Bringing True Decentralization to Ethereum (ETH): Infura Has Got to Go
For a long time, Infura has been one o Ethereum's most popular, but also most controversial, technologies. It is responsible for handling over 13 billion code requests every day, and it also allows developers to enter the blockchain without running a full node. However, according to experts, it is also the reason why so many dApps were created on ETH network, due to the simplicity of interaction.
So, what is the problem? Well, the problem is that Infura is operated by one single provider, which is the ConsenSys development studio. Not only that, but it also relies on Amazon's cloud servers. As such, many consider it to be the weak spot of the Ethereum network, one that might bring to its failure if left unresolved.
Simply put, every dApp that has any connection to Infura would stop working if Infura was ever turned off. This is a big concern, and while Infura's contribution to ETH ecosystem and its importance to the network is undeniable, a centralized service cannot be allowed to serve as a pillar of the developer community.
This is why numerous developers have decided to try and distance Infura from Ethereum by attempting to bring forth decent alternatives. However, while there have been numerous attempts to achieve this, a proper method that can fully replace Infura has yet to emerge.
dApps Depend On Infura
Right now, there are around 11,803 full nodes on Ethereum. About 5-10% of them are Infura-based. However, Infura is under constant maintenance, which makes it highly reliable, and also more than capable of handling huge amounts of traffic. In other words, the issue is not how Infura does its job. The problem lies in the fact that it is a centralized service.
The reason behind using a service like this is its ability to handle large quantities of data in the first place. A full active node on Ethereum needs to be able to handle as much as 1 Terabyte of data, which is clearly much more than a regular laptop can handle. By using Infura, developers can focus on the software, and not worry about hardware or storage.
The same is true for users who often use a tool called Metamask for holding their coins. Metamask depends on Infura just as much, which means that almost all dApps do too.
Another issue regarding Infura is privacy, as it uses Amazon's storage. Also, if Amazon ever decided to cut ties, most of the functioning dApps on Ethereum would stop working.
Reaching True Decentralization
As mentioned, there are numerous attempts to reach an alternative to Infura and cut the network's dependence on it. One such potential solution is a new code library called LightJS, which was released by Parity Technologies. The goal is to convince developers to create light clients instead of relying on Infura. Light clients require less storage and less dependence on hardware, while the degree of decentralization increases.
The hope is that less and less future dApps would choose to connect to Infura, and would instead go for light clients.
Other projects that are trying to reduce dependence on Infura include VIP node, Dappnode, D-node, and others, all of which aim to convince people to run their own full node. Many believe that this is not already the case as there are no rewards for running full nodes. This is something that VIP node aims to change, and hopefully, attract developers.
D-Node aims to create a market between node operators and developers, but it also wants to decentralize their economic relationship. To achieve this, it uses DAO — Decentralized Autonomous Organization.
As for Dappnode, it was created in order to allow developers to set up their local networks and design it in a way that will make dApp deployment easier. Some of these projects, especially VIP node, also receive funding from Infura itself.
Solving The Problem
ConsenSys, the firm behind Infura, is also one of Ethereum's largest startups. It also announced that it is funding an Incubator project, which is to cut Metamask's dependence on Infura. And even Infura itself is attempting to help by adding different cloud providers so that Amazon would not be in full control of all ETH dApps.
Everyone is trying to add more decentralization to Ethereum, even Infura itself. It originally became popular thanks to the nature of Ethereum's platform, which is capable of producing and executing dApps instead of just using the cryptocurrency for making transactions.
In addition, the hardware is of great importance for Ethereum network, as Ethereum also stores a sum of all computation that happens on the platform. This is called “state”, and it keeps growing as more users interact with the platform. This makes hardware more expensive, as well as more difficult to use, which also doesn't encourage users to get involved.
These issues have been noticed and recognized a long time ago. While different methods were developed in order to resolve them, they all have their own flaws that need addressing. Even Ethereum's co-founder, Vitalik Buterin, proposed a method of fixing some of these issues which comes down to rewriting the underlying incentive so that users would get rewards for running a full node.
This is one of the changes that were included in a new upgrade called Ethereum 1x, which is expected to be implemented at some point in 2019.