, but they also control the software and could divert the users to a different fork if they wanted, right?
- atemerev 9y ago
  Well, no. They can release the reference client, but most users rarely employ it for transfers (most Bitcoins are either in cold storage, or in online wallets like Coinbase). Anyway, after the hardfork, you'll get your balance in duplicated both chains (as all transactions before the hardfork are visible). The market forces will decide then how much each part costs.