If people are going to depend on github being up for their CI workflows, etc., there should be serious effort expended at the network and cache layer to be suit

If people are going to depend on github being up for their CI workflows, etc., there should be serious effort expended at the network and cache layer to be suitably reliable. It's probably fine to not be able to do developer-level actions for hours, but even a 5m outage in deploying other systems is unacceptable for most businesses.