Even
seemingly straightforward
questions, for example «is X safe» and «is X not safe» seem like (effectively) asking the same
question but they may require entirely different analysis and give different results — because of uncertainty in the data, in the results, the
logical conclusions / inference required etc. etc..