One advantage I think the parent is alluding to, is that zero-based indexing makes it a bit easier and more natural to generalise between continuous (functions

One advantage I think the parent is alluding to, is that zero-based indexing makes it a bit easier and more natural to generalise between continuous (functions on R) and discontinuous (functions on N) situations. Since the latter are often used to approximate the former, this is quite handy.