A lot of incident reports have flight attendants or copilots leaving to try to make visual confirmation of things that it seems would be better suited if there were some actual visual feedback.

And presumably if the camera doesn't work as part of the pre-flight checklist for whatever reason, you're not going anywhere until it's fixed/replaced.

part of it is that there's a backup for most everything required.

The tower sees it with a set of binoculars. Hopefully.

I'm talking about things they can't see, which I've heard in recordings as well.

For something like this they'd gain something for sure - time and redundancy

You'd need potentially an unrealistic number of cameras. Even if you decide on a case by case basis, you still have to weigh the risk that every component adds, and with a finite amount of money to spend on risk reduction you want to get the most bang for your buck.

I'm operating under the belief that more time to respond to issues is always a plus. Do you disagree?

Good enough is good enough.

It's a cost, which is why I'm trying to understand what the ROI would be.

> Good enough is good enough.

In cases like this, yes, but ... planes do crash due to mechanical failure and it's not unusual for there to be confusion from instrumentation where visual confirmation could help.

To answer your earlier question, not to split hairs, but I disagree with "always". You can inadvertently increase risk when trying to reduce it, or unwittingly prioritize things that offer less bang for your buck. That's why risk analysis is important.

I want to be clear I'm agreeing with you that some cameras would be useful in general. I think my only disagreements are that eyeballs are fine for this scenario, and that "more" isn't always meaningfully useful or necessarily better.

Not that people shouldn’t help but you gotta be careful with maneuvers you don’t typically make, even if trying to be helpful.

This is how I got my commercial pilot's license, and I'm a senior software engineer.

It just "clicked" with my brain to follow specific procedures ("algorithms") for absolutely everything. Checklists, strict rules, "flows", handling emergencies. It all felt natural to me.

They will throw random stuff at you during checkrides. Pull the power back on one of the engines right after takeoff saying it failed, fail instruments that you were using to navigate, blindfold you ("foggles") and put you almost upside down and then say "recover!".

But you have everything so ingrained in your mind by that point that it's almost robotic. You just look at the inputs (almost upside down, engine #2 is gone, no attitude indicator, whatever it is) and know what the output is supposed to be (roll to unload Gs, lower the nose, full power, check the standby AI, etc).

I happen to work well when there are strict rules and procedures. If this, then that.

Like the situation OP says: you scan instruments to analyze the situation, determine what to do to recover and apply that procedure. If it doesn't work, or something else goes wrong during that procedure, you adapt to another procedure. Remember, that during every flight with an instructor, you're being trained on one thing or another, so after a while all the "emergencies" seem routine. You'll be turning onto Final to land and suddenly your instructor will decide that your flaps failed so you have to land without them, or just as you're flaring for a landing he'll tell you to go around, etc.

It certainly results in well-trained pilots, but it also gets very boring.

The immediate reason to stop was the fact that the more I did it, the more nauseous I got, but by then I had already decided that flying was not nearly as exciting as I expected it to be. Stopping was an easy decision to make.

That said, I also have friends who still totally love it.

In fact there's an exception that proves that rule: there was not an algorithm to handle pitot tube ice causing an MCAS failure on the 737 MAX, because the type certificate was shared with earlier aircraft that didn't have an MCAS to fail.

And two planes crashed because pilots didn't have an algorithm to follow to tell them what was happening when their trim went crazy.

I've been wondering for a while when I read stuff like this, why don't modern airplanes have exterior cameras that cover every surface? I watched a video recently about a flight where one of the engines literally detached and the pilots didn't know the extent of the damage until after. Why not just have cameras they can quickly pull up to check stuff like this?