Quantcast
Calendar Interface - Scott Adams' Blog

Calendar Interface

Okay, this has been bugging me forever. I’ve been trying to figure out why entering a calendar event on my smartphone or Google calendar or Outlook feels so…annoying?

It takes too many clicks. That’s one issue. But I always feel more annoyed than I would expect to feel from a few extra clicks. So I have a new hypothesis. Let me run it by you.

I think entering a calendar event is, for your brain, the equivalent of patting your head while rubbing your stomach at the same time. In both cases, the components are individually easy. They only become annoyingly when combined. Some thought processes don’t combine well.

By contrast, listening to music and washing dishes at the same time is easy. So is walking and chewing gum, at least for most people. Fortunately, some parts of the brain do work well together.

I started thinking along these lines when I realized I couldn’t drive a car and navigate too if my wife added the complexity of asking me to imagine a spatial task, such as how we might arrange the patio furniture. The spatial processing part of my brain would get overloaded and I would drive right past my turn.

Hypnotists take advantage of this effect too. Some inductions cause the subject to activate several brain regions at once. This technique quickly “exhausts” the brain and puts you in a frame of mind where you just want someone to tell you what to think.

Now back to calendars. When I think of months, I reflexively picture a circle, with January 1st at the top and June at the bottom. That uses the spatial processing part of my brain. When I think of a day within a month, I picture a wall calendar grid with four horizontal weeks. That’s a different spatial model. When I think of the time of day, I think of a round clock with two complete cycles for AM and PM. My smartphone unhelpfully adds another spatial model by making me enter times in a sort of slot machine interface with rolling windows, which causes me to imagine a tire shape, with the tire heading toward me. Meanwhile, the other options I need to click are spread around the screen and require a mental scavenger hunt, which is another spatial task.

Add to this spatial overload that my calendar likes to present itself sometimes in a month format, and other times by week. Worse yet, on some of my calendar interfaces the months scroll in a left-right orientation, and on other interfaces the months scroll up-down.

I also have to jump back and forth between my keyboard and my mouse. That’s another brain complication with spatial implications. Individually, every component I mentioned is simple. It only becomes a problem when they are clustered.

So what’s the solution?

First, you want to separate your keyboard tasks from the clicking tasks. I want to do all of my clicking first and my typing of descriptions at the end. My Droid calendar interface asks me for the “What,” then makes me click something else, then I must rotate the roulette wheel, then go back to typing for the “Where” and the “Description.” Meanwhile, my brain stubbornly believes the “What” box and the “Description” box should be the same thing. If I’m entering “Bob’s birthday,” what is the “What” and which is the “Description”? Stop screwing with me, Google!

Now let’s consider my brilliant new calendar interface idea. The date and time need to be arranged on a single screen using the same spatial metaphor. On the left you have years, starting from the current year at the top and ascending. The user clicks one to select. No scrolling needed.

On the same screen, to the right of the years column is the month column, with January at the top, descending to December at the bottom. The user clicks a month, no scrolling needed.

To the right of the month column is a column for the hour, with Midnight at the top, descending through the hours. The range between 6 AM and 6PM would be shaded to call out the normal workday range. Click once for the hour.

Next is the minutes column, from zero at the top to 55 at the bottom, in 5-minute increments. The 15/30/45 increments would be larger for easy selection.

With this model, time is organized from big (year) to small (minutes), from left to right, just as you would expect. And it requires your mind to stay in one spatial model throughout. It might require some tweaking to make it finger-friendly for smartphone screens, but that seems doable. Perhaps on the day column, where you need to fit 31 choices, touching it once zooms the general area of your choice so the next click can pick the date with more specificity, then it unzooms.

And I suppose you need another column at the end to pick the length of the meeting.

While I’m complaining, get rid of the “All day” default. If I want a meeting with no start or stop, let me specify the date without the time and move to the next screen.

I’m sure my approach has its own problems. The main idea is to keep the user’s brain in the same spatial processing model throughout the process.

On a related note, I wonder if this brain overload problem is more important to our daily lives than we notice. For example, I have a hypothesis that ping pong can cure attention deficit disorder. A good player will focus intensely on the ball at the moment of impact and instantly switch his attention to the other player, and the larger strategy, until the ball is struck on the other side of the table. In the course of a match, a player quickly switches from intense concentration on the ball to a broader concentration and back hundreds of times. Does the brain of a ping pong player become extra good at controlling concentration as needed?

And is it possible to manipulate people, in terms of marketing or sales, by cleverly overloading the resistance areas of the brain with the sort of thought processes that are meant to exhaust it? That’s probably happening now, but in more of an accidental and trial-and-error way. What if we discovered that asking for a favor works best when the other person is involved in a certain type of mental task?

I need to go combine some spatial planning with some language processing tasks and make a comic now.