One of the most important arguments for Blockchain technology is its decentralized architecture. It provides security avoiding a single point of failure and censorship resistance as no single authority can decide and impose its criteria over the group. While this is true from the architecture standpoint it is quite evident the industry is starting to centralize and most of the big chains are "in the hands" of a few businesses. Bitcoin mining is controlled only by a few mining pools. Ethereum and several EVM nodes are quite concentrated at Infura and Etherscan with its "scan" white label solution is the preferred service for most of the L1 and L2 Ethereum-based chains.
Yes, you can set up your own node for many chains and access the chain directly with existing devices such as DappNode but the truth is 99% of traffic uses Infura and Metamask to access those chains. The result is whenever any of those services suffers, the whole chain suffers and whenever they decide to change the access rules, everyone will have to either find a different path or pay any additional cost, whatever the form of the cost is.
The infrastructure layer is key in this industry and there are only a few groups working in that direction. Last afternoon I had the pleasure to meet with one of them which is creating a competitor of Etherscan. Their solution is pretty good and they are getting ready to jump into their first client. Big chains depend too much on Etherscan as their chain data solution and having a competitor will only improve final user access and will make this type of service grow.
Although most of the users do not even know what a blockchain data scanner is, it is key to check chain transactions and account states, as well as directly access smart contracts to request data or interact with them.
I would love more and more projects would jump and develop in the infrastructure layer of W3 as we need to push some more into decentralization.