I wonder if curation will eventually be the "job" of the protocol itself or if all of the algorithmic curation will happen on the client side 🤔

Seems like it'd be best of being on the client end with open source and fair/visible curation rules