Is that even knowable? Like, I know they called it “The Astonishing, Unfailing, Bell System” but if they had an outage somewhere did they actually have an infrastructure of “canary phones” and such to tell in real time? (As in, they’d know even if service was restored in an hour)
Not trying to snark, I legit got nerdsniped by this comment.
They absolutely did. Note that the reliability estimates exclude the last mine because trees falling and the like but they had a lot of self repair, reporting, and management facilities.
Engineering and Operations in the Bell System is pretty great for this.
Is that even knowable? Like, I know they called it “The Astonishing, Unfailing, Bell System” but if they had an outage somewhere did they actually have an infrastructure of “canary phones” and such to tell in real time? (As in, they’d know even if service was restored in an hour)
Not trying to snark, I legit got nerdsniped by this comment.
They absolutely did. Note that the reliability estimates exclude the last mine because trees falling and the like but they had a lot of self repair, reporting, and management facilities.
Engineering and Operations in the Bell System is pretty great for this.
Running a much simpler system with much more independent nodes.
It's a lot easier to keep packets flowing than to keep non-self-contained servers serving.