Double Spend Notification



Summary:

The author of this post is questioning why a blockchain is necessary and discussing Ripple's lack of one. They explain that with blockchain-based transaction confirmations, the miner who finds the transaction block has full authority to decide which transactions are real and fraudulent, and even has the option to not include particular transactions, leading to network confirmation taking more than 10 minutes or even an hour for multi-block confirmations.The author proposes a solution where only 51% of nodes in the network need to confirm a transaction as legitimate for it to be confirmed by the entire network, with malicious nodes being excommunicated. They discuss two challenges: informing everyone when a transaction hits 51% confirmation and dealing with a triple-or-more spend.The first problem can be solved by having each node see the transaction twice and the first node to see it hit 51% making the announcement. The second problem can be reduced to a majority-wins problem. If a transaction is more than 6% ahead of others, it wins, and if no clear majority is hit, the miner can choose the transaction with the most votes at the time. This would only occur for a transaction if the spender of the coins attempted a double spend.In a separate email, Peter Todd discusses a new type of double spend defense where you can produce a child transaction that pays 100% of the double spend to fees, hurting the double spender but not profiting from it. Todd also proposes a mechanism where you credit the customer account immediately with half of the deposit amount and burn half to fees if the customer tries to double-spend, leaving the other half to pay for what they did spend. This mechanism could be used by BTC->alt-coin exchanges and services like SatoshiDice.


Updated on: 2023-06-06T17:46:09.418213+00:00