The first question was the most interesting: Is it a mistake to use the events of the recent past as a method of predicting the future? since this seems to be

The first question was the most interesting: Is it a mistake to use the events of the recent past as a method of predicting the future? since this seems to be a veiled criticism of Kurzweil's entire methodology -- he sees an exponential trend and extrapolates it way out into the future with no consideration for the fact that all exponential trends in the real world eventually hit a wall somewhere. Kurzweil, unfortunately, misses the point and uses it as a cue to go on about exponential trends again.