Section 5.2.3 of the draft paragraph 2:

A simple-type-specifier followed by a (empty) pair of parentheses constructs
a value of the specified type.  If the type is a class with a default
constructor, that constructor will be called; otherwise the result is the
default value given to a static object of the specified type.

Take this program as an example:

#include <iostream.h>

struct A {
  int x[10];
};

main()
{
  A avar = A();

  for(int i=0; i < 10; ++i)
    cout << avar.x[i] << " ";

  cout << endl;
  return 0;
}

As I understand this the array inside A should be filled with zeros.
I don't think this is how existing compilers work.  So how will this be
accomplished?  As I see it, either a) a real static object of type A is
created (which is guaranteed to be all zero) and copied to avar; or b)
the compiler has to generate a "everything to zero" copy constructor.
Both don't seem too implementor friendly.

Seems to me, if the programmer wanted his members initialized to ANYTHING
he would supply a constructor or initialize each instance.  What is the
rationale for this?

Comments?