No, UI is the New UI

Today I stumbled across “No UI is the New UI,” an article extolling the upcoming demise of the traditional Graphical User Interface, in favour of text messaging interfaces like Facebook M and “Magic.”

Tony Aubé writes:

The rise in popularity of these apps recently brought me to a startling observation : advances in technology, especially in AI, are increasingly making traditional UI irrelevant. As much as I dislike it, I now believe that technology progress will eventually make UI a tool of the past, something no longer essential for Human-Computer interaction. And that is a good thing.

While I think conversational interfaces, whether powered by natural or artificial intelligence, have a long and prosperous life ahead of them, I don’t think they should replace traditional interfaces in most cases.

Conversation is fantastic for certain tasks, like requesting or negotiating certain broad information. Negotiating with software about which restaurant you’d like delivery from is probably much nicer than filling out a web form, but you can do so much more with a computer.

(Also, have you noticed how many of these conversational UI products in North America are really really first-world-problemy? Arrange my flights, send me food, clean my house. Oof!)

He continues,

As a designer, this is an unsettling trend to internalize. In a world where computer can see, listen, talk, understand and reply to you, what is the purpose of a user interface? Why bother designing an app to manage your bank account when you could just talk to it directly?

I’ll tell you what the purpose of a user interface is: it’s to provide much richer information, most of it visually, and to allow for deeper interaction with the thing you’re trying to understand. Let’s look at both of these:

The most important part about the Graphical User Interface is that it’s graphical. The eye is crazy fast at soaking up information. The eye can see shapes and colours, can determine hierarchies of importance and can compare choices like nobody’s business. We also have a type of work dedicated to they eye that’s been worked on and studied for centuries, called Graphic Design.

For spoken-word conversational UIs, you get one morsel of information after another. You can’t go back, you can’t make comparisons, you’ve got to remember it all. Everything must be described. From this mode, “huge” and “tiny” are arbitrary sounds whose meanings don’t seem as different as they are meant to be.

For written-word conversation UIs, the information is a little bit spread out, you can technically read backwards, but you’re still left with arbitrary symbols (we know them as letters and numbers) trying to relay information.

That’s the first part, the visual richness of graphics. The second part is how we humans interact back with the computer. In the graphical user interface, we have pointing devices (mice, fingers, pencils, etc) for indicating our interest and for exploring information. This allows us to directly manipulate the thing we care about. We can point at them, select them, move them, apply things to them. The list goes on. This lets us arbitrarily manipulate things in space, whereas with conversational UIs we end up needing to manipulate things in time, like we do word after word and sentence after sentence.

The overarching theme here isn’t that graphics are better than text, or buttons are better than SMS, it’s that these different interfaces force us to think in very different ways.

We change with our tools, with the media we use for communication. We communicate with one another, yes, but we also communicate with ourselves. The representations we choose help us think. The graphical user interface as we know it today is far from perfect. It will continue to change, but graphics are here to stay as an important part of our society.

See also Bret Victor’s lil post about graphics and computers.

Speed of Light