Ironically, emails and essays seem to be out more error prone than mathematical proofs... 1. Proofs are intended to be correct. The vast majority of computer

Ironically, emails and essays seem to be out more error prone than mathematical proofs... 1. Proofs are intended to be correct. The vast majority of computer programs are intended to be practical, and are far far larger than the largest math proofs. Where correctness matters, computer programs have fewer bugs. 2. Nearly every non-computerized mathematical proof is not even a proof, it is a sketch that requires a lot of hand waving and logical jumps to be interpreted. Likewise, specs have fewer bugs than concrete program, because they aren't complete.