How is it then that over almost seven years all of our statisticians, programmers, disciplinary experts and other data scientists missed all of these warning signs and failed to adjust how they were using Twitter? Have we simply stopped looking at the data we analyze? Or are we just chasing fads?