Block confirmation with N-1 state root hash

[OnedgeLee]

This discussion is about manner, advantages, disadvantages and difficulties of N-1 confirmation.

For now, block indicates state root hash, which is from state after transactions of current block executed.
We have several advantages if we use state root hash before the transactions...

• Both of validators and non-validators will calculate state transition during appending time, so timing of block sync will be more close.
• Validators can propose / validate block within shorter time, and consensus will be much more robust.
• ...

But we have to handle...

• State root hash of current state just floats on the storage, and does not appear on the block explicitly.
• Meaning of state root hash of the block is changed
• ...