Why We Need To Think About Zero-Confirmation Transactions

Firstly, let me recap on exactly what a zero-confirmation transaction is, as summarised by CCN:

“Bitcoin transactions are committed to the blockchain about every 10 minutes. Zero-confirmation transactions do not reside in a block yet. Instead, they reside in the memory pool of miners. Until a block is mined that includes the transaction, it is said to have ‘zero-confirmations. Once included in a block and written to the blockchain the transaction has one confirmation”.

Often, many merchants and exchanges won’t accept transactions until they have a confirmation, sometimes they require as many as six, this means they can be left in limbo, unable to develop until a (or several) confirmation(s) have been made and the transaction has been written to the blockchain. The main reason for this is to prevent Double Spend Attacks, instances that occur when a transaction is broadcast numerous times throughout the Bitcoin network, the transaction that is accepted by the highest number of miners is the one which will go on to be confirmed, however if times right, these sort of attacks can leave fake transactions that will eventually move on towards confirmation, these transactions are technically invalid and thus hold no value. At the recent Satoshi’s Vision Conference, Tom Harding discussed Native Respend Resistance, an idea that works within Bitcoin Cash specifically to reject double spend attacks, or, respends. According to Harding, Respend Relay is the key to this, through digital wallets issuing an alert when a respend is detected. It does this by simply hesitating before confirming the transaction, when the transaction is picked up, the network will be monitored for a short amount of time to ensure no respends occur, if this is the case the payment is accepted, otherwise, it’s rejected. Harding also discusses how this idea has stemmed from a comment by Satoshi in 2010, referred to as the ‘snack machine’ comment. This refers to the problems that comes hand in hand with fast transactions, one of which is the ability to duplicate transactions temporarily, to fool nodes into accepting respends. Harding is highlighting that there is still a risk within Bitcoin Cash, in that zero-confirmation transactions that are in limbo, can lie as potential respends, giving more incentive to miners to prevent these transactions making it through the blockchain does seem like the only viable solution, a statement initially proposed by Hal Finney in 2001, according to Harding’s talk. With Bitcoin Cash in mind, the risk of respends is high, there’s no doubt about it, in his talk Harding validates these opinions, his solution, the Respend Relay in line with Finney’s recognition of offering greater incentives to miners for spotting these attacks should hopefully reduce these risks if implemented quickly and effectively. Featured Image Source: The Blue Diamond Gallery - Great Photo By ImageCreator