(Local | Global) fungible assets in Multi VM

[conr2d]

Description

Fungible assets in Noir can be classified as Local and Global assets.

Definitions

1. Local Assets:
   Fungible assets whose states are entirely managed within a single VM's context. These assets are designed to operate independently, with no shared state across other VMs.

2. Global Assets:
   Assets where the primary state transitions are managed by the engine itself while exposing interfaces for operations within individual VMs. Advantages of Global assets:

• Lower costs: Key operations are centralized in the engine, reducing VM-specific overhead.
• Shared state access: The engine allows multiple VMs to interact with the same asset state seamlessly through exposed interfaces.

However, Global assets come with a significant caveat: contracts must transition their state to use our contract SDK (instead of managing state independently), requiring developers to modify their contracts to comply with this approach. This introduces a learning curve and a proprietary workflow, which might not align with our protocol’s goal of simplicity and broad compatibility.

Alternative Approach: Cross-VM Asset Management

Instead of Global assets, we could explore a mechanism where Local assets remain fully managed within their respective VMs but are made accessible across VMs through an adapter layer.

The adapter would:

• Maintain metadata about the asset's state location (i.e., which VM manages the asset) and provide instructions for accessing it.
• Facilitate cross-VM calls to interact with the asset, ensuring that even non-native VMs can recognize and work with these assets.

For example:

• A standard ERC-20 contract could be deployed on the EVM.
• To interact with this asset from another VM, the engine would require registration (or upgrade) of the asset's metadata.
• This process would enable the engine to coordinate interactions while maintaining compatibility with existing contracts and minimizing modifications.