On preview I see a bunch of folks have already chimed in this morning while I was typing. So some of this is redundant.
There have been industry discussions about something like the OP proposes. Or at least a “distress mode” which would start transmitting data when things started getting weird / dangerous.
The essential point and this applies to all the threads on all aviation events any time is that there’s only so much “budget” for adding safety to the system. I put “budget” in quotes, because I’m not just talking about money. There’s only so much bandwidth available in the RF spectrum. There’s only so much space in the avionics compartments to add new boxes. There’s only so much engineering talent & regulation writing ability and international cooperation, and … to go around.
And every time some talking head on TV starts a campaign to “fix” some perceived problem that’s pretty much a one-in-a-million event, resources are diverted from the things which will actually, demonstrably, save lives by preventing the “almost accidents” that happen every day.
Had MH370 had 100% real-time CVR/FDR download & 100% real-time position tracking the exact same thing would have happened to those folks. Nobody would have lived or died one second differently than how it really happened. The only thing that would be different today is that a billion people’s media-fed morbid curiosity would be a little more satisfied, and the CTers would have a head start on inventing their own competing version(s) of events.
The industry spends a very big “budget” every year on the things which do enhance statistical safety. It’s a very big system, with insane reliability standards and massive needs for backwards compatibility and universal international buy-in. So things move slowly.
For many proposed improvements, we’re already dealing with a system with failure rates down in the 1 in 10 million bracket. When things work that well, there are a lot more ways to introduce an unexpected bad consequence than to introduce an improvement. Doctors may have adopted it first, but “First do no harm” is a good motto for this process too.
But the real action is in the “almost accidents”. Right now, today, every day every aspect of most flights is being recorded onboard. Hundreds of megabytes per hour of data. On the newest jets its more like a couple gigabytes per hour. And all of it is downloaded a few days later while the jet is parked and later dug through looking for anomalies. Both crew performance and aircraft mechanical issues.
Out of these tens of thousands of flights per day per aircraft type and per company, patterns emerge. Every driver knows of that one nearby bad intersection where car accidents seem more frequent and you yourself have had a couple close calls. Aviation is now so safe we don’t have enough accidents to use them as a good guide to what *really *needs fixing. But we can now watch the “close calls” and direct resources to addressing them. Because they are what *does *need fixing.
It’s been a decade since we had the last US collision between an aircraft on a runway and something or someone that shouldn’t have been there. But right now there is a HUGE effort to adjust procedures, equipment, and mindset nation- and world-wide to reduce the frequency of close calls, which had been high & rising from, say, 2005-2010. Billions is being spent on ATC equipment aimed specifically at this threat and major engineering is ongoing on new generation cockpit tools to also reduce the risk. Meantime many procedures, some subtle and some pretty significant have been put in place to add more layers of backup & error avoidance to this safety concern.
But since nobody has been killed, CNN and therefore you, are 100% unaware of this effort. But it is bearing fruit and the so called “runway incursion” rate is dropping fast at most locations.
There are still some hard cases to fix. Chicago O’Hare is about 2/3rds through a multi-decade multi-Billion dollar effort to build new runways and taxiways in a less collision-prone layout compared to the 1940s-shaped reality it is today. Between buying land, getting suburbs to accept noise where it wasn’t before and all the rest, this isn’t easy.
Boston Logan is next. It’ll be 10 years before we get through the politics. Oddly, one of the biggest objections locals have is that since nobody is dying, it ain’t broke so it doesn’t need fixing. It’s hard to convince a hostile audience that reducing the error rate from one in 5,000 to one in 50,000 is worth them having jets directly overhead every 2 minutes UFN. But that reduced error rate converts into a few statistical lives saved every decade.
That’s the reality of how aviation safety proceeds. I’d never claim, and neither would the experts, that every dollar is always spent on the 100% closest snake and with god-like omniscient wisdom. But a lot of good people are putting a lot of serious thoughtful non-political, non-kneejerk effort into it.
And the results are paying off in myriad ways you don’t see when you don’t crash. But we see them all, because we’re doing them all.
An oddity is that as we wring out the last few thousandths of a percent of systemic predictable risk (no exaggeration), what’s left are the bolt-from-the-blue events. The next aircraft to go down may well be hit by a meteor. And we’ll be here on the Dope debating how to meteor-proof airliners or develop a real-time meteor reentry detection and warning system, complete with a planet-wide grid of instant 100% reliable secure communications.