Don wrote:
What has to be kept in mind is that you are predicting a quake that has the potential of causing damage and death. When this kind of a situation takes place and you see the potential for a large quake you can revise your prediction before the main event takes place.

Dennis replies:
Back in January of '96 when I was being evaluated on sci.geo.earthquakes, I had posted a prediction with smaller windows then I normally use. I did this since most, not all, quakes fell within the smaller time frame to improve on the probability of getting a hit by chance. Normal is 10 days. At the end of the smaller window nothing had occurred so I extended the window to the full 10 days. This didn't go over to well and the extension was treated as a separate prediction and not as an extension.

Dennis