After reading John Gruber's thoughts about Tantek Çelik's Three Hypothesis of Human Interface Design yesterday, I felt compelled to contribute to the subject. For those of you mildly interested, but scared of large words, I recommend reading the Daring Fireball post. You'll get the general idea.

At first I considered sending an email to John Gruber about how I overcame iCal's short comings. He is understandably irritated by the interface and all the damned tabbing required to add an event. But then I got to thinking about how drastic a change it was to interact with iCal this new way, and how much muscle memory factored into the process. I thought it might be more helpful to tell the whole internet.

First, the iCal tip.

Adding events to a calendar in iCal can be very aggravating when you interact with the event in the info drawer. There are 13 text fields that need data. Entering text into all of these fields is a pain in the ass, whether you tab or point and click. I would venture to say that pointing and clicking is more of a pain in the ass than tabbing and typing, because the targets are so small.

A much easier method to add events to iCal is to switch to Day view (command+1) or Week view (command+2), and click and drag over the area of the calendar in which you are scheduling your event. When you release the mouse button, the date and time fields are populated, and the event name field is active so you can simply type the name for your event. It really couldn't be much easier. I believe this was the way Apple envisioned the majority of their user base interacting with iCal.

This method doesn't work in Month view, however. This is my preferred default view, and probably the reason I stuck with tabbing and typing event dates into the info drawer for so long. In the various places I have worked, I have noticed that most people prefer their default view be set to Week. I can only speculate that Apple came to the same conclusions in their user testing for iCal.

In my opinion, the cognitive load of the click-drag method of adding events to iCal is drastically less than the tab and type method

1. Click the desired date
2. Switch to Day or Week view (command+1/command+2)
3. Click and drag over the desired region
4. Type the name of the event

1. Double-click on the date of the event in month view.
2. Type the event name.
3. Tab past Location.
4. Tab past “all-day” checkbox.
5. Tab past Month.
6. Tab past Day.
7. Tab past Year.
8. Enter the hour.
9. Enter the minutes.
10. Swap the AM/PM.

Certainly switching from pointing device to keyboard has it's own cognitive load costs, but this is a single switch (at least for those of us who can mouse with our right hand and command key with our left) which I believe off-sets all the typing and tabbing.

My Hypothesis (Decreases in human interface cognitive load are proportional to repetitive use.)

Fitts' Law excludes trained movements, or what we in the rhythm section call “muscle memory”. I think this is fine for the basic concept it is intended to outline—low level movement in a single direction. While it may be difficult to factor this concept into the design of a computer software interfaces,

Tantek’s hypothesis alone are not entirely exempt from this factor. I believe a fourth hypothesis is in order. Tantek defines cognitive load as:

how mentally easy/hard it feels to do something.

It is my belief that this mental load gradually decreases the more you actually interact with something. I would actually expect that this particular hypothesis has already been extensively tested and proven to be law. I expect this because it happens to me every day—or more realistically—every week when I rehearse with my bands.

We rehearse on a regular basis so that it becomes easier to play songs. Songs that we're not quite comfortable with require more effort to remember, and often aren't included on the set list until the cognitive load required to perform them is sufficiently low enough to pull them off without biting our tongues and looking at our hands while playing them.

The cognitive load put into playing older songs mostly goes to thinking about how much you wish you could run to the restroom or sit down for a second. In fact, I don't think I could tell you the keys, tempos, dynamics, or various nuances to the arrangements without sitting down to think about it. I recall most songs on the fly, while I'm playing them. I imagine most musicians do the same. When you've been playing certain songs for years, they seem to just play themselves, no mater how tired or drunk you may be. There is no cognitive load, because there is no thinking.

Go ahead and ask me about the cognitive load levels I had when I switched from playing slab to doghouse, or from a 1/4 size bass to a 3/4 size, or from nickel strings to gut. I'll tell you about lugging two basses to every gig and making up set lists that minimized instrument switching and maximized hand endurance.

Ask me to compare them to the levels I have now. Now that I haven't picked up an electric base in about two years... Now that I can play four hour sets without batting an eyelash... Now that I don't have to worry about blisters...

Thanks to regular rehearsals, I now play double bass and iCal equally as well. It's all about the muscle memory.