You may have read or not that on September, 15th, Yoshio Kushida, a self-taught Japanese radio astronomer, went public predicting there would be a strong earthquake in Tokyo within the next two days. His theorizing in a nutshell: Earth magma movement is different before an earthquake, influencing ions in the earth atmosphere leading to detectable differences in the electromagnetic field.
He was not taken very serious by most researchers. With the news yesterday (and last week), some researchers already have said their: 'Kudos to...' and publicly announced they would now have a closer look. Kushida was off in location, magnitude and time, but he was too close to shrug off his idea.
It may turn out to be a one hit wonder (many earthquakes in Japan, predict one for next year and you'll be right) or it may turn out to be the onset of making earthquake prediction as good as one-week weather prediction, for a start. We'll see. However, it shows how science works: Make a successful prediction and you will not be ignored any longer. However, before your ideas get accepted finally, your track record must be much better than one hit.