The difference between
Interaction design is a different kind of design
The brilliant author and observer, Scott Berkun (@berkun), asked (on Twitter) the provocative question:
What is the greatest design lesson you’ve ever learned?
There were many varied replies, and I finally weighed in with my answer:
The design of the behavior of digital systems is not at all the same kind of design as visual, print, architectural, industrial, or any other design discipline. There are many similarities, but they are qualitatively different.
This provoked Víctor David (@victordavidmx) to inquire of me:
Apart from the fact that we are dealing with humans and their complexity and not with inert matter, what would you say is the difference that makes a difference?
Here is my answer:
It’s not about the complexity of humans, and matter-based objects are often anything but inert. It’s about how humans perceive behavior.
Humans are non-deterministic. That is, the way any given human will act or react to the world cannot be predicted with any certainty. Physical objects on the other hand do behave predictably. Maybe not with perfect, uniform predictability but with enough that humans have evolved two different sets of cognitive frameworks for dealing with the world: our perception of, and our reaction to other people, and our perception and reactions to everything else that is not a human. I’m no psychologist but researchers know that humans who cannot differentiate between humans and other things are…special. I believe that autism is related to this anomaly.
If a coffee cup falls off of the table, we might be surprised, but if that cup begins to dance and sing our favorite song, we are gobsmacked. If a person jumps off of the table we are amused, and if they begin to dance and sing we are further amused. Just like Kahneman’s two modes of thinking, we have two modes of perceiving things in the world (at least two): our perception of human behavior and our perception of everything else. Of course, being human, all of the dividing lines between these two forms of cognition are wide and gray.
Physical objects do not have non-deterministic behavior. Rocks just lay there. Clouds move in random yet expected ways. With use, your toothpaste tube may deform in a pattern different from the way my tube deforms, but from a cognitive point of view, they deform in the same predictable way. And they don’t leap up and sing.
Mechanical devices are like that, too. A naval warship is a highly complex machine, but machines are physical objects whose reactions to outside stimulus are as predictable as the squeezing of a toothpaste tube. We comprehend naval warships by using the same part of our brain that we use to comprehend toothpaste.
Digital systems do not behave in a deterministic way. Now, all you rationalists will argue that digital computers are absolutely deterministic and you won’t be wrong. But from the point of view of human cognition, their behavior is sufficiently complex and their behavioral rules are sufficiently obscure that we comprehend them as non-deterministic. Generally, this happens at a low enough level in our thinking that we aren’t consciously aware of it.
The non-determinism that we don’t consciously perceive is subtle. If your computer simply did random things in response to your pressing keys on the keyboard, you’d just throw it away as a bad business. But it doesn’t escape your mind that when you are running Snapchat, the places that you touch the screen, and the reactions that result, are very different from the reactions you get from Instagram even when you touch the screen in the exact same place. Your mind registers this as non-deterministic. The part of your mind that is engaged when you use objects with digital behavior is different from the part of your mind that is engaged when you don’t. I believe that the human brain uses the same cognitive tools for perceiving digital behavior as it uses for perceiving the behavior of other humans. Stanford researchers Nass and Reeves refer to this as “the media equation.”
Design is in everything. It would be impossible to argue that the design of a physical object is any more or less difficult or demanding than the design of the behavior of a digital object, and I certainly do not want to demean or diminish the skill and experience required to create beautiful, desirable, and functional buildings, machines, clothing, images, and other physical artifacts. While I don’t wish to diminish, I very much do wish to differentiate between these two categories of design simply because the human mind comprehends them differently.
Of course, digital objects almost always have physical components, and — increasingly — physical objects have digital behavior, so, once again, the dividing lines are wide and gray.
Those people tasked with designing the behavior of digital systems must imagine how humans will act on and react to the behaviors that we create, and they will base their actions and reactions on cognition-for-non-deterministic-behavior. The word for that is interaction, and we design it. That’s why I use the term “interaction design” to describe the craft of designing digital behavior. Other designers, with other names, design for other parts of human cognition: how our fingers detect the difference between the various levers in an airplane cockpit, or how our aesthetic sense is stimulated by the flowing folds of a gown, or how our legs and our backs recline into a sofa, or how toothpaste behaves in a tube. These other design disciplines are just as demanding and difficult, but they are qualitatively different.
There is an enormous gulf between the nature of designing for the different ways that humans perceive their world. Even though we call them both “designers,” they perform qualitatively different tasks, with divergent tools, in disparate ways.
Years ago I wrote a book about this stuff and why it matters. It’s called The Inmates are Running the Asylum.
