> In computer applications, rational values have limited precision,
regardless of how they are implemented. It is well known, for
example, that the value 0.1 cannot be represented exactly in the
floating point format standardized in IEEE 754 (the most common
format, and also the format used in this extension library).
In applied science and engineering, this is relatively unimportant
because the input values often originate from measurements or
approximative computations, which are imprecise to begin with.