A lot of discussion around blockchain focuses on Layer 1 and Layer 2 networks, but modern Web3 applications depend on a much larger infrastructure stack.
A typical application may rely on several independent systems.
RPC infrastructure provides access to blockchain nodes.
Indexers transform raw blockchain data into something applications can query efficiently.
Oracles bring external data such as prices, weather, or real-world events on-chain.
Data availability layers make transaction data accessible for rollups and other scaling systems.
Bridges move assets or messages between networks.
Sequencers order transactions in many Layer 2 architectures.
Each component introduces its own trust assumptions and failure modes.
This creates an interesting question: when we call an application “decentralized”, how much of its supporting infrastructure also needs to be decentralized?
An application might use decentralized smart contracts but depend on a centralized front end, centralized RPC provider, centralized sequencer, and a small oracle set.
Technically decentralized does not always mean operationally decentralized.
Which infrastructure layer do you think receives too little attention from users and developers today?
My guess would be RPC infrastructure and indexing, because most users never see those layers until something stops working.