Hey Andy! Thanks for the question… Most tokens that follow a standard like the ones you mention have the storage integrated within the same interface as the functions that govern the logic for how to manage that storage. We see this as a limitation, where the logic and storage are bound to one another by the standard itself.
This ERC standardizes the interfaces that abstract the implementation from the storage. To give you a clear example:
Let’s say you have 100 DAI on your account. This value is stored on a mapping within the ERC-20 contract, and you modify or access it by using the other functions listed on the standard (transfer(), balanceOf(), etc)
Now, in this scenario the assets are represented by a uint number, associated with your account. Your account has 100 DAI, whenever this storage says you do.
What we propose, is that the functions that modify this storage can be abstracted away from the data. For instance, you keep the storage on one contract and apply the logic on a different contract.
Back to the question: how does this work in ERC-7208? You can have Properties with storage values within the ODC contract, and you can have Property Manager contracts that implement the logic. Why is this a desired thing? Because by separating the logic, you get more versatility, at the cost of a delegate or delegateCall. Example of this would be: we can store a Property with value uint= 100 for your address, and then this Property is managed by two different Property Managers contracts, each of them implemented with a different interface. i.e. ERC-1155 and ERC-20. This is a theoretical scenario, of course… but it would mean that you both simultaneously own an NFT (1155) and 100 of an ERC20 token. Each Property Manager works as its own token contract, it just delegates the storage to the ODC.
Hope this makes things clear(er)