Meni Rosenfeld explains prerequisites for fork arbitrage:
http://fieryspinningsword.com/2015/08/25/how-i-learned-to-stop-worrying-and-love-the-fork/
If coins in one fork have to be decoupled from the coins in the other fork through Coinjoin before they can be moved (without someone maliciously using the signed tx to trigger the same move in the other fork), it seems far too cumbersome to have that persist indefinitely.
However, with the futures market idea where people send their coins to the exchanges and buy/sell rights to the coins in each branch of the coming fork, redeemable after the fork happens, it seems like only the exchanges would need to do the Coinjoin transaction. Maybe they'd only have to do one big one, then they can arrange the payouts.
Moreover, since the futures market would essentially act as a prediction market, it may be very clear which fork would win quite a while before the fork actually happened, and - for instance - miners could use that information to decide to start mining on XT or BU. That would be miners responding to investor intent. Then no worries about decoupling because the other chain is dead.
Thoughts? An alternative would be changing the signing algo completely, but that also seems extreme.
Reasoning backward from the astonishing invulnerability and robustness Bitcoin would need to withstand attacks at trillion-dollar market caps, I imagine a world where forks and spinoffs are happening all over the place, potentially, with the ultimate goal to make Bitcoin truly as close as possible to being "just a ledger with balances controlled by private keys" from the user's perspective, with any specific protocol abstracted away from being essential to that, so that users simply think of their privkey as their money access key and know they can choose from a variety of software/networks to achieve the actual movement of money to other addresses.
http://fieryspinningsword.com/2015/08/25/how-i-learned-to-stop-worrying-and-love-the-fork/
If coins in one fork have to be decoupled from the coins in the other fork through Coinjoin before they can be moved (without someone maliciously using the signed tx to trigger the same move in the other fork), it seems far too cumbersome to have that persist indefinitely.
However, with the futures market idea where people send their coins to the exchanges and buy/sell rights to the coins in each branch of the coming fork, redeemable after the fork happens, it seems like only the exchanges would need to do the Coinjoin transaction. Maybe they'd only have to do one big one, then they can arrange the payouts.
Moreover, since the futures market would essentially act as a prediction market, it may be very clear which fork would win quite a while before the fork actually happened, and - for instance - miners could use that information to decide to start mining on XT or BU. That would be miners responding to investor intent. Then no worries about decoupling because the other chain is dead.
Thoughts? An alternative would be changing the signing algo completely, but that also seems extreme.
Reasoning backward from the astonishing invulnerability and robustness Bitcoin would need to withstand attacks at trillion-dollar market caps, I imagine a world where forks and spinoffs are happening all over the place, potentially, with the ultimate goal to make Bitcoin truly as close as possible to being "just a ledger with balances controlled by private keys" from the user's perspective, with any specific protocol abstracted away from being essential to that, so that users simply think of their privkey as their money access key and know they can choose from a variety of software/networks to achieve the actual movement of money to other addresses.
Last edited:
