Who exactly determines how an IP address should be split to ensure that a massive network doesn't collapse under its own broadcast traffic? Furthermore, this process involves "borrowing" bits from the host portion of an address to create smaller, more manageable logical segments. Why is mastering the binary math of subnet masks and CIDR notation the most critical technical hurdle for aspiring network engineers?