3 ms·
That's a narrow take. Detecting an earthquake early and alerting millions of people quickly are both hard problems, and both had to be solved for this to work.
by pythonaut_16 4y ago
That's a narrow take.
Detecting an earthquake early and alerting millions of people quickly are both hard problems, and both had to be solved for this to work.
Much of the hard work on Google's side just happens to overlap with other use cases that they've already scaled.
- ilyt 4y agoWere it be built from scratch earthquake detection is still harder part than message distribution. "deliver same message to subset of subscribers" scales horizontally pretty easily.
- pythonaut_16 4y agoBe my guest to go build one then. Oh but if you're truly committed you'll need to avoid Cloud computing resources, avoid Twillio, don't use any Google services, no Firebase Cloud messaging... You probably shouldn't use Android phones either... And remember any delayed or failed messages put people's lives in danger... Or we can just admit that Google as a whole had some impressive achievements here alongside the Earthquake team.
- lttlrck 4y agoCome on, this is a mean-spirited mischaracterization of what they wrote.
- pythonaut_16 4y agoI thought it was fair given their oversimplification of what delivering messages at scale requires. The point isn't that the Earthquake Detection isn't impressive. Why do we have to deny Google any credit here? Is it just because they're a big tech company? IMO it feels like another "I could build Dropbox in a weekend" comment.