A relatively simple fix that could prevent planes from disappearing

It's important to note that the substance of the article here is not about improving tracking of airplanes while they are flying (which actually does very little to keep them from disappearing if a catastrophe stops them from flying) but is actually about tracking weather more accurately by gaining real-time data from other aircraft in the vicinity. This suggestion might indeed have helped prevent the most recent Air Asia crash, and perhaps also the Air France crash from several years ago. It takes good information to respond to rapidly developing storms at high altitude over the ocean, and to advise pilots how to change their courses (along heavily trafficked airline routes) to avoid both dangerous weather and collisions with other planes.

Can someone explain to me why they can't just stick a GPS receiver wired to Iridium satellite phone and duct tape it to the back of the plane? I.e. it shouldn't be neither hard nor that expensive to send a few dozen bytes of telemetry every minute via Iridium. Why anything like this is not happening?

The Microsoft project is a way of figuring out current upper wind conditions from data available from the aircraft that fly at those upper altitudes ... which is pretty cool, but doesn't help plan fuel loads by itself. For that you would have to use those derived wind speeds to generate a better forecast. You need to know what the winds are going to do in the future. Getting the upper wind forecast wrong doesn't cause a crash. At the most it might cause a diversion.

The Microsoft idea has nothing to do with the sort of bad weather that people think might of been an issue with Flight QZ8501. That bad weather mostly consists of thunderstorms which are small unpredictable areas of vertical air movement.

Drones are not likely to be of much help here as they normally operate at low altitudes.

This article really demonstrates the level of idiocy that armchair experts contribute to the discussion. It's as if the author had to publish a piece on the disappearance and use it to generate as many click bait responses as possible.

Firstly, for ground covered aircraft there is already extensive weather radar (the same used to demonstrate clouds on your tv's weather graphics on the evening news) which shows what is happening live. This is tracked by most airline operators and the hotspots can be rerouted from central if necessary.

Secondly most newer aircraft have a forward looking weather radar that can warn them as the plane flies of the immediate vicinity (though they don't have full visibility of what's on the other side). At the very least they can use this to do what happened here; request a variation in the path of the flight.

Thirdly the main problem of weather is in areas not covered by weather radar. This is where drones would be most useful; but that is also the problem: they need sufficient fuel to get them to stay in or around one place and unless there is automated return to base and refuelling options it isn't going to happen.

Fourthly the planet is big. Really big. You may think it's a long way down to the chemist at the end of the street etc. the point is knowing what the weather is like tens of miles away isn't really that useful for identifying the kind of cells causing bad weather which are generally known about in any case. To get the kind of required resolution you either have to have lots and lots of them, or send a leading drone out in front of the aircraft's intended path waving a red flag. Except to do that it has to fly at a similar speed to the jet which means each drone is going to cost many millions In hardware and fuel for each flight.

Lastly no pilot is going to want to fly through a mid air drone field because the chance of disaster occurring due to high speed impact with a stationary object is much higher than that of suffering weather damage. The reason planes fly at different altitudes depending on which direction they are travelling is to avoid meeting traffic coming the other way; generally because if you notice it it's already too late to avoid it.

I love the idea this article proposes, basically Waze for airplanes where real time weather data comes from the actual flights themselves and shared via traffic controllers or the planes themselves.

The most terrible reality is that the capability for tracking virtually all long-haul airframes (99.999%) already exists. It's called Inmarsat FANS-1/a ADS-C, which is now being offered to all airlines for free, post-MH370. It was present on the MH370 airframe (but for a cost), and could've located it to within tens of square miles instead of millions.

http://www.inmarsat.com/news/inmarsat-provide-free-global-ai...

http://www51.honeywell.com/aero/common/documents/FMS_DirectT...

http://members.optusnet.com.au/~cjr/CPDLC.htm

I am a bit surprised that he did not mention big data.

But there is point in this talk.

Of course this system will not cost 300 as it has to be dependable and it must kept properly calibrated.