Saw the NOVA special on Flight 370's disappearance.
Very interesting explanation of the IMMARSAT effort there - but I have to say, it looks a lot like counting straws in a haystack. The delta in the numbers that were derived to get a sense of direction and distance were a few parts per billion. I realize that they reduced the data until a 'trend' could be identified that gave some idea as to probabilities, but it seems kind of fringe, in practical terms. The satellite was actually moving due to end-of-intended life low-fuel issues, for one thing.
Fascinating though, how they could measure the time the signals would need to travel from airplane to satellite, etc. I got some idea that given the amount of time until the last 'handshake', the bird would have to be well up fully into either the northern, or down in the southern arc; if north, that, if true, would have put it up in Kazakhstan somewhere - or conversely, roughly where the search has been occurring along the southern arc.
Sounded good, but one still wonders how firm this logarithm distilling really is in terms of accuracy. As one searcher put it, there is some mounting doubt about ever finding the plane in the southern Indian ocean.
Might it take 77 years to find this one too? I surely hope not.