If you haven’t already and you want an interesting weekend project, get a shooting timer and check how fast your hotkeys actually are compared to the mouse.
Back in 2009 someone told me, a 12 year Linux user at that point with a huge memory of key bindings, about plan 9 and their research into efficiency and speed of using key bindings over a mouse. I couldn’t believe it, they found that there was no appreciable speed benefit of using keys over a mouse.
When I say I couldn’t believe it I’m speaking literally. I wasn’t able to comprehend how what felt so slow to me could not be beaten by what felt so instantaneous! So I got a shot timer and tried it on my own.
A shot timer is a stopwatch that waits a short but random number of seconds before beeping then counts up until the requisite number of bangs have happened or some other thing that makes it stop. They were designed for shooting (as you might imagine) where you want the person to start in some state of readiness and time how long it takes for them to get on target and shoot without being able to “count” themselves in.
Keys were slightly (fractions of a second) faster for things I had designed my workflow around and used all the time like changing desktops, opening terminals, moving stuff between desktops.
Mouse was faster than opening a terminal and typing the name of a program to launch it, and moving files around was faster with the mouse almost all the time. A combination of mouse and “universal” keybindings (ctrl-c,v,x,z) made it faster all the time.
Across flux/blackbox, xvwm, windows and osx, the box window managers were fastest with keys, mostly because you can’t do much with a mouse in em at least the way I was running, xvwm, windows and osx were either dead even on average or faster with a mouse.
It’s a fun thing to try if you’re ever wondering how much time you’re actually saving.
Hey, that’s actually a very nice project, and to be honest, I can kinda imagine that the saving is minimal if there at all, in terms of time. Partially, I think this is also due to the fact that we are talking about super small amounts of time anyway! Moving files around I think it’s totally fast with a mouse, and in general I still do it like that.
For me speed is really a secondary thing, it’s about ergonomics and limiting my movements. Chances are, I am already writing on the keyboard when I want to do something, so it might not be faster to switch to browser with mod+2 and back to terminal with mod+1, but it’s less movement to find the mouse, rotate the shoulder (my split kb is open at shoulder width) etc.
Also I think I would argue that requires less focus because it’s inherently more mechanic as an action compared to find a button and click, or dragging and dropping something. Either way, it’s for sure something interesting to look at!
Next time I’m around the family for awhile I wanna repeat it with a blood pressure monitor. I always feel more intense when using keyboard shortcuts, the question is if I’m also running a higher blood pressure when using them.
If you haven’t already and you want an interesting weekend project, get a shooting timer and check how fast your hotkeys actually are compared to the mouse.
Back in 2009 someone told me, a 12 year Linux user at that point with a huge memory of key bindings, about plan 9 and their research into efficiency and speed of using key bindings over a mouse. I couldn’t believe it, they found that there was no appreciable speed benefit of using keys over a mouse.
When I say I couldn’t believe it I’m speaking literally. I wasn’t able to comprehend how what felt so slow to me could not be beaten by what felt so instantaneous! So I got a shot timer and tried it on my own.
A shot timer is a stopwatch that waits a short but random number of seconds before beeping then counts up until the requisite number of bangs have happened or some other thing that makes it stop. They were designed for shooting (as you might imagine) where you want the person to start in some state of readiness and time how long it takes for them to get on target and shoot without being able to “count” themselves in.
Keys were slightly (fractions of a second) faster for things I had designed my workflow around and used all the time like changing desktops, opening terminals, moving stuff between desktops.
Mouse was faster than opening a terminal and typing the name of a program to launch it, and moving files around was faster with the mouse almost all the time. A combination of mouse and “universal” keybindings (ctrl-c,v,x,z) made it faster all the time.
Across flux/blackbox, xvwm, windows and osx, the box window managers were fastest with keys, mostly because you can’t do much with a mouse in em at least the way I was running, xvwm, windows and osx were either dead even on average or faster with a mouse.
It’s a fun thing to try if you’re ever wondering how much time you’re actually saving.
Hey, that’s actually a very nice project, and to be honest, I can kinda imagine that the saving is minimal if there at all, in terms of time. Partially, I think this is also due to the fact that we are talking about super small amounts of time anyway! Moving files around I think it’s totally fast with a mouse, and in general I still do it like that. For me speed is really a secondary thing, it’s about ergonomics and limiting my movements. Chances are, I am already writing on the keyboard when I want to do something, so it might not be faster to switch to browser with mod+2 and back to terminal with mod+1, but it’s less movement to find the mouse, rotate the shoulder (my split kb is open at shoulder width) etc. Also I think I would argue that requires less focus because it’s inherently more mechanic as an action compared to find a button and click, or dragging and dropping something. Either way, it’s for sure something interesting to look at!
Next time I’m around the family for awhile I wanna repeat it with a blood pressure monitor. I always feel more intense when using keyboard shortcuts, the question is if I’m also running a higher blood pressure when using them.
I could check with my fitness tracker, it has a “stress level”. Might not be super accurate though.