There has been many accounts in recent weeks of the dangers of a 51% attack on the bitcoin network and what that would mean. Indeed the mining pool Ghash.io did manage to grapple the majority from the network which sparked doomsday folks to start calling for the end of days for bitcoin. Fear not, for there are several reasons why this would not happen, and there are many reasons why a majority in the bitcoin processing network doesn’t mean the end of the world.
Double spending attacks.
As long as a majority of CPU power is controlled by nodes that are not cooperating to attack the network, they’ll generate the longest chain and outpace attackers.
The first thing that people are quick to point out is that if a group of people owned more than half the network hashing power, then they would be able to double spend coins. While this is technically true, it is realistically impossible. The reason boils down to human nature. A mining pool is a organization that distributes the hashing load over many participants. Each one of these individuals is contributing to the team effort to solve a block and in return is being paid by pool for their services. They have no incentive to contribute their hashing power if it became known that the pool that they were contributing to was in reality using their majority to cause double spends, which would erode trust in bitcoin, and thus erode the value of their payouts. Satoshi developed the mining system with careful attention to the balance of incentive and rewards for all bitcoin network participants. This system does not change just because the role of one of these participants has became distributed. The dis-incentives for a single entity to launch a 51% attack on the network remain the same if the entity was pool of individuals controlled by a single entity. Once the network were to notice a large number of orphaned blocks coming from a certain majority pool rejecting other transactions, then I believe a large number of the offending pools participants will leave the pool to join another. Assisting in a double spending attack as a member in the pool has no benefit to any given individual. The double spent coins would belong to a small few individuals given the number of transactions per block, and the vast majority of the pool members will only see the value of their bitcoin eroded by the loss of confidence in the network once the attack was noticed.
Transactions would be small
The transactions that would possible to be double spent would be of small amount of coin. This is because in order to execute a double spend attack, one has to pay for a good with some coin, have the goods delivered, and then have the transaction backed out. Most retailers selling goods of significant value will usually wait for several confirmations before confirming the sale and delivering the goods. The only merchants that may not would be ones selling items about the value of a coffee. Not much of a benefit considering the value loss in bitcoin due to a theoretical double spend attack.
Any double spend attack, once discovered and publicized, would immediately get the mining pool put onto a blacklist, and thus others on the network would ignore blocks being published by that pool. This feature is not currently built into the protocol, but it may be in the future. Bitcoin, being a distributed network, like the internet, is very resilient to damage, and can re-route around damaged sections.
Summary, 51% majority attacks:
Not as bad as people are lead to believe.