2 ms·
But constructivism does not give up continuity. All of the traditional definitions carry over. What you do have to give up, though, are discontinuous function
by btilly 14y ago
But constructivism does not give up continuity. All of the traditional definitions carry over.
What you do have to give up, though, are discontinuous functions on continuous sets. They are not well defined for numbers that are the result of an infinite sequence converging near the border of the discontinuity. You can still give them a formal treatment, of course, but they aren't really functions.
However if you're attached to your step functions, never fear. It would not be the first function-like thing that isn't a function. For another example consider the well-known derivative of the step function, the Dirac delta. Not actually a function in classical mathematics, though it is certainly useful.