Let a have a look.
A friend of mine, in circles that would know, after some discussion with somebody else who I don't know, thought that if the plane had been transporting illegal banned radiotive, biological or chemical weapon, and something went wrong, the pilot had military training and would have ditched the plane where it did. That is his theory, but it is important to find it to prove what "actually happened".
Now reality, they claim a position based on some servicing signal sent to satelites. At first they claimed not to know if the path was over the indian ocean or towards Eurasia. Then they claimed to have determined it was to the Indian ocean. How do we know any of this is true, or not in some way mistaken. If the above scenario of something being covered up where true, it could have been turned towards a destination going towards Eurasia, with all aligned nations misreporting it. If terrorist related and a mistake made, it could be sent the same path, with everybody dead from lack of oxygen. In all these cases independent verification, and in this case, verifications that these are actual recordings of the aircraft at the time, can yeild some answers.
Now, let's get towards some much less controversial options, which I'm pretty much hoping might be it. Signals are such little things and anylsis with only so much resolution. So, mistakes can move the zone to where the plane is not. Altitude winds and local winds, can affect the length of flight, making it fall short, left, right or long. The pilot can choose a undesirable altitude or heading, and do so after the last ping, even fly in circles, go straight up or down, turn engines off at anytime, or dump fuel and end up way short. We don't know.
Now, I pretty much have said this all along, that a detailed micro analysis of current, wave and wind movements against the shape of items washed up on African beaches in 3D simulations, can back track to likely point of origin of the crash zone (if we have the data and processing power). But, then again, how do we know that the items are actually from the plane, and also not just dumped in the ocean.
Now, the earth is being photographed by satelites. Some are amazingly detailed, some are not. It is possible to see the effect of a much smaller item than the satelite's pixels, especially over time. Discolorisation, or odd brightness level, of a pixel indicates something might be there. It is possible to track items from crash to where they wash up, and in reverse, over subsequent pictures. The size and shape of items will have an average effect over adjacent pixels depending on its orientation and placement relative to pixel boundaries, wave action, sun angle and weather. But basically a computer program could be made to detect and track the more visible items back to the crash site. Volunteers could scan the Indian ocean for crash debris from shoots on the day or week. If it can't be found, then it probably didn't happen at that location.
The sea floor. The cheapest way to scan existing and new areas are by drone. Using suitable wavelengths of light to penetrate water and pickup reflections, in a photographic or lidar like fashion, you can pickup images. Far, far cheaper than what was done, when you apply cost over future crash scenes. There is a scientific drone that works on little energy that glides up and down the water column, using the pressure differences to reduce energy consumption (plus put a solar cell on it). Combined with a suitable energy source it could glide along the bottom for short stretches come up transmit data recharge (though certain nuclear options available to US and Chinese navies are an option) would eventually get through a search area. The issue is that over a certain period of time things can get covered, moved and/or float down a trench or other feature and be covered. So time is of the essence.