Why is a blockchain hard to change, and what separates open, private and shared networks? Here are the key principles.
Once a transaction has been added and confirmed, it is in practice permanent. Each block is locked to the one before it with a cryptographic fingerprint.
To change an old entry, an attacker would have to redo the work for every later block while controlling the majority of the network. That is so costly that in practice it does not happen on large networks.
The ledger sits on many machines, not on one server. There is therefore no single place that can be attacked or switched off to stop the system.
Users hold their own cryptographic keys. Whoever controls the key controls the assets, which is also why secure storage of keys matters so much.
Public blockchains such as Bitcoin and Ethereum are open to everyone. Anyone can take part, see the transactions and help secure the network.
Openness brings a high degree of trust and resistance to censorship. The drawback is that such networks are often slower and less private.
A private blockchain is run by one organization, which decides who may take part. It offers control and privacy, but loses much of what makes blockchains unique.
A consortium blockchain is run jointly by a group of organizations, such as banks or trade finance players. It provides shared records without one party holding all the power.
Whatever the type, blockchains share the same goal: a reliable record of transactions without a central authority. The difference lies in how much openness is traded for control and speed.