On Saturday while out in London, I made two bathroom visits during the course of the day. Here’s what happened.
In the first bathroom, after faffing around trying to get the tap at the sink to work, the guy at the next sink realised there was an automatic sensor above and to the left of his tap… it was hardly obvious.
“Why do they have to make these things so annoying?” I said. He just laughed, and went to dry his hands. Then he said “you should be OK with drying your hands – they’re just paper towels, it’s easy.”
Haha… like I wouldn’t have figured out the tap eventually. Stupid evil tap.
In the second bathroom later on that day, still scarred from the first visit, I flapped my arms around the tap for a good few seconds trying to get this second tap to work. Then I noticed that it had a handle on it :-(
Did the day get better?
That evening, I went to a restaurant, and the front entrance was through double glass doors.
I tried to push the right door. Nope.
Then I tried to pull it. Nope. It wasn’t going to budge.
I then tried to push the left door hopelessly. Nope.
Pull the left door? Hallelujah! I’d figured out how to enter a restaurant! Only my fourth try!
The observant among you will note that if I’d been on the other side of the doors, I’d have been right first time :-)
Anyway, although that’s a lovely story, do I have a point?
Well. Why are simple things that we use every day so needlessly difficult and unintuitive? Is somebody playing a trick?
There’s a book that talks about this exact topic. It’s called The Design of Everyday Things, by Don Norman. It was written quite a while ago, but given my experience on Saturday, no one has listened.
Norman, a psychologist, talks about how users tend to blame themselves when they can’t figure out how to use a given instrument. You know what they say… anger leads to hate etc.
Humans make errors. It’s a sad fact of life, but true. Some errors are made after conscious thought, and these are called mistakes.
Another kind of error is a slip, which is an unconscious error. How many times have you accidentally walked out of a room to do something only to forget what it was as soon as you left the room? Or put your laundry in the bin rather than the clothes hamper? Or driven to work when you meant to drive to the gym?
Humans have been blamed for a lot of bad things that have happened… accidents, disasters, some even of the nuclear variety. But quite often, these could have been avoided if the system had been better designed to be easier to use and harder to make mistakes.
Norman explains that when we look at an object, we immediately start getting a sense of what we can do with it. Users develop what is called a user model, and covers aspects like:
- affordances: if it has a handle or buttons, we know it can probably be picked up or pressed
- constraints: these might be physical constraints preventing certain actions, or we might discover other types of constraints such as logical (we wouldn’t want to do that… that would break it) or even cultural (we were taught never to do that!)
- our own personal knowledge of how it’s likely to work based on our past experiences.
Now humans are pretty smart compared with other animals like dogs or chimps. We’ve got hands, which means we can grasp objects and do things with them like twisting, bending, shaping etc with varying force. So objects can have a lot more affordances for man, than merely being something to lick or bark at or pee on. Chimps have hands too, but aren’t as smart as us so struggle when it comes to building models of how objects can be used.
As smart as we are, we’re still only human. So a key goal when designing something well is to design it in such a way so that the user is able to build a good user model, and use it to predict the effects of actions. There should be a natural map between actions and expected results. This can even remove the need for lengthy instructions for how something works.
- The design of a steering wheel is such that turning it left or right by varying amounts causes the car to turn in the same direction by a proportional amount. A steering wheel has a natural mapping between action and result. It’s very easy to learn how to steer a car, even some dumb kid could do it.
- A typical office phone might only have about 15 buttons, but these can combine to give hundreds of non-obvious commands. How are you supposed to know that the command for putting someone on hold is e.g. #97**? It’s not obvious. It has to be learned and memorised. It requires knowledge in the brain rather than knowledge in the world.
Having figured out how to carry out an action, what happens when that action is actually carried out? Well some feedback would be nice. It would be nice to know that the action had the intended effect. With a car steering wheel, it’d be obvious from the car actually turning in that direction. But the Scary Evil Phone is potentially even more annoying than Sheila. It doesn’t even have a screen to display feedback – you have to try and make sense of any sounds it sends down the earpiece.
Design for ease of use, and make it harder to make mistakes and do the wrong things. Nobody needs to feel stupid.
This is especially true for computer software. Because even though computers are our most complex and powerful tools, meaning that there’s a lot of potential for things to go wrong, we have the power to program a computer to follow any behaviour that we specify. So we can choose to make systems that are easy to use and provide excellent feedback. And we can do this with good design.