Today’s NY Times has a familiar “The-Future-Is-Here-But-With-Caveats” pop-sci article on the Predator drone fleet: Drones Are Weapons of Choice in Fighting Qaeda. A new point is that the military seems to be moving more towards quick-and-dirty ad-hoc solutions that work mostly as expected, as opposed to the expensive and protracted perfectionist way.
As is common in time-cutting and cost-cutting, good UI design in this case is the first to go:
Pilots, who fly them from trailers halfway around the world using joysticks and computer screens, say some of the controls are clunky. For example, the missile-firing button sits dangerously close to the switch that shuts off the plane’s engines.
… [T]he service plans to eventually shift to simpler and more intuitive ground systems that could allow one remote pilot to control several drones. Now, pilots say, it takes up to 17 steps — including entering data into pull-down windows — to fire a missile.
It appears that we still haven’t learned that the most brilliant technical solution rarely, if ever, translates directly into the user-centric domain without considering the design issues. It’s all too easy for an engineer to drop in a Yet Another Gray Button somewhere on the screen that calls a function, and consider the task complete.
It’s interesting to note that the problem with the engine kill switch is mirroring a classic aircraft safety case back from the nineties. This time, though, the switch is close to the missile-firing button. That can’t be good, can it? Not only would you not want to kill your engines instead of firing, but if you’re in a situation where you’re about to kill your engines because of a malfunction, the last thing you want is to accidentally fire a missile.