For a long time, many people have worked on the problem of creating decentralized digital money. All pieces of the puzzle precede Bitcoin except one: how to ensure that monetary units can not be worn more than once . This problem is itself a derivative subproblem how to make digital money that is monetarily useful by controlling money creation.
Bitcoin each previous solution involved having to rely on a third party or an infinite inflation . Third parties may fail or rebel and an infinite monetary inflation destroys properties why someone would want to use digital money.
There are algorithms that are decentralized Byzantine consensus before Bitcoin, but all participants are defined in the algorithm. A third part where you have to trust, albeit decentralized, remains a problem.
What does the test work ?
Test work accomplishes two things.
A, it allows the consensus emerges in a group of participants not defined . No lists of miners who are allowed to create blocks: Miners can come and go without implying a disruption of the system. The old miners can not prevent new mining compete with them.
And two, creates a cost for a double impossible to avoid expense .
To make a double spending on Bitcoin, a miner must create (at least) two versions of the block at a certain height. Since the blocks are only accepted if they contain a valid proof of work, there is no way that a miner can avoid paying the cost of energy expended in the creation of extra blocks.
In a period of time that a miner could have created n blocks, if you decide to make a double spending can only hope to obtain the corresponding n-1 block payment, so you will lose money if the benefit of double spending does not reach compensate for that loss. That behavior is what Bitcoin can promise with regard to the resistance to double spending nor more nor less.
Resistance double spending does not depend on the number of miners. Even if there were only one miner, the miner also lose money if efectuase double expense for a lower value to extend the chain invested in a block less than it could have spread if not incur a double expense.
Right here it is where academic cryptographers throw up their hands and shout that Bitcoin is broken because it only provides a linear guarantee instead of multilinear security guarantees. This happens because they are applying incorrect threat models.You need super-linear properties of security to ensure that an individual can encrypt your emails safely. Trade, however, does not need that degree of certainty.
Entrepreneurs and businessmen manage risks consistently and the economy works despite the risks can not be reduced to zero . In terms of risk of double spending, Bitcoin is enough to say, "I can define under what conditions someone would lose money to reverse a payment". While the risk is predictable, it can be accurately marked its cost, and while that can be possible to conceive an economy.
Even if an economy can operate with a double protection against relatively small expenditure, should not we try to improve it?
Yes, but only if the attempt to get something better not break anything else. Given the objective of "decentralization at all costs" will produce more resistance to double spending needed by the economy, and usually at the cost of restricting the ability to make payments -that say why a coin there first place-.
And what of resistance to censorship?
Miners lose money by not including lucrative transactions in the blocks, in the same way to lose money by double spending, so Bitcoin has the same resistance to censorship (DoS attack) as double spending.
Users can reduce their risk of being target of a DoS attack accustomed to proper hygiene privacy so that the attacker does not know which transactions belong to the entities it wants to attack.
What if the miners derailed?
The miners sell a product to their customers. Right now, customers are almost exclusively investors in Bitcoin. Miners get only profit if your customers continue to buy your product, and if stopped producing what investors want, miners can be fired.
Dismiss miners in a system like Bitcoin is complicated. You can not eject a miner individually; only all of them at the same time. The most obvious way that all investors and users can do this is to agree a test of different work from a certain point where "all" means "a sufficiently large part of the economy so that even those who disagree have to unite to avoid further losses. " After this step, the existing mining industry would collapse due to the absence of consumers and would be replaced by a new one.
Is it dangerous that investors have so much power?
Yes it is.
Right now, a dangerous split liquidity between Bitcoin and non-Bitcoin goods must pass through the points centrally regulated control. That is a serious vulnerability. The way out of this situation is to grow the Bitcoin economy as quickly as possible so that liquidity is provided by the direct exchange of goods and services rather than through exchanges. It is much harder to make a law ordering millions of people not accept Bitcoin to close a dozen bank accounts respected.
This is achieved by having more transactions within the block chain generating benefits for miners. The best way to protect against an attack Bitcoin is to build a self -enclosed economy need hundreds of transactions per second, then thousands, then millions .