When things work, until they don’t

So I quit my job a little more than a month ago. I did cancer research, and I decided to go into user experience design. A number of people, when I told them, kind of scoffed at the idea and probably thought that I had to turn in my hippie idealist card. Design? Isn’t that just making something pretty? How is that going to help people, besides making Apple fans even more insufferable?

There’s actually way more to design than that. But the hard part is, good design is invisible. It just works, and you forget about it 10 seconds later, going on with the rest of your day. But bad design can frustrate you, it can stop you from doing something that you NEED to do, and it can even be deadly.

So what is design? Wikipedia has a fairly straightforward definition, which I will paraphrase: it is a plan for someone to achieve a unique expectation, that considers context, cost, environment, and safety. So far, already beyond making something pretty! What I’m interested in, is the human aspect. How to design something to be easier to use.

Let’s take a door. People can look at a door and know, right away, what makes a door a door instead of a window or an opening for a car or a giant hole in the wall. We walk through countless numbers of doors a day.

How many of you have pushed on a door that could only be pulled? Or walked into a door entirely? How many have blamed yourself for doing that? Guess what? It’s not your fault – it’s the door’s! Or rather, the people who designed that door.

Humans are really good at establishing patterns. You walk through one or two doors of different types, you understand how it’s supposed to work. It’s almost unconscious. But a designer has to know what those patterns are to have it built correctly.

Here’s a list of patterns for doors: You pull to enter, you push to exit. Bar or plate means you push. Handle means you pull. If all else fails, we can see which side the door hinge is on and figure it out from there. It all works…until it doesn’t. Break any of those patterns, and people don’t know what to do with your door.

And for Pete’s sake, don’t make a window the size of a door. I went to a fancy lounge the other day where the bar area was surrounded by door-sized windows. The staff had taped a warning to one such window, on the end, where people had undoubtedly tried to walk through: “Glass door. Be careful!” Gee, thanks, a lounge with a danger of human bird strikes.

So walking into a door kinda sucks. But it’s still just an inconvenience. It can get worse. Those of you keeping up with the news for the past week or so will be aware of my next example: healthcare.gov

Countless numbers of people could not use the site, and I’d bet a good number of those people NEEDED to be able to use it. They needed to get health insurance. You just could not get to the website. So what happened?

Well, on the outside, it looks pretty nice. But ah, design also considers how it works on the inside, too. They made one particular critical error in how the whole thing works. It was so fundamental that even I, who’s been in the field for really only a few months, went, “Oh, you’ve got to be kidding me. I can’t believe this.”

You had to register for an account before you could browse plans.

First of all, would you shop at any store in the first place if they required you to leave an ID at the door? What would happen if say, Walmart tried to pull that off on Black Friday? That was the same thing.

Sure, it’s the internet, but people still don’t like giving you personal information before they absolutely need to, and bottlenecks can still happen. We don’t expect them to happen anymore because of all the huge popular website we use that DO work. Again, it works….until it doesn’t. Time will tell if they’ll be able to fix everything in time but…mmmm, we’ll see. [Note from 2015: Turns out they did all right!]

Healthcare.gov is something that is clearly someone else’s fault, but going back to the door example, there are other design issues that are often thought of as “our fault” that can be much more serious than walking into a door.

Most accidents today are chalked up to human error. “Oh, if so-and-so had just done this right, this wouldn’t have happened.” And so we go on with our lives, proud of ourselves that of course we wouldn’t make that same mistake. We’re not stupid.

Not so fast. Air France Flight 447 fell out of the sky over the Atlantic back in 2009. The Airbus involved had all the fancy gadgets that was supposed to make flying easier and safer. It worked…until it didn’t. The sensors iced over and the plane took itself out of autopilot, which it was supposed to do, but the pilots didn’t know which sensors were accurate and which ones weren’t.

Ultimately, it WAS human error that took the plane down – one of the pilots kept the plane climbing, which put it into an irreversible stall. But…only one of the pilots made that mistake. There were 2 other more experienced pilots with him. Why didn’t they stop him? They did not know what he was doing. The first pilot had kept his stick locked to a climbing angle, and in Airbuses, the two pilot sticks don’t physically match each other. In Boeing planes, they do. You can tell exactly what your copilot is doing in a Boeing, but not an Airbus.

Yes, there were other things they could have done to ultimately prevent it, but why burden pilots with more training for something that is ultimately a design flaw?

I don’t think we should just accept this. I don’t think that we should put all the responsibility on the user. Just what is the point of getting all this knowledge and developing all this technology if we can’t help each other be better?

So I encourage you all to design. Make something better. Not just pretty, but work better for you. Maybe you won’t get to design something as serious as an airplane, but we can plant the seed. And together, we can all get past accepting that things will work, until they don’t.


This was a originally a speech I gave at Crown City Toastmasters