Poor UI Design Can Kill – Air Inter Flight 148, a Harsh Lesson Learned

Part of the vim cult's belief is the Never-Leave-Home-Row. That is, where you first rest most of your fingers on the asdfjkl; keys. Your hands are 99% of the time if you're not actively typing in the document. If Vim didn't have modes then you'd have to drop this concept. But of course, mistakes in vim don't normally result in dozens of people dying. If something happens that I didn't expect I just glance to the lower-left corner of my terminal to see what mode I'm in.

A common path to fluency with vi is to live in command mode. When you type something, immediately press escape out of habit. When you want to type something new, develop a habit of pressing a/i/o with confidence that you're already in cmd mode.

If you find yourself navigating in insert mode, that's a red flag.

But you also neatly capture the reason for modes: they provide more capabilities and make a tool more useful. Again the main UI/UX problem isn't that there are modes -- we deal with modes all the time: inadequate vs outside voice, kitchen vs. bathroom activities, behavior with your BFFs vs. your boss or a customer, stadium vs. museum. It's just that these are so totally and fully integrated into our experiences that we don't often think of them as modes. But they are. And the behaviours are learned.

The trick in UI is to make the distinctions so obvious that the user responds like this. Behavior in a given mode is obvious, using the wrong behaviour a glaring error.

I´m currently flying A-320, I´ve also flown B 737(300 and 800), and MD88. It´s true that at the beginning the new airbus received a lot of heat from pilots. After all they were trying to reinvent cockpits UI from 0. This led to a series of design accidents (due to initial design mistakes ) like the one on the article.

The initial idea it´s not bad: reducing weight (removing all mechanical and hydraulic controls), inventing the glass cockpit, simplified system management, etc..

It took over a decade to polish the cockpit (and the whole airplane, that had tons of electronics related failures all around) to it´s actual safe and workable state. Those issues, plus the idea that pilots had to readjust your whole idea of a cockpit, were the ones that created that opposition. The opossition it's almost gone now, although in general we pilots still prefer Boeing aircraft (at least in Europe). Airbus airplanes fly very nicely and are great too.

But now that we are used to the new IU, the concept problems are still there.

Main diference between flying normal fly controls and a moving auto-throttle and airbus fly-by-wire:

-Airbus throttle levers don´t move while the auto-thrust is changing the power setting. All the other airplanes as far as I know do it.

-Airbus side-sticks don´t move when the autopilot or the other pilot is giving an input. Also as somebody explained above, the inputs are added, you can end with a double (or zero input), if you forget to push the priority button (in stress situations like last second corrections to land it´s easy to forget it).

Why this is a mistake? because when flying a modern airliner you have an excess of information to process, all at once. A moving auto-throttle let´s you know what the engines are doing through your hands, without having to look at a tiny indicator at the instruments. Also it let´s you override the auto-throttle briefly because you need more power or less power at a given moment, without having to disconnect it. With airbus auto-thrust either the airplane or the pilot has the control, and you have to manually takeover. This may be dangerous at an approach that it´s already hard enough with cross winds or heavy rain.

The same apply to the side-stick. Normal autopilots move the fly controls at the cockpit while actuating them. This let´s you know what the autopilot it´s doing. And of course when a pilot it´s flying manually, the other pilot knows exactly what he is doing ALL THE TIME. He can slightly correct a maneuver without having to take over the controls from zero.

Airbus has wiped a layer of fundamental information available to the pilots, increasing the information overoosd through the vision. It´s the equivalent of driving a race car with a playstation control. Most of the time this information it´s not used(like in cruise), but you really, REALLY need it when things become hairy close to the ground. In fact I routinely disconnect the auto-thrust to land, with the Boeing and MD88 I kept it engaged.

Airbus knows this, but it´s too late to change 20 years of airplane development. They´ll need to change too many things (systems, procedures, certification, instruction), also right now it´s easy to change from flying let´s say a A320 to a A 330, due to having very similar cockpits and systems (that´s a great idea). The conversion course it´s short and relatively cheap. Changing the controls would make this fleet changes much more expensive and time consuming to Airlines.

Edit: overall small changes to improve clarity.

Stalling and flying into the ground have been two common failure modes since flying began and they often happen in sequence. I would think they would be high in the list of misuse cases any aircraft designer would consider.

"we should refrain from being quick to judge which is more or less intuitive."

We should not be quick to judge which UI is more effective, but by definition we can be quick to judge which is more intuitive. In fact, the more time you spend with a system, the less qualified you are to comment on whether it is intuitive. Judging whether a UI is intuitive should take very little time and no expertise past understanding basic operations. I'm certainly not qualified to judge, but I'd think anyone who has experience with modern flight systems would be qualified to comment on whether a particular design was intuitive.

And that is an obvious problem. It's literally one of the first things you cover when you begin learning how to fly. The instructor will sit you down and walk through how you hand off the controls and agree who's flying the plane at any given time. Likewise, when flying with another pilot you always do a positive handoff. "You have the plane." "I have the plane." The physical link between the two sets of controls is an important backup mechanism that ensures this happens.

Not linking the two sets of controls is insane. It's unrelated to the different approaches to fly-by-wire. Taking two radically different control inputs and simply averaging them together makes no sense.

The caps lock key does a very similar thing to shift, and is not very useful on a modern keyboard. In my experience it is more often engaged by accident than on purpose.

To restate my analogy: the problem is when you can run the same function twice with the same input and get a different result. Or when the same variable can mean two totally different things depending on some external context.