Here's a quick summary: user actions are saved onchain but the computations associated with those actions are computed offchain (i.e. not by hived nodes but by 2nd layer app code). Since all the data necessary to reconstruct the state of the 2nd layer app is stored onchain, the 2nd layer ecosystem is decentralized as long as the app is open-sourced because you could run the 2nd layer app code on your system and get the same results as anyone else would, hence there is no centralized server required.
You are viewing a single comment's thread from:
thanks that helped :)