[Bitcoin-development] Double-spending unconfirmed transactions is a lot easier than most people realise
tomh at thinlink.com
Wed Apr 23 22:29:33 UTC 2014
On 4/22/2014 9:03 PM, Matt Whitlock wrote:
> On Tuesday, 22 April 2014, at 8:45 pm, Tom Harding wrote:
>> A network where transaction submitters consider their (final)
>> transactions to be unchangeable the moment they are transmitted, and
>> where the network's goal is to confirm only transactions all of whose
>> UTXO's have not yet been seen in a final transaction's input, has a
>> chance to be such a network.
> Respectfully, this is not the goal of miners. The goal of miners is to maximize profits. Always will be. If they can do that by enabling replace-by-fee (and they can), then they will. Altruism does not factor into business.
The rational miner works hard digging hashes out of the ether, and wants
the reward to be great. How much more valuable would his reward be if
he were paid in something that is spendable like cash on a 1-minute
network for coffee and other innumerable real-time transactions, versus
something that is only spendable on a 15-minute network?
There is a prisoner's dilemma, to be sure, but do the fees from helping
people successfully double-spend their coffee supplier really outweigh
the increased value to the entire network - including himself - of
ensuring that digital cash actually works like cash?
More information about the bitcoin-dev