My previous home computer was an IBM laptop. It was an X-series ThinkPad, which means it was so small it didn’t have room for a trackpad. Instead, the mouse was controlled with a pointing stick in the middle of the keyboard: a mini joystick, which I would call a “nipple” to make people borrowing my laptop feel uncomfortable. I loved that pointing stick.

After Lenovo bought IBM’s computer line, the quality decreased. Dirt-cheap netbooks because the hot new thing. So when my old IBM died, I replaced it with a 12″ netbook (just before that market segment disappeared) with a trackpad.

When I use the trackpad to select text, I often miss the first character of the word. I decided that trackpads just suck at selecting. It turns out that I misunderstood one of the basic metaphors of Windows, Icons, Mice and Pointers (WIMP) user interfaces:

I’ve been selecting text by clicking on the first character I want to select. It turns out that graphical user interfaces are designed to select the gaps between characters.

When the mouse moves over text, the cursor changes into a vertical I-beam. The I-beam represents a movable vertical-line text cursor, like in a word processor. So to select text, you click to drop the vertical-line text cursor between two characters and then drag to another text cursor position.

I figure my misunderstanding came either because:

• I don’t pay attention to the cursor image and just habitually click on things that I want no matter what the image is
• I sometimes type at a command prompt where the text cursor is an underscore that takes up a full character

Now that I realize this, my selecting has improved. I accept that humans need to meet computer intefaces half way, so this isn’t a UI failure, it’s my lack of knowledge. But I can’t help but wonder who else has this problem and whether user interfaces should be designed the opposite way?

Neo-classical economics assumes that people are rational agents: given a choice, they will always choose the option that maximizes utility (wealth, happiness, etc). And yet, as with most things in economics, actual observation shows this to be a poor model for reality. Instead of rationality, it turns out that “satisficing” is an excellent model: given a choice, people will choose an option that does a “good enough” job of maximizing utility.

I like to think of satisficing in terms of computational theory: people make choices using an algorithm that yields increasingly accurate answers the more computational resources are devoted to it; but the optimal answer would require huge resources, so a “good enough” answer must suffice.

I believe we not only satisfice when we’re making decisions but also when we implement them. The concept is closely related to “settling”. We constantly decide that what we have readily available is good enough, whether it’s paint colour or a mate.

Life has sparse survival decisions and many inconsequential decisions between essentially interchangable choices. Most of the interesting decisions in life are aesthetic choices and for these we consistently satisfice.

I’ve been wondering lately if satisficing increases stress? Is it frustrating to know that greater effort could achieve a more optimal answer? Do our brains’ decision processes dislike being halted before they complete? Life is a series of aesthetic choices and it pains us to make each one = life is suffering, the First Noble Truth.