Blockchain, the best-known type of distributed ledger technology (DLT), has evolved enormously since 2009 when Bitcoin first nestled itself in the public consciousness. The objective of DLT is to enable us to trust the message, not the messenger.
This simple statement represents radical change – it is now the ledger itself, not its controller, that we place our trust in when making a transaction. So, what is DLT? It comprises of three words, all operative, and the first two words – distributed ledger – also define the meaning of the third. DLT is a set of technologies related to the operation of a distributed ledger that allows us to trust the message, not the messenger. The fact that consensus is the outcome of a collective, decentralised procedure will enable us to remove trust from the individual parties.
Today, we commonly refer to three "Layers" of DLT, but as will become apparent, this is confusing in its ordering and places one Layer into the picture that shouldn't be there, as it defeats the decentralised purpose of DLT and adds risk. We'll propose a logical framework that addresses the issue in the conclusion. Let us first elucidate the current state of things. We present layers in the chronological order they were introduced.
The foundation: Layer 1
Layer 1 is composed of the most widely known example of DLT, that is, blockchains. Their core functionality is to provide a distributed consensus procedure to ensure that transactions have occurred in a valid order so that, as mentioned in the introduction, people can trust the message rather than the messenger. However, Layer 1s do not communicate well with one another, so each relies upon their ability to capture and tie to themselves various projects as captive clients. They are silos, and the more such protocols emerge, the more complicated the creation of a single infrastructure becomes.
The most successful Layer 1 blockchain - Ethereum - suffers from high operational costs and does not scale for general use. Even post the "Merge", Ethereum will operate at 20 transactions per second, taking 20 minutes to settle transactions and a $20 "gas" fee to transact, which doesn't solve its scalability issue. The need to overcome Ethereum's shortfalls has led to the prominent re-emergence of technology as old as computing itself: batching of transactions. By art of marketing and at the cost of earnest communication, this has become known as a "Layer 2."
Scaling through Layer 2
Layer 2's collate transactions and then use a Layer 1 – such as Ethereum – to provide a consensus confirmation, notarising a large number of transactions as one. This is batching. Batching does create a ledger, but it isn't decentralised. Layer 2 services are centralised and demand considerable trust on the user's part.
Those risks are present whilst waiting for the batched notarisation of the transaction. Once notarised, the asset remains at risk within the (unregulated, usually uninsured) Layer 2 asset pool. As assets are not individually notarised, they cannot be segregated from the pool of assets in the 'layer 2' provider's asset pools until it is extracted to a Layer 1 protocol.
Further, that risk between transaction and settlement is not momentary – most Layer 2's take many hours to notarise the transaction and settlement; some takes days, some as much as a week. Simply put, if you are using batching to scale blockchain, then you invalidate the use of distributed ledger technology entirely. The whole point is that you can trust the message, not the intermediary messenger at scale, with near-instant settlement.
An infrastructural "Layer 0"
Layer 0 is - or should be - the infrastructure layer of DLT. It is the base upon which all things can be built without having to concern ourselves with writing code to comply with every network. It enables the DLT functions to fall away from sight and become an operative resource pool.
Many protocols are now claiming to be Layer 0 protocols. However, they are basing this upon their provision of bridges - which they operate and control, thus re-instating third-party trust - between the protocols they connect. As with Layer 2's, this forces the user to trust the intermediary and again, undoes the purpose of DLT. Interoperability allows users to enjoy the best features and services from the networked Layer 1 protocols, with none of the downsides. The impact of adding more protocols when you have interoperability is not to add complication. Rather, it adds scale - the capacity of each networked protocol adds to the whole.
Such interoperability would transition Layer 1 of DLT from being a fractured, siloed network of ecosystems, with a wide variety of programming languages and standards, to becoming a common resource layer. Such a protocol dispenses with the need for a Layer 2 solution to scale blockchain. Every transaction is provided with decentralised, public consensus. Finally, to be considered infrastructure, it should also be open. Fees are applied only to the cost of deploying and using the infrastructure rather than driving up the price of governance or a staking token.
Conclusions
An analogy to complete the picture: If we think of mobile phone networks, the Layer 1 resources are the mobile phone masts. A Layer 0 in this context would be like a virtual master network connecting you to the nearest and most suitable mobile phone mast to deliver the required service.
- Resource Layer (network of L1s)
- Infrastructure Layer (interoperable L0)
- Platform Layer (a mix of industry-specific services and layer 0 supports that assist the deployment of applications)
- Application Layer (services and products at the user level)
