Decrepit Old Fool

In my office there’s a clipboard that everyone uses.  It tends to get laid down all over the place, so time is wasted hunting for it.  I put up a hook on the outside divider wall of my cubicle and hung the clipboard on it.  While it is below eye level (it’s a low divider) it is still visible for the entire length of the room, and is the only object in about 12 square feet of featureless fabric, right next to the desk surface where most people set down the clipboard.  Yet at least two people have been unable to find it while it was hanging on the hook.

There’s more.  Time is often wasted looking for a pen to use writing on the clipboard.  So I tied a pen to the clipboard using bright pink string.  Today I handed the clipboard to someone, with the pen on the board’s clip.  I watched as the person looked at the clipboard, patted his shirt pocket looking for a pen (the bright pink string was touching his hand) and then studied my desk for several moments to find a pen and fill out the form which was directly under the pen.

This is an interesting problem.

All of us have had the experience of not seeing an object because it was not in its normal context, or was in a different orientation than expected.  It happens to me all the time – keys, hat, or a tool that I just set down moments before.  My grandmother had an expression; “If it was a snake, it would have bit you!”  I take the existence of that expression as evidence that this interesting problem is also a widespread, longstanding problem.

It’s easy to build a stereoscopic camera, but object recognition is a difficult computing problem.  It’s not even all that easy for our hunter-gatherer, pattern-seeking brains.  We might see the face of Mary in a cheese sandwich, but miss the eye-level sign that says in block letters: “Please use other door”.  And since both work and commerce depend on pattern recognition, this interesting, widespread, longstanding problem is also of considerable economic importance.

Consider the placement of fire extinguishers, the design of signs, icons, and door handles.  Visit the detergent aisle of your supermarket: it’s a chaos of bright colors where nothing stands out.  But it isn’t just the lack of contrast: even a splash of bright pink string might not be noticeable on twelve square feet of drab fabric.

Whether you advertise detergent, design websites, or write baking instructions, there are several shortcuts to handling attention, perception, and recognition.  The cheap, easy ones – bright colors, moving, blinking lights, and arrows, pretty much amount to tying on bright pink string, and they often don’t work.  That leaves more subtle solutions.

One is the creation of visual triggers.  It takes time to make a corporate logo that gets instant recognition, and changing that logo is playing with the company’s bottom line.  The same is true of safety icons.  How to make them culturally universal, or for that matter, mean the same thing to individuals in the same culture?  Not everyone has the same neurological makeup.

From Design

An effective shortcut in most cases is to include faces.  For deep evolutionary reasons, most people notice faces.  But take a look at any modern magazine and you’ll see that even that hard-wired shortcut can be overused.  As Scott McCloud says in Understanding Comics, greater degrees of abstraction makes for more universality… up to the point where the face is no longer recognizable as such.  But the abstraction has to go pretty far before that property is lost.

Another shortcut is conventional position.  You may invest energy finding something the first time, but cognitive parsimony installs a time-delay before that investment can be triggered again.  So you’ll start by looking where you found it before.  A set of controls or objects that are in a consistent place (as I hope the clipboard hook will become) will, over time, reliably save time.  This is the reason we expect web pages to have basic controls along the top and outer edges, with the logo hot-linked to the home page. 

Still another is extreme simplicity.  Google’s text-entry box in the middle of a nearly blank screen is like a clerk behind a desk when you enter a blank room where everything else is painted white.  The first – almost the only – thing you can do is walk up to the clerk and ask a question.  And yet, the clerk knows the answer to almost every question you could possibly ask.  All you have to do – as James Cromwell said to Will Smith in I, Robot, is ask the right question.  Other wise the answer may be as cluttered as the room is simple.

Still, the Google page is not nearly as blank as it appears on first glance.  I count twenty controls on it, of which two are drop-down menus.  Here Google is combining nearly every trick in the book to push usability about as far as it can go – conventions, position, and making the “website” do the work.  Because while the Google home page appears to be a website, it is really the business end of an inconceivably large and powerful computing system. 

The handling and exchange of information on the scale of the web is far too vast for (analogues to) bright pink string.  For one thing it is no longer possible to prioritize information.  What will be important?  About the best we can do is work on metadata and systems to make sense of it. 

What I wonder is: how far can Google’s model be explored for other sites?  Could an e-commerce site use it?  A college? Might all future competition for web usability be in the arena of intelligent response?  Could future web users come to expect it everywhere?  Might we become accustomed to querying our physical environment as we query glowing rectangles today? How will education be affected when information, once valuable, is in limitless supply (and therefore by definition, cheap) and the most important thing will be… the right question?