I think we agree.
Your own "anecdata" shows how catastrophic DNS failures are and why it's justified to expect 100% uptime with proper redundancy. Too bad you had providers not architecturing properly this key part of their infrastructure to the point you had to build your own solution.
