“A Design Epiphany: Keep It Simple” is a New York Times article getting some play on the blogs. Given my new drum to beat, it won’t surprise you that I think this notion of “simplicity” plays right into what I said about “managing expectations” and making your designs explicit. A simple design will meet a user’s expectations, because they’ll be able to grasp what it does.
There’s a spin on this, though, that’s kind of alluded to, but not really addressed in the article. It’s the paraphrased Einstein quote, “”Make everything as simple as possible, but not simpler.” What is not acknowledged in this article is what to do when something is, well, complex. A common error is to try to “simplify” it by reducing the elements for interaction — saying having a single button serve more than one function, depending on modes. So while the interface might look simple, actually using the damn thing can be a pain (I’m thinking of radio alarm clocks right now).
This came up at the 2003 IA Summit, when Mark Bernstein reminded us all that “Multivalence is Not A Vice.” This is a challenge that information architects have that interaction designers and product designers don’t really have to grapple with. Information, in the form of words and messages, have an inherent complexity, because they have an inherent ambiguity — they can mean different things in different contexts to different people. There’s rarely, if ever, a one-to-one mapping. We have to deal with the fact that a variety of people will approach and use a single piece of content in a multitude of ways, and that that content will, in turn, spur a multitude of responses on the part of the reader. (Off the top of my head, think of a recipe. Typically it’s “used” as a set of instructions for cooking, but it can also be used as a guide for making a grocery list, or a communication tool to suggest a meal to someone else. Right there, that one recipe affords a wider variety of uses, of interactions, then, say, the Rabbit Corkscrew.)
I don’t know where this tangent has taken me. But I’m going to stop now.
For better or for worse, interaction design has been distilled to “keep it simple”. But that’s not its goal. The goal is to create interfaces and interactions that support target users’ goals. To create interfaces and interactions that map well to the way target users think about accomplishing certain tasks. If that is done well – the result may be an “elegant” and “simple” design. “Simple” is in the eye of the beholder.
I’m on a project now where part of my duty is to constantly nag about simplicity. But I’ve started to think that isn’t quite the right word. The nest of information structures, user goals, and corporate goals we’re dealing with is complicated. Making something simple would be a disservice.
What we’re really after is what programmers and engineers usually mean when they say “elegance.” We’re after the smallest number of features and interface elements than can make the complexity accessible. This sets us on a quest to invent rules of interaction that have broad reach and power. So, for example:
Foreground / background
there’s objects you can manipulate in the foreground, and zones of organization in the background.
objects that can be picked up respond to rollover by indicating that they’re ready to go; they’re not glued down.
when an object is picked up, zones that are willing to receive that object indicate their willingness to do so.
All objects and zones provide a control which exposes the several ways the user could choose to view them right now.
And so on. What we’re hoping is that in the end the interface will be discoverable and memorable, so people will wind up saying, “it’s simple!” But there will certainly be scores of features and views. It won’t be simple. We’re aiming for elegant.
All that said, I’ll probably still keep preaching the term “simple” around here, because it somehow people on the team respond to it better than when I say “elegant.”
This is the long way of agreeing with you and Einstein.
The last comment was certainly SIMPLE but in no way elegant. Just doing my bit to get it off of Peter’s main page.