Reference Information: Only One Fitts’ Law Formula – Please!, Drewes, Heiko. Conference on Human Factors in Computing Systems.
Summary: Fitts' law describes the mean time it takes to move a pointing device into a specific region. In the field of Human-Computer Interaction, it is one of the view formulas which is based in hard science. However, since it's inception, the exact formula for Fitts' law has been heavily debated, even fit Fitts himself.
The problem with having multiple formulas is that a paper's author can pick whichever formula makes his or her hypothesis work. Drewes also points out that, "At the moment, referring to Fitts’ law pro- vokes a long discussion and shifts the focus away from the original research question."
For such a pivotal formula in the field of Human-Computer Interaction, no consensus has yet been made on which formula represents Fitts' law the best. Unlike other fields, the controversy over Fitts' law doesn't stem from people using different assumptions to define the same concept. Instead, all of the current formulas work from the same premise but represent it differently. Some use different sets of variables, while others simply add onto Fitts' original law, adding in variables which they consider important enough to be included. Over the course of the paper, Drewes presents his argument for why there should be one standardized formula representing Fitts' law.
Discussion: Whenever I've heard of Fitts' law in the past, I always heard people mention it in the context of Window's Start Menu, or OS X's menubar, citing "The easiest thing to hit with your mouse cursor is a corner." I never realized that Fitts' law described hitting any arbitrary point with a pointing device.
Some of the math toward the end of this article was a little hard to follow, since everything's been abstracted away. However, the diagrams included, such as the one above, aided greatly in my understanding of the topic.
Drewes article is well laid out, and doesn't assume any previous knowledge on the topic. Over the course of ten pages, Drewes is able to establish what Fitts' law is, why it's important, the various ways it's represented, and finally, why you should believe is stance that there needs to be one agreed upon formula to represent Fitts' law.
I think it's interesting how any Fitt's formula can be used to fit the users hypothesis and that's acceptable. If the HCI wants to be viewed as a fully functional scientific community then using any formula sounds wrong. So, maybe it is best if one formula is standardized for the subject.
ReplyDeleteIt is interesting that there is no standard formula that everyone uses, perhaps so that results may be skewed in such a way that provide a correlation when none is found. HCI seems more like an "art" than anything, as the author says in the final couple of paragraphs. I thought it was comical how he said at the end of the day, it's all just sort of common sense. Don't make buttons too large or too small.
ReplyDeleteI thought it was quite hilarious that there exist like 4 equations to determine the pointer location and his attitude toward it. It reminds me of Happy Gilmore when the guy was like "Damn you people, this is golf!" but instead it would be like "Damn you people, this is science!"
ReplyDelete