Your father thinks that blacking-out is similar to the movies, in that there are fast acting tranqs and other things that will 'knock out' a person and then they wake up with a headache. Anesthesiologists take at a minimum about 12 years of training before they touch a patient, even for 'low risk' dental procedures. Trying to black out a plane full of highly agitated people until the pilots can land or 'do something' about terrorists in the act is a total fantasy and may cause more deaths from false alarms and faulty wiring than the terrorists would have caused.
[0]https://en.wikipedia.org/wiki/Choking_game [1]https://en.wikipedia.org/wiki/Cerebral_hypoxia
The interesting thing for me, reading the comments is that my actual comment (that if security costs 10x to defend and 1x to perpetrate, then the system will by definition fail) has no feedback. I think about ways to "even" the cost on these things. Biometric security is one (although it too has its issues). Using something like Clear[2] at an air port to insure people are who they say they are, and having vetted that person via other mechanisms prior to their flight, seems more efficient and harder to subvert than the current system. It converts the problem from "150 unknown actors are getting on a plane" to "150 known actors are getting on a plane." Does it make it difficult enough for the attackers? They have to forge biometrics and disguise the actual flight they are taking. Not sure if it completely levels the field but it raises the cost for the attacker.
[1] He also thinks that every state should allow and encourage people to carry hand guns so that "crooks and crazy people would think twice."
[2] See http://clearme.com basically at signup biometrics and other data are used to confirm identity and to take a picture. When you arrive at the airport your boarding pass is scanned (an presumably validated) and the operator gets a picture of what you should look like given your biometrics. So if you steal someone else's fingerprints you don't look like them.
That's only if you have an infinite supply of evil doers. The cost in operational experience and manpower can deplete the number of operatives that can pull off a complicated attack.
Passing out from lack of oxygen is dangerous.
This is why medically induced hypothermia is sometimes used on heart attack patients, so that the brain doesn't swell up and kill itself after normal oxygenated blood circulation is restored.
So let's not casually mess around with air passengers' oxygen supply, mmkay? It would almost be better to just flood the passenger cabin with vaporized marijuana extract. The risk there is that someone might stage a fake attack just to turn the plane into a giant bong.
We know the "no O2" method is very rarely needed, simply from observed statistics.
We know that O2 masks come down more regularly than that. Such stories are not hard to find. Here's a list: http://www.avherald.com/h?search_term=Cabin+pressure&opt=256... . Looks like a few dozen per year.
If there were a switch like you suggest, what is the probability that it will engage by accident, and prevent the masks from otherwise deploying as they should?
I agree with the author's thesis that if it costs flyers 10x the effort to fly than it does the evil doers to do evil, then the evil doers are winning that battle. And the existing practices are due for a new look over. But not sure that eliminating them completely will achieve what you want.