Categories
Uncategorized

Mouse Faster

Keyboard vs. The Mouse, pt 1

We�ve done a cool $50 million of R & D on the Apple Human Interface. We discovered, among other things, two pertinent facts:

  • Test subjects consistently report that keyboarding is faster than mousing.
  • The stopwatch consistently proves mousing is faster than keyboarding.

This contradiction between user-experience and reality apparently forms the basis for many user/developers� belief that the keyboard is faster.

Which brings the more general question, are VI/EMACS really better editors/ides than say VisualStudio?

[via mpt]

6 replies on “Mouse Faster”

Interesting that what Tog suggests is basically what GNOME does:

“3) User-specified command keys are the best solution to the current problems, leading to the following guideline:

Guideline: All command-keys should be user-specifiable. The developer can and should supply an initial set, but the user should be able to overrule those choices.”

I.e. you can easily switch hotkey combinations via the menus in GNOME. Of course, they do it in a crappy way and it is not universal, but there ya go.

In general, though, I see Tog’s point. I think that he may be right that weird key combinations are probably slower than mousing. However, there are three (no four!) things that I think qualify his statements:

1) work done in 1989… may want to redo that research today

2) what about simple and repetitive key combinations? I know I can go WAY faster when doing REPETITIVE things when I have key combos. Like, if I *know* that I am doing a bunch of bolds, command-b is so much faster than pulling down the menu. Now, on individual commands, he may be right.

3) I think the post about subjective time is very important… good interface is what the user thinks is good interface AND increases productivity: there is a reason the windows world is stuck on alt-x-y combinations: they FEEL right. The sweet spot is when you can combine both feeling right and being right.

4) I don’t get RSI in my hands, i get it in my elbow from mousing too much, or holding the mouse incorrectly. I think you have to consider RSI in this study, to the good or ill of the keyboard or mouse.

It’d be interesting to see a test like this specifically for unix/gnome. I have a feeling that a lot of the way that apple does things is the reason for the strange data. Apple doesn’t have something like emacs or vi, ie: designed for non-keyboarding. All their software is (AFAIK) designed to be used with a mouse, and based on my own experiences, non-mouse behaviour is stuck on as an afterthought or given the regard of a second-class citizen. IE: The suck involved in os/x keyboard based menu navigation.
This is “Ok” though, because these days non-mouse interfaces are far between, and you’ll never see one on a mac. http://xvsxp.com has the menu-nav skinny between os/x and winxp.
Engel – good points.

When I read Tog’s account of mouse is faster than keyboard I laughed. It’s so broad its funny … but it ends up sad in the end. I hear it quoted as absolute truth, and people eat it up. People forget to think.
After reading Tog’s book, I had to time myself. I used a few common applications where speed matters: gvim, textpad, bash, and explorer. Saving a document (gvim, textpad) was always faster with keyboard shortkuts. Browsing and executing files was sometimes faster by mouse, sometimes by keyboard (bash vs. explorer). Advanced tasks (find, replace) were faster via keyboard (gvim, textpad, bash). An informal test, but measured.
My conclusion: Mouse control /can/ be faster than the keyboard, for certain actions. Keyboard control can also be faster than mouse, for specific actions. Knowing which is which is key.

Umm, mx, if you have ever heard of this little thing called the Heisenburg (sp) uncertainty principle, then you know that your experiment is even more laughably incorrect than Tog’s could ever dream to be.

Tog said that a keyboard shortcut is faster once you have decided to use it, but that in everyday use switching “modes” to use a keyboard shortcut is slower than the mouse. It is the mental process that takes so long, not actual physical process of hitting the keys. Timing yourself would only be timing the keystroke, not the decision-making process. There is no way that you could accurately record yourself making decisions, because the decision to record a decision would keep you from making a correct amount of time in the decision. Can’t be done, dude.

This is not to say I agree with Tog on this point (I don’t), just that the “science” behing your statement is just really, really, horribly wrong.

Yes, my tests were small, and as easily flawed as anyone’s. I did test whole actions though — like producing the same document accross tools, using either mouse, or keyboard for command-based items. The point was that some things are faster one way, as shown by some andectodal evidence. The Apple tests were as bogus as mine, and the general statement is proof.

Comments are closed.

Categories
Uncategorized

Mouse Faster

Keyboard vs. The Mouse, pt 1

We�ve done a cool $50 million of R & D on the Apple Human Interface. We discovered, among other things, two pertinent facts:

  • Test subjects consistently report that keyboarding is faster than mousing.
  • The stopwatch consistently proves mousing is faster than keyboarding.

This contradiction between user-experience and reality apparently forms the basis for many user/developers� belief that the keyboard is faster.

Which brings the more general question, are VI/EMACS really better editors/ides than say VisualStudio?

[via mpt]

6 replies on “Mouse Faster”

Interesting that what Tog suggests is basically what GNOME does:

“3) User-specified command keys are the best solution to the current problems, leading to the following guideline:

Guideline: All command-keys should be user-specifiable. The developer can and should supply an initial set, but the user should be able to overrule those choices.”

I.e. you can easily switch hotkey combinations via the menus in GNOME. Of course, they do it in a crappy way and it is not universal, but there ya go.

In general, though, I see Tog’s point. I think that he may be right that weird key combinations are probably slower than mousing. However, there are three (no four!) things that I think qualify his statements:

1) work done in 1989… may want to redo that research today

2) what about simple and repetitive key combinations? I know I can go WAY faster when doing REPETITIVE things when I have key combos. Like, if I *know* that I am doing a bunch of bolds, command-b is so much faster than pulling down the menu. Now, on individual commands, he may be right.

3) I think the post about subjective time is very important… good interface is what the user thinks is good interface AND increases productivity: there is a reason the windows world is stuck on alt-x-y combinations: they FEEL right. The sweet spot is when you can combine both feeling right and being right.

4) I don’t get RSI in my hands, i get it in my elbow from mousing too much, or holding the mouse incorrectly. I think you have to consider RSI in this study, to the good or ill of the keyboard or mouse.

It’d be interesting to see a test like this specifically for unix/gnome. I have a feeling that a lot of the way that apple does things is the reason for the strange data. Apple doesn’t have something like emacs or vi, ie: designed for non-keyboarding. All their software is (AFAIK) designed to be used with a mouse, and based on my own experiences, non-mouse behaviour is stuck on as an afterthought or given the regard of a second-class citizen. IE: The suck involved in os/x keyboard based menu navigation.
This is “Ok” though, because these days non-mouse interfaces are far between, and you’ll never see one on a mac. http://xvsxp.com has the menu-nav skinny between os/x and winxp.
Engel – good points.

When I read Tog’s account of mouse is faster than keyboard I laughed. It’s so broad its funny … but it ends up sad in the end. I hear it quoted as absolute truth, and people eat it up. People forget to think.
After reading Tog’s book, I had to time myself. I used a few common applications where speed matters: gvim, textpad, bash, and explorer. Saving a document (gvim, textpad) was always faster with keyboard shortkuts. Browsing and executing files was sometimes faster by mouse, sometimes by keyboard (bash vs. explorer). Advanced tasks (find, replace) were faster via keyboard (gvim, textpad, bash). An informal test, but measured.
My conclusion: Mouse control /can/ be faster than the keyboard, for certain actions. Keyboard control can also be faster than mouse, for specific actions. Knowing which is which is key.

Umm, mx, if you have ever heard of this little thing called the Heisenburg (sp) uncertainty principle, then you know that your experiment is even more laughably incorrect than Tog’s could ever dream to be.

Tog said that a keyboard shortcut is faster once you have decided to use it, but that in everyday use switching “modes” to use a keyboard shortcut is slower than the mouse. It is the mental process that takes so long, not actual physical process of hitting the keys. Timing yourself would only be timing the keystroke, not the decision-making process. There is no way that you could accurately record yourself making decisions, because the decision to record a decision would keep you from making a correct amount of time in the decision. Can’t be done, dude.

This is not to say I agree with Tog on this point (I don’t), just that the “science” behing your statement is just really, really, horribly wrong.

Yes, my tests were small, and as easily flawed as anyone’s. I did test whole actions though — like producing the same document accross tools, using either mouse, or keyboard for command-based items. The point was that some things are faster one way, as shown by some andectodal evidence. The Apple tests were as bogus as mine, and the general statement is proof.

Comments are closed.