?

Log in

No account? Create an account
Previous Entry Share Next Entry
A Quiz Designed to Give You Fitts
device
jducoeur
Today's link comes via my co-worker Bob Peterson, who pointed the engineering group to it. It's not so much about programming this time; rather, it's about UI design. But really, if you're a programmer coming anywhere near the UI, it's well worth reading.

I recommend going through the page as it says: take the quiz seriously, try to come up with the answers on your own, and then continue on to the answers. The point of the exercise is to demonstrate a basic principle of UI design that is very often overlooked, and he makes a bunch of good cases for why it should be paid attention to. (Personally, I got a few of the answers right off the top of my head, but didn't suss the common thread until it was pointed out.)

Some longtime engineers will undoubtedly look at this and decide that it is pointlessly persnickety -- that the recommendations violate How Things Are Done, and are more trouble than they're worth. But part of what separates a truly great UI from a merely functional one is re-examining your assumptions frequently, and paying attention to those persnickety details. (One reason why our new check fraud product is going to win in the marketplace is that we *are* spending the time to design and build an unconventional but kickass UI...)

  • 1
would have been harder if the answers were nto in the subject line...

Would have been harder if I had never heard of Fitt's Law. Would have been much harder if I didn't read Ask Tog every few months.

Well, I suspect that the point here is that there are an enormous number of people out there designing UIs who either don't know the Law in question, or don't think about it often enough. I think the quiz does a good job of motivating the underlying idea...

So, like...does this website explain who decided that when I am using keyboard to navigate a dropdown and I want Gilbert, I hit g and it goes to g, then I hit i and it goes, not to gi, but to i?

No, it says that a dropdown should not be so long that you need two letters.

Sad that this is still relevant after 10 years.

I still remember Fitts' every time I use a non-mac system and have to slow down to ge the menubar.

That said, either I'm unusual or I don't agree with Tog's listing of the speed order for the 5 magic pixels. I mean, the first one I agree with, and I agree with the answers, just not the bonus order. Maybe I have a stiff wrist from trackpad usage, maybe I just move my hand more from the elbow and shoulder, but the upper right corner is much faster for me than the lower right, and is my fastest corner - I just swing my arm out. But otherwise, the ordering seems correct.

Yes, I was puzzled by the order of the corners too. I would think that the fastest corner would be the bottom-left, using the biceps to swivel around the elbow, followed by the top-right, using the triceps ditto.

Anyway, I Am Not A User Interface Designer, but I did surprisingly well. I didn't know the name "Fitts' Law" (although I dimly recall something called a "Fitts List", which lists things humans are good at vs. things machines are good at, and recommends allocating tasks that way). But I nailed several of the questions, and was on the right track for several others.

I was thinking about this as I used my laptop tonight. I suspect it is a function of where you put your mouse/pad in relation to the pivot point (elbow). If your mouse is far enough outside the scope of your body, you have already twisted your arm out to grab the mouse, and the angle to the outside is to the bottom and the inside is towards the top. If you keep the mouse close to the inside of your body, then when you swing your arm, the arc is the other way, being the other side of the forwardmost point.

Laptops with their trackpads centered enforce a custom of moving the pointer from the inside of the body's reach. I may position my mouse in a similar orientation out of learned response.

I agree about disagreeing about the order. But I don't mouse with my elbow, but rather my wrist. So my order is BL, TL, TR, BR. I don't think it matters that I am left handed, because I mouse with my right.

Yeah, I also differed slightly from his order -- I don't think his logic is quite as sound there as in much of the article...

  • 1