NEAR Protocol

NEAR is a proof-of-stake blockchain built around sharding and usability. It splits state and processing across shards that grow with demand, and replaces hexadecimal addresses with human-readable account names — a design choice that says more about its priorities than any throughput figure.

Nightshade shards the block, not the chain

Rather than running separate chains, every block contains chunks produced by each shard, so a single chain of blocks is maintained while work is divided. Validators verify only their assigned chunk. This keeps a unified chain history while distributing the processing load, and it means shard count can change with demand rather than being fixed.

Accounts people can read

Addresses are names rather than hashes, and accounts support hierarchical sub-accounts and multiple keys with different permissions. A key can be limited to specific contracts and spending limits, so an application key cannot drain a wallet. This is real security architecture, not cosmetics — most losses in crypto come from a single all-powerful key.

Fees partly return to developers

A share of gas fees is paid to the contract that was called rather than entirely to validators, giving builders a revenue stream tied to usage. Almost no other chain does this. It changes the economics of building an application from purely token-issuance-funded to partly usage-funded.

The remainder is burned

Fees not rebated to contracts are destroyed. Combined with issuance to validators, net supply change depends on how heavily the network is used — the same broad structure Ethereum has, applied on a sharded base layer.

Where it competes

NEAR's argument is that mainstream adoption fails on user experience before it fails on throughput, and its account model is a direct answer. The counterargument is that most liquidity and tooling gravitated to EVM chains, so usability advantages have to overcome a substantial ecosystem gap.