|Quoting zanl188 (Reply 133):|
Denying a pilot planning suicide by plane those extra minutes accomplishes what exactly? He/she will simply crash the plane in another manner which precludes any action by other crew members
To preclude any malicious actor from having any more than a minute or two to act is a pretty good improvement over not doing anything to stop them short of shooting the plane out of the sky if detected by the ground with all ending up dead. Every minute counts planes can hit a ton of targets within a few minutes of their cruising position let alone the hours they could have if not detected in time.
|Quoting 2175301 (Reply 134):|
The base problem is "human relations" based; and technical solutions have proven to be horrendous in their ability to solve human based problems. It is precisely because engineers and scientist tried to apply technical solutions to human based issues that the respect for engineers and scientist declined so much from 50-75 years ago.
One of the defining factor of human life is the mediation and use of technology throughout our lives. Planes themselves are a perfect example of the fusion of technical expertise and human interaction to connect the planet together. Could you imagine human society without any technology because I don't think it is possible to separate human factors from technologies it is bi-directional.
Doors are also part of the "human relation" and were invented ages ago to control access and are too a system of technologies and inventions refined over the ages. We close a door when we want privacy, security, isolation, protection. And we open a door when we want to welcome people and so on. People who abuse a secure door to lock out others to perform nefarious actions to the severe detriment of those people don't deserve a use case.
The base problem is not just a human relations problem there is 0% chance you can both predict and stop someone from snapping. Improving screening processes and addressing pilot stress factors is all good but doesn't stop someone from snapping one day due to a completely random reason which we cannot control for because we are all human. (People are not computers)
The assumption behind the combined door override is that we can trust people as a group which accepts the possibility for bad actors and mitigates attacks from within. In essence it is still very much a human solution as you can't just have a pile of keycards thinking for themselves.
|Quoting 2175301 (Reply 134):|
In reality, the 2nd person in the cockpit - regardless of qualifications - is a human based solution; which is the right approach in how you solve human based problems.
It gives the person with a human based problem another human to talk to. That in itself forces a change. As for the concept that the other person needs to be able to win a fight. Extremely unlikely there would be a fight. If it even looked like a fight might develop all the 2nd person has to do is open the door and if needed scream for help. I assure you that there are people in the front of the aircraft who would jump to help.
The two people in the cockpit solution is a procedural solution not a pure hearted noble "human solution" anything that involves a human would qualify door override included since it involves the trust in the entire human crew after all.
It is a flawed solution because it knows that one of the two could be the attacker but just hopes no one person would attack the other but if a person is willing to kill 149 people at minimum knocking out a person is not unheard of. The only cases it will stop are borderline pure moment of opportunity cases, the moment any degree of planning occurs then it breaks instantly. In the GW case if the captain did not goto the wash room as the FO suggested it would have been very likely the FO would have "gone to the wash room" and then upon return knock the Captain out cold, under 2 person rule the second person would have already left and the door closed and automatically locked. From what was stated this crash was premeditated, how much planning he did in his head or somewhere yet uncovered is not known.
|Quoting 2175301 (Reply 134):|
In the end there are no perfect solutions for every problem. Humans are ultimately fallible and so are technical systems and parts. But, I suspect that just having a 2nd person in the cockpit eliminates at least 80% of these kinds of crashes; without creating any other risks (and your technical solution has other risks that have to be accounted for).
By the way; I am an engineer who works in the nuclear power industry. The biggest problem in nuclear power is human based and not technical. Just like aircraft there have been many decades of technical improvements. Most of the biggest events in the last 40 years have been traced to human factors (even if the decisions were made decades earlier not to design for a known on average 90 year event); and the biggest gains in event response have come from dealing with and addressing human factors.
Throwing random percentages when it is obvious any malicious actor who pre-planned the act would just ignore random number and break the 2 person solution in any case. If you want random numbers the 2 person solution at best would stop 1% of attacks which have zero planning and occur as a moment of insanity of which the second person could snap the other out of it.
You are open to say what these "risks" are for the updated door logic. I openly say a 2 person system doesn't solve the element of surprise issue nor the pre-meditated act issue it at best helps people feel better but doesn't stop the problem from happening again.
No the biggest problem in nuclear power is that you can just make a reactor that cools itself without any intervention. (Passive safety) Human factors are part of the design process a human error can be amplified or enabled through a mistake in design sometimes even just the user interface. (In this case the door)
Unfortunately for humans and planes you can't just passively let it figure out what to do as it isn't as simple as perfectly random nuclear decay generating heat which needs to go somewhere.
The human brain is many orders of magnitude more complex than any nuclear reactor. (A nuclear reactor is just a giant kettle albeit much more complex than the one in your kitchen) And a plane can't just sit underground as that wouldn't be very useful. (While a nuclear reactor could just sit around generating heat, heck you can have natural nuclear reactors no human no technology even needed that is how simple it really is)
|Quoting morsecoder (Reply 146):|
The IP addresses of websites you visit are usually stored in your computer log files and router memory as well. Sometimes on a rolling basis (i.e. some number of most recent sites) or sometimes until the logs are deleted. Your ISP will also have a record of the connections made to your IP address. In the U.S., I think the standard is to keep that data for 90 days. I don't know how it works in Germany, but here, if the police have a search warrant from a judge, it's all accessible.
More important details would be held by the big data folks Google, Facebook, Bing, ... will store all the website you visit, the search terms you enter, the referral headers, where you accessed it from, your account information, targeted ad profile, for technically forever (after a while it gets mixed into the great big data pool for the automatic algorithms to munch on) Once you have his ISP
account info and get a court order every company he touched is just going to turn over the entire data trove of information they have. (Many websites, ads, social media links, comment middleware also keep track of your website activities down to the click/scroll/dwell level to estimate gaze and focus on parts of the webpage) This information is much more fine grained than just what sites he visited or frequented. You can build a persons online persona(s) just by looking at those records. (AOL learned that the hard way when they released "anonymized search inputs" which was quickly broken when people figured out the identities of a few in the dataset)
An online advertising company has a much better picture of your life than say any government or ISP
because they are paid to create targeted ads specific to your tastes/searches/habits which form an automatically generated advertising profile.
|Quoting flyingturtle (Reply 148):|
In Germany, the provider can only store data that is necessary for technical troubleshooting, and to calculate and send out the bills.
Google, doubleclick (the ad company google bought long ago), and websites mostly operate outside of Germany so they still collect information like crazy. His IP history is probably already with them as well (even if he connected with from a public hotspot)
|Quoting AirlineCritic (Reply 150):|
First, the important thing. There are so many ways in which pilot can cause structural problems, unrecoverable flight envelope, quick attitude changes near the ground, etc that I doubt the door designs themselves will have an effect. If you can't sit idle pondering what you have done for 8 minutes or 7 hours... you will perhaps crash the plane on take-off. Or hit the other pilot in the head when he is not suspecting it. And many other things.
We can improve the system, but we are close to as good as it is going to get. Without an ability to see inside people's heads... you will be relying on the skills and professionalism of the pilots and the organisation that hired them. Just the way it is.
P.S. I agree that specific new door designs deserve their own thread.
I don't think I've been posting specifics in reply to any of these calls for the specifics to go elsewhere. The design has been fleshed out and now it is just relevant replies to people saying they don't think anything could be done.
First the important thing is this GW case did not end with an in flight breakup nor did it happen at takeoff or landing. The door concept is designed to make malicious acts harder and give them no time to act other than instantly. This puts a big dampener on planned attacks with specific timings and targets. I would rather the only plan be to instantly crash on near the ground which is at least remotely survivable than allow for hours and minutes of flight towards some other target of choice with means certain death with long foreknowledge of all on board of impending doom and total helplessness.
Allowing for the possibility of specific timings and targets without any risk of the plane being retaken is allowing for another 9/11 at worst and at best having military pilots shoot a passenger jet out of the sky. (Not something I find particularly ideal or even remotely close to good enough)
We don't need to see into people's individual heads is the entire point there isn't any realistic way to do so and it is totally invasive even if you could do it. We rely on the skills and professionalism of the crew and company as a group just the way it always was.
|Quoting namezero111111 (Reply 153):|
This accident started long before the caption got locked out of the cockpit. While I think the 2 people in the cockpit at all times is a sound idea for reasons other than suicide, I'd be very surprised to see door design changes resulting from this accident given what we know now. This accident chain is more easily and effectively broken beforehand.
Changes to medical certification / psych tests are more likely - possibly anonymous counseling programs for depression. (not sure if LH or other airlines have that for depression; they normally for for alcoholism).
Another possibility would be that a psych test is mandatory should a crew member be "reported" by other crew members for erratic behavior, say, 3 times in a year. This lends itself to abuse though, and hence may be a bad idea to begin with.
Enhanced medical testing won't fix anything, people will just hide symptoms and lie in an interview or eval you just can't force people to tell the truth nor can you detect lies with certainty. This accident started long ago and was enabled, amplified by the door not opening for the captain or cabin crew.
Why would anyone go to an anonymous counselling program with the simulations desire for increased doctor reporting. Regular clinic visits are already confidential and anonymous counselling in that way already in place.
Psych tests cannot detect someone who is planning an attack and wishes to hid those plans, they can't even tell if a person is lying about whatever they are saying. And a three strikes your out system means instant abuse as you recognize yourself. (Also no other pilots or co-workers actually thought he was unstable and he had no reported problem from the existing reporting structure that allows for anonymous reports on other members of the company, hence this safety net already failed because people are not so simple)
|Quoting BackSeater (Reply 154):|
solution whereby crew members would vote with their permanently assigned secret PINs for (or against with a bogus entry) overriding the lock
I would strongly advise against any static pin code which crew know in their memory as there would be cases where the pin code leaks or the database you use to load the pin pad with the unique pins gets compromised and updating the per crew code list would create logistical nightmares. The existing door lock-down covers the pass code database leak by allowing for forced lock-down so in the crew override case you have to make sure the pins can't leak out. My concept covers pin protection with an alternate but as effective means.
|Quoting A332DTW (Reply 161):|
Granted breaks are usually no onger than a couple months, but the general premise that breaks in trainig should raise red flags I think is overblown.
It is beyond overblown it is probably dangerous to even suggest. Saying any deviation from the norm is a red flag is not going to help at all. Are we going to have pilots refuse to take a break and just double down on their training rather than risk being perceived as "unstable" (stress, burnout risk would sky-rocket). Even having a bad dream or not being happy with your company or work should not cause people to report them as unfit then they will just stop talking to people all together and bottle it all up (we all know where that is going to end). What this case shows is that detecting psychological instabilities is inherently a very difficult task in which the measurement/process itself (the testing, rules, regulations) can alter the outcome in any number of ways some of which may be good, neutral, bad. There is no easy fix on the mental health side of things because humans don't work like a two state system of off/on, locked/unlocked, sane/insane.
[Edited 2015-03-29 15:37:14]