[brewmarche]

> The usual "0.1 + 0.2 != 0.3" is _not_ imprecision.

Correct. That’s using the wrong base. Decimal floating point doesn’t have that problem.