> So what comes next? Since you provided a source for your numbers, I'll bite. Formalize results into mechanical proofs that can be verified by computers so th

> So what comes next? Since you provided a source for your numbers, I'll bite. Formalize results into mechanical proofs that can be verified by computers so that we build a library of computer proofs. You can't bullshit a computer. I'm not sure how it would work for statistical results, but defining a formalized standard might be a good first step for deriving numbers from raw data instead of relying on the authors to calculate the statistics themselves.