Which makes me wonder... did they just pulled a Murphy's, or are the services so unstable that they go down if no one's overlooking it? Maybe the services already go down multiple times a day, but the outage is short?
Also, these guys are in engineering. They are very likely not even directly involved when there are outages. They build the systems and protocols to avoid and recover from outages, but don't actually perform the work themselves. It's developers vs. IT.
Correct, it's pretty much impossible for an outage to not be noticed and the GMail on-call being automatically paged.
SREs at GMail are engineers, yes, but they're very much directly involved with fixing outages - not so much at the 'try turning it off and then turning it on again' level, more the 'redirect all traffic away from this cluster into a different one, while we roll back the broken update'.
SRE is a combination of problem-solving when there are outages, and building tools to 1) automate away the manual jobs involved in massive-scale system administration so that outages are less likely to occur.
Doesn't sound like an organisation that would miss services going down even briefly.