You will be able to explain why rewriting bitcoin's history is expensive and what confirmations mean.
Darren, a 29-year-old engineer, has S$2,000 he is curious about putting into bitcoin. A colleague tells him the exchange will credit his deposit "after a few confirmations", and another one says the whole ledger could be hacked one day by someone with enough computers. Darren nods at both and understands neither. By the end of this lesson you will be able to explain both to him.
Lesson 1.1, Bitcoin is a shared ledger that nobody runs, showed how a payment is signed, broadcast and bundled into a block by a miner. This lesson is about why, once it is in a block, it is very hard to change.
A hash is a short fingerprint produced from any piece of data. Feed the same data in and you always get the same fingerprint. Change a single character, even one digit of one payment amount, and the fingerprint comes out completely different. Nobody can predict in advance what fingerprint a given piece of data will produce. You only find out by running it.
Each bitcoin block contains the hash of the block before it. That is the "chain" in blockchain. Picture a stack of numbered receipts where every receipt has the previous receipt's fingerprint printed at the top. If someone edits receipt 500, its fingerprint changes. Receipt 501 now carries a fingerprint that no longer matches, so 501 has to be redone, which changes its own fingerprint, so 502 has to be redone, and so on to the top of the stack.
On its own, that would only make tampering obvious. Anyone could still redo the whole stack on a laptop in a second. What makes it expensive is the puzzle from lesson 1.1.
To add a block, a miner has to find a hash that meets a target set by the protocol, such as a fingerprint starting with a long run of zeros. Since nobody can predict a hash, the only way is to keep changing a spare number in the block and trying again, over and over. This is proof of work: the valid block is proof that someone spent a great deal of computing effort finding it.
Now go back to the attacker who wants to change block 500, perhaps to undo a payment they made. They have to redo the work for block 500 and every block after it, and they have to do it faster than the rest of the network is adding new blocks on top of the honest version. Nodes follow the chain with the most accumulated work, so a slower attacker never catches up.
To pull ahead, the attacker would need more computing power than all the honest miners put together. That means buying or renting enormous amounts of specialised hardware and paying for the electricity to run it, all to attack a network whose coins would probably fall in value the moment the attack became public. This is why people describe bitcoin's security in money terms. The ledger is not protected by a password or a company's promise. It is protected by the cost of outworking everyone else.
There is a limit to this argument, and you should know it. The protection is only as strong as the honest computing power behind a chain. Smaller blockchains with fewer miners have been rewritten in this way, because renting enough power to overwhelm them was affordable. When someone tells you a blockchain is "unhackable", ask how much it would cost to outwork its honest miners.
When your payment is included in a block, it has one confirmation. Every new block added on top of that one adds another. Each extra block is another layer an attacker would have to redo, so the chance of your payment being reversed shrinks quickly as confirmations pile up.
That is why an exchange does not credit Darren's bitcoin deposit the moment he sends it. It waits until the deposit has a certain number of confirmations, and each exchange sets its own number and states it in its deposit help pages. With blocks arriving roughly every ten minutes on average, a deposit can take a while to show up even when nothing has gone wrong. A small payment between friends might be accepted after one confirmation. A large one sensibly waits for more.
The same logic explains why a confirmed mistake stays a mistake. If Darren sends coins to a wrong address, every new block buries that payment deeper, and there is no operator who can step in and pull it back.
Proof of work deliberately burns energy, because the cost is what makes cheating unattractive. The bitcoin network as a whole uses a very large amount of electricity, and that is the main environmental criticism of it. Estimates of the total are published by university research groups and vary with the method used, so if the figure matters to your decision, look up a current estimate and read how it was made.
Supporters reply that some mining runs on surplus or renewable power. Critics reply that the energy buys a payment system that handles far fewer transactions than card networks. You do not have to settle that argument to understand the design. Just know that the security and the energy use are the same thing seen from two sides, and that lesson 2.1 covers how Ethereum moved away from mining for this reason.
Darren's two colleagues were each half right. Deposits wait for confirmations because each block makes reversal harder, and the ledger can in principle be rewritten, but only by someone willing to outspend the honest miners. You can watch the first part happen on any public block explorer: search for a transaction, and the page shows how many blocks now sit on top of it. Older transactions show thousands. A brand new one may show none yet, and you can refresh the page and watch the count start to climb.
Find a transaction on a block explorer and note how many confirmations it has and how long ago it was included.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).