Did you read the entire article? The author says that floating point math is both 1) fine 2) not the cause of the issue under discussion. Floating point is a ti

Did you read the entire article? The author says that floating point math is both 1) fine 2) not the cause of the issue under discussion. Floating point is a tiny digression in a much larger piece. > Floating point arithmetic might be weird, but it’s very consistent and well-specified: the IEEE 754 specification is rigorously implemented. So it’s not arithmetic...