What is blockchain? What makes a good blockchain? Is blockchain just a buzzword? Is a blockchain really trustless?
Let's start with the last one. The word trustless is often used to describe blockchain, however it is not really trustless — it would be better described as decentralised trust or systemic trust (trust in the system).
Rather than trust an individual (person or organisation) the trust is provided by the very system itself.
You do need to trust the transaction processors (aka miners) for the chain, and also the programmers that wrote the software being used. However you don't need to trust the individual processors or software, only that the majority are truthful. A characteristic of blockchain is that an individual bad actor, controlling only a small portion of the network, cannot take the network down.
A blockchain network can increase systemic trust through:
- A large number of processors (e.g. public chain)
- Multiple clients (i.e. multiple development teams)
- Open source (allow visibility and the option to reject changes and go a different direction)
Only two major blockchains rate highly on these trust criteria: Bitcoin (BTC) and Ethereum (ETH). There are strong reasons why these have strong support.
Some of the others are trying, but not yet there, and worryingly some of the type cryptocurrencies are not decentralised at all, but actually controlled by a single central organisation.
The risk of centralisation
Without these three characteristics, the network is at risk of centralisation, e.g. if there is only a small number of processors, then they can manipulate the system. Or if there is only a single client, written by a single team (or single author), then it is also at risk of being manipulated.
Ensuring code is open source both allows easily visibility and validation, as well as somewhat of a backup if the single client implements undesirable features, then it can easily be forked.
If trust is not decentralised, then the overhead of running as a blockchain is an unnecessary cost. Or if centralised trust is not an issue, then a distributed ledger (or database) may be a better solution.
A small trustless consortium may exist, but paying fees to run on a public chain (leveraging their decentralised trust) may be simpler than running your own hardware.
What is blockchain?
There have been many definitions of blockchain, ranging from a simple description of the data structure (just a chain of blocks), to a more complete definition that includes elements of decentralisation and public availability.
The key, in my opinion, is that the system provides a solution to provide systemic trust, without the need to trust a central authority.
A network, such as Bitcoin or Ethereum, has a lot of structure and coding designed to provide this systemic trust; it is the core of their design. If you have no need of such trust, such as a network controlled by a single organisation, then the overhead of this structure is unnecessary.
It would be like adding a seat belt, airbags, and roll cage to a living room chair and then advertising it as ANCAP vehicle safety rating of 5 stars. Sure, it might technically have been tested to that, but if you are just going to stick it in your living room then the cost was unnecessary, and it is never actually going to protect your from a car crash.
Evaluating the top blockchain networks
So, how well do the major networks meet these definitions? The following table is based on available data in November 2021:
| Name | Processors | Open Source | Clients | Notes |
|---|---|---|---|---|
| Bitcoin | 8 - 12,000 | Yes | Bitcoin Core (75%), Armory, Bitcoin Knots, Specter | |
| Ethereum | 3,000 | Yes | Geth (60%), OpenEthereum, Nethermind, Besu, Erigon | |
| Binance | 1 | No | Unknown | originally an ERC20 issued by Binance; now a privately run token |
| Tether | n/a | No | n/a | minted by a single organisation; claimed backed by reserves |
| Cardano | < 1,000 | Yes | Cardano Node | |
| Solana | 1,000 | Yes | Solana | |
| Ripple | 30 - 150 | Yes | rippled | network is decentralised; tokens are owned/sold by Ripple |
Bitcoin and Ethereum are the two clear leaders against the trust criteria. Both have multiple open source client implementations, requiring any change to be negotiated amongst multiple stakeholders, and a large network of processors.
Cardano and Solana meet most of the criteria, but have not inspired multiple client development teams, so have some centralisation risk for changes. However both are open source, so could be easily forked if needed.
Ripple is an interesting case with what is effectively decentralised permissions — there are no processor incentives (mining), but each processor has a list of the other processors they (individually) trust. The systemic trust is not based on an agreed algorithm (with economic incentives), but on mutual trust between the core participants. Furthermore all the tokens are issued/owned by a central source.
I think Ripple has an overall fairly weak level of trust, but is probably suitable for it's intended purpose (a consortium of payment providers), just not as a general blockchain.
The remaining two of the top cryptocurrencies, Binance and Tether, are almost fully depend on the goodwill of their issuers. There are probably many reasons why those companies should honour their goodwill, to remaining an ongoing business.
However, this is no different than, say, depositing your money in a bank, where you are relying on the bank to securely hold your money; they simply are not systems based on decentralised trust. They are both vulnerable to unilateral changes by their issuers.
Use cases for blockchain
Blockchain, when defined as a network that has decentralised trust, has real uses cases where a central trusted entity is not available or not desired. There are proven use cases as cryptocurrency, and for various associated distributed finance uses, where have distributed trust can be beneficial, e.g. resistant to outside influence.
Bitcoin and Ethereum satisfy these criteria, as do similar systems such as Cardano and Solana.
There are, however, many blockchain-like systems that do not feature a core based on systemic trust — in these cases it certainly seems that they are jumping on the blockchain bandwagon.
I would caution people to look at any project claiming to be blockchain and evaluate if they really need to complicated distributed trust systems that blockchain is built on. If they are not using them, e.g. there is a single issuer or network controller, then you have to question why they are using a more expensive blockchain solution, when a simple distributed ledger or even a regular database, may be better.