if you regard pure math as "the real thing" when it comes to mathematics, then it is a shame in my country that the first exposure that students have to this is

if you regard pure math as "the real thing" when it comes to mathematics, then it is a shame in my country that the first exposure that students have to this is in 2nd year of university studies, if they opt in to analysis & abstract algebra. primary school, high school and the more applied university classes drill rote learning and ability to compute, but don't teach ability to work with abstractions, unpack definitions, extend definitions to new settings, search for, present and critique proofs. from the perspective of wanting to share the beauty of pure math with a wider audience, it'd be great to introduce earlier (at least optionally). on another hand, introducing widely applicable tools for conducting and evaluating experiments would probably be far more useful to the world -- i'd be tempted to chuck out or defer a lot of early rote learning of trig, calculus, integration and replace them with probability, statistics, causal inference, experimental design. or at least offer both as different streams. arguably, this would be more useful to a lot more people than concentrating on pure math.