From earlier postings:
... have all instrumentation output, voice and even possibly video (flight deck and cabin) continuosly downlinked to ground stations or uplinked to satellites enroute.
1) In addition to traditional "black boxes" there would be transmitters in planes that would transmit flight data to a network of ground stations 3) Since the transmitters don't have to survive an impact they wouldn't have to be terribly expensive so you'd be able to intall them even on relatively small aircraft. 6) Overseas flights might need to switch to some sort of satellite system. This would probably be more expensive so would probably be limited to larger aircraft.
This capability has existed since the early 1980s. It is called ACARS, which stands for Aircraft Communications Addressing and Reporting System. I developed firmware for the Bendix system purchased by Delta, Piedmont and Northwest Orient in 1985. Those same systems (which sold for, if I recall correctly, about US$18,000) are in use today. Little, if any, of my other career work has had similar longevity: it is a (sobering) reminder of how long it takes to make improvements in commercial aviation. You can learn more about modernization efforts, including satellite, overseas, and encryption capabilities, by surfing over to www.arinc.com and searching on the keyword ACARS. I do not know whether the airlines are interested in using this to augment the cockpit voice recorder but it would make sense. Twenty years ago, the system was limited to 2400 baud and it was essentially used as a "time clock" for automating the process of tracking hourly air crew wages. The other two primary uses were to provide rapid turnaround on jet engine performance parameters (which reduced service costs), and to expedite bi-directional weather reporting (pilots could retrieve or file weather reports into an automated system). The aviation industry was fun to work in back then but more recently it has been no joy at all (the last project that I participated in was ETMS, which is a USA radar tracking network whose control room is in Kendall Square, Cambridge, MA): too much bureaucracy and too much money to get far too little done. It just doesn't seem hard to me to come up with a snazzy database system to keep track of 10,000 airplanes in real-time with data, audio, and video feeds. A dot-com could've done the equivalent task on a few tens of millions of VC money in the space of 3 years, whereas the US government manages to plunder $billions and waste decades. Pardon me for venting. But if someone wants to give it a try, perhaps for some country other than the USA, then I'll come join you on building a "real" aviation network. -rich