professor. >> i was also going to say that i'm still troubled by the fact that if it was flying for five hours, it should have gone further than the search area they're talking about. so i do think what you would do is look at the directions and go with the flight speed the airplane was flying and propagate that out to a search little further out than they're showing. so i don't know if there's something i'm missing. >> david gallow that gets to the point you raised earlier. andy pasztor, in terms of your reporting that it was a manual change of direction, how can they know that it was manually done? >> because one of the -- the system that sends up data from the aircraft to a satellite was not switched off at that point. it took them maybe between 14 and 20 minutes or something around that time, whoever was doing this, to make the transponder stop working and disconnect the satellite communication system. and in that period of time is when the course change occurred