> Binary floats are unable to represent 1/10 just as decimal floats are unable to represent 1/3. That is true, but most humans in this world expect 0.1 to be r

> Binary floats are unable to represent 1/10 just as decimal floats are unable to represent 1/3. That is true, but most humans in this world expect 0.1 to be represented exactly but would not require 1/3 to be represented exactly. Because they are used to the quirks of the decimal point (and not of the binary point). This is a social problem, not a technical one.