NutBerry - smart contracts on Layer 2

Layer 2
1

NutBerry(Jelly) - Offroading Ethereum Transactions

I’m thrilled to announce the upcoming release of a permissionless layer 2 solution
with support for smart contracts, aka the NutBerry Project.
The design is similiar to what we generally call rollups but it intentionally has the same transaction encoding
and signing scheme as Ethereum transactions to be one-to-one compatible with the existing ecosystem.

The most anticipated feature is the possibility for on-chain EVM verification, that makes it possible to
run smart contracts on a permissionless / trustless layer 2 solution.
Though, the runtime has some restrictions like not be able to call other contracts.
I call that a state-minimized EVM or LEVM - Lean Ethereum Virtual Machine.

So, unlike PoA, PoS or other forms of custodial { child, side }-chains, the NutBerry offroading engine as I call it
is not another blockchain. Data availibilty is fully archieved on the root-chain and the contract is able verify and replay
all transactions either through directly finalising a blob of transactions on-chain (if you are in a hurry)
or via an interactive computation verification game to offload the computation but
enforce correctness on-chain for a given transaction blob.

Roadmap

  • 1st Milestone
    Supports only the ERC20 token standard.
    (Already does that via a fixed smart contract)

  • 2nd Milestone
    Going to support the ERC721 standard.

  • 3rd Milestone
    Support stateless smart contracts.

  • 4th Milestone
    Stateful smart contracts, aka smart contracts with support for storage.

All resources will be published on https://github.com/NutBerry
once the first milestone is complete.

Ask me anything :wink:

3 Likes
2

:muscle:Released :ship:

2 Likes
3

LEVM - Lean Ethereum Virtual Machine

sounds cool :partying_face:

1 Like
4

@johba Thanks for asking :slight_smile:
I’m cross-posting the question here for everyone else:

is nutberry using optimistic rollups?

At the moment I would call NutBerry a half-optimistic rollup :laughing:
The reason is that it can verify a block by replaying on the root-chain but NutBerry doesn’t use state-roots or intermediate state roots yet.
And if the block is too big to be replayed on the root-chain, NutBerry relies on the verification game alone to resolve the outcome.
(Btw, using the verification game is default modus operandi)
But after implementing a gated computing model I can basically remove the negative aspects of the dispute game,
together with state-roots NutBerry will actually become a optimistic rollup system :slightly_smiling_face: