I think the assumptions that go into the design of such protocols matter regardless of transparency and what could cause failure is the bias towards the assumptions of the people, or machines, devising the protocols. There are unknown unknowns going forwards and significant enough social change that any fixed set of protocols may have trouble keeping up. A potential solution would be baking change into the protocols but that is, I think, very difficult. I'm not making an argument against open government or protocols or progress just that weather forecasting is difficult especially over the longer term.