Ejection Decision (2016) | Hacker News Reader

Ejection Decision (2016)

verticalmag.com

89 pointsTomte14 comments

> More than one pilot being interviewed stated that his decision to eject wasn’t made in the heat of the moment. The decision was made years before, in training, after careful thought. “If I encounter these conditions, I will take this action.” These guys made their decision early. When they encountered the conditions they acted. Without thought. Without hesitation.

This is also good advice for making ethical decisions, especially when situations or people in power suggest otherwise.

Not the first time I've heard somebody say this.

If you want to do the right thing under severe temptation to do otherwise (at least consistently) you you need to have made that decision in advance, otherwise the situation will overwhelm your ability to process it and make good decisions. If you don't you're effectively leaving the outcome to chance, or perhaps whether you ate a good breakfast that morning.

Also helps when dealing with difficult/angry/argumentative people, or when you're arriving at work into what you know is a difficult situation.

A few examples would be illustrative. It's unclear how immutable beliefs about moral judgements are a good thing. That seems like the basis of prejudice, for example.

Of course it can be the basis of prejudice. The only difference between “I will not kill except in self defense no matter how angry I am” and “I will not serve any black people no matter how desperate for cash I am” is the goodness of the choice being made.

It’s a tool. Like any tool, you can use it for good or for evil.

Decisions like "I will not sleep with a woman who is too drunk to climb a flight of stairs unaided" and "I will run rather than using this utility knife in a fistfight" are much easier to make objectively when you're calm and sober, rather than waiting until your blood stream is full of alcohol, adrenaline and testosterone.

Yeah this is basically the gist of it. In the heat of moment, at the apex of a moral dilemma, it is likely that your body will have a lot of things operating upon it, and a large cognitive load to track just to keep tabs on the environment.

If you mix in the obligation to make an important moral decision, especially one that sacrifices evolutionary benefit for abstract social benefit like "I will not engage in a violent reaction when I feel offended" or "I will not copulate with this apparently willing and attractive potential sexual partner", well, we could say that it's likely that /usr/bin/moral_override will time out and your body will default to the path of least resistance for itself in that moment.

Thus, rehearsing for good outcomes in high-pressure situations is like tuning a configuration file to ensure reasonable behavior in known potential failure modes. You can hope that you won't hit the failure modes described (or you can plan ahead, knowing you eventually will, which is the better option), but you want to make sure that your system will behave appropriately when those modes are encountered. Leaving the behavior undefined or trusting the defaults is too risky for anything important.

I'm not the OP and don't have any solid examples off the top of my head, but I'm going to riff a little bit while I process the concept myself.

It seems to me that by explicitly writing down the decision tree and reflecting on it, you could attempt to identify whether or not there were prejudices baked into it.

As a contrived example based on the article and comments: should the fact that the patient is a 6-year-old girl vs. an overdosing opiate addict make any difference on whether or not the conditions you're flying in are safe? Or should it be based on visibility thresholds etc.

In most combat sports such as MMA you avoid training illegal moves, like biting, gouging, kicking a downed opponent, Etc. Under pressure you resort to what you’ve trained to do, and if you’re training for a street fight then you’re going to do something disqualifying.

You could possibly reinforce your biases and prejudices by "preprocessing" on the fly value judgements. On the other hand, you will never confront your biases and prejudices if you never attempt to reflect on then and what you should do in various situations. They will only become apparent to you after you have already made the biased or prejudiced decision, in the heat of the moment, without the luxury of time consuming introspection.

Train like you fight, fight like you train. That is more or less the doctrinal position of modern militaries, and it works. From sillouette target practice, to live fire drills under stress, and yes, ejection from a doomed aircraft.

HEMS is one of the few areas of emergency services that is _actually_ a dangerous line of work. It's on par with commercial fishing and logging in terms of fatalities per 100,000 workers (as opposed to "regular" fire/EMS/law enforcement jobs, which are safer than being a traveling salesman). I'd love to do flight EMS, but it's not a risk I feel I can responsibly take with a wife and kids (to put that into perspective, "crawling into burning buildings" falls on the other side of that line for me)

A large majority of the fatal incidents are "scene" calls (as opposed to inter-facility transfers), and stem from pushing past the edge of the envelope like this article describes. There is also some sad irony in the fact that the patients that are most likely to induce this level of pressure (those closest to death) are also ones that are least likely to benefit from the additional risk taken (i.e. they're gonna die anyway).

Around here the majority of scene medevacs are flown by the county sheriff's office. All of their pilots are former military aviators, and are very, very good at flying. I've watched them put helicopters places I wouldn't fly my quadcopter for fear of losing it... Their skills have definitely made the difference between life and death for patients I've handed over to their flight crews. I have a lot of respect for what they do, but I also worry for my friends that fly...

The training video linked off the article is also worth a watch: https://www.youtube.com/watch?v=Aa1Ba_NEobs

This reminded me how fascinated I am by disaster reports.

Into Thin Air got be fascinated with mountaineering, but the next few books I got were all accident reports.

We have a similar thing at work: if we're deploying, and start getting paged for error rates, the default action is "roll back!" We explicitly don't try to figure it out before we roll back or try some heroics to fix it or whatever. Just roll back!

Reply on news.ycombinator.com