> float default is NaN, why... just WHY!!!!! > char default is 0xFF.... Invalid values, perhaps? Other than making initialization mandatory, this seems like a

> float default is NaN, why... just WHY!!!!! > char default is 0xFF.... Invalid values, perhaps? Other than making initialization mandatory, this seems like a reasonable way of catching them. For integer types it's not really possible because all their values are valid.