Mouse vs touch-pad vs TrackPoint: Fight!

Benji Shults
10 min readOct 5, 2019

--

…in which we focus on objective measures of productivity from the perspective of software engineering.

Everyone has their preferences. You may be used to one thing and you may not want to change. There’s no accounting for taste and this article will not try.

Instead, I aim to use objective, technical, measurable criteria to compare the three most popular pointer-control and scrolling devices.

Setup

I assume that if your machine has more than one of the pointer control devices available, then all but one of the three are disabled.

TrackPoint

Lenovo ThinkPad USB Keyboard with TrackPoint

Many high-end laptops come with this technology including Lenovo, Dell, Toshiba, and Sony.

For the TrackPoint, I will assume that it is configured as it typically comes out of the box: tapping it is NOT a mouse click and adaptive acceleration is enabled. Further, I assume that there are three hardware buttons below the keyboard as there are in every TrackPoint keyboard I have encountered.

Touch-Pad (TrackPad)

Apple TrackPad 2

For the touch-pad, assumptions are more difficult because of the variety of configurations. I will assume that the touch-pad is configured with a good set of gestures including two-finger scrolling.

Mouse

Dell no-frills mouse
Apple Magic Mouse 2

For the mouse, I will assume that the mouse has at least the standard button(s). On a Mac, this means one button. On a PC, I will assume that there are two buttons and a scroll-wheel.

Productivity/Functionality Assumptions

How much time and effort is required to accomplish something with the device?

I am assuming that time spent with your hand is an unproductive position is waste. If we could eliminate it, we would be more productive.

I also assume that you know how to touch-type. Fingers on the home row is your position of highest productivity if you are a software engineer.

I assume that you learn and use the keyboard shortcuts of the tools you use. If the shortcut is easily reachable from the home row, then the keyboard shortcut is going to be faster than any of the pointing devices.

If you don’t know how to touch-type or you don’t learn keyboard shortcuts, then the productivity gain you will get from learning those two things will probably make more of a difference than switching your pointer device. Go do that, then come back here.

Under these assumptions, it follows that we will measure the time and effort of using your pointing device from the point where your fingers are on the home row to the point that your fingers are on the home row again.

An aside on keyboard usage and configuration

An objection is raised that some keyboard shortcuts require moving your fingers off the home row.

One example of this is keyboard shortcuts that require the use of the control/Ctrl key (which are many.)

Consider that

  1. the control/Ctrl key is frequently used as part of many keyboard shortcuts
  2. the control/Ctrl key can be difficult to reach with your fingers on the home row
  3. the CapsLock key is never needed and very rarely used

This is easily solved by mapping control/Ctrl to CapsLock. For Windows, I recommend Microsoft’s Sysinternals. This is an excellent suite of tools and it includes a utility two configure the behavior of the CapsLock key. The standard Linux distros and the Mac OS provide utilities for this as well.

A second example is the function keys (F1-F12).

My recommendation is to configure your tool to use more reachable shortcuts. Emacs and Eclipse are great examples of tools with reachable keyboard shortcuts. IntelliJ Idea can be configured to use a keyboard mapping similar to either of those two tools.

Productivity/Functionality Comparison

Moving the pointer

How much time and effort does it take to get from starting position to moving the pointer then back to the home row?

With the TrackPoint, all of your fingers stay on their home key except one. That one finger moves to a comfortable position one-and-a-half keys away from its home… closer than the number keys. Mouse clicking is done with a thumb which only needs to slide down from the spacebar.

Time and effort: negligible.

With the touch-pad, one hand must leave the home row completely, move off the key area and onto the touch-pad. Once the pointer is moved and the pad pressed for a click, that hand must return to its position on the home row. If done naturally, this is probably less than a second wasted each time.

Time and effort: a little less than a second each time the mouse must be moved.

With the mouse, it is similar to the touch-pad except that, generally, the hand needs to move further and the whole hand has to move to control the pointer. If done naturally, this is probably more than a second of waste each time.

Time and effort: more than a second each time the mouse must be moved.

Can you move the pointer while in starting position? (E.g., while typing.) Unless you use the TrackPoint, you’ve probably never even considered this possibility because you cannot have both hands on the home row and one hand on your pointing device at the same time.

So, with the TrackPoint, not only do we not waste time moving hands on and off the keys, the time spent moving the mouse can overlap time spent with typing with both hands on the home row.

These considerations are so enormous that the other technologies have a lot of ground to make up.

Scrolling

All three types of devices support vertical scrolling.

With the TrackPoint, one thumb slides off the spacebar to the middle button and one index finger slides over to the nub. With the TrackPoint, there is no limit to how for one can scroll without interrupting the flow to reposition a finger.

With the touch-pad, as we’ve already discussed, we have to move a hand off the home row. But what’s worse, we must reposition our fingers every page or so!

With the mouse, it is similar as with the touch-pad. The finger must be repositioned every page or so.

Another considerable advantage to the TrackPoint. If you are watching a screen capture that involves someone scrolling, you can tell if they don’t use the TrackPoint by the pauses in scrolling. That’s waste.

What about horizontal scrolling?

If you use a device that doesn’t support it, the you don’t know how useful it is. Obviously, this is useful when text extends off the screen horizontally. (Yes, you can move the pointer to the scroll bar — if there is one — and scroll that way, but that is waste compared to having a hardware gesture.) When you have a lot of tabs open, horizontal scrolling allows you see other tabs in the tab bar without needing to move the pointer to the scroll arrow.

Every TrackPoint implementation I’ve used supports this.

With a touch-pad or a mouse… YMMV. Some support it, some do not.

Clicking

Start simple: how much time and effort is involved in a simple click?

If you’ve read up to here, you know that for the TrackPoint, this involves a tiny movement of the thumb while all other digits can stay on the home row.

For a mouse, obviously, one hand must move well off the keys.

For a TrackPad, well, there are various designs. Some make it as difficult as it is for a mouse, some as easy as it is for a TrackPad, and some are in-between.

Now, what about modified clicks? E.g. Ctrl-Click, Atl-Click, or what have you.

With the TrackPoint, one finger (or thumb) moves to the modifier key and a thumb moves to the hardware mouse button below the spacebar. Most modified clicks can be accomplished with seven of eight fingers remaining on their home keys.

With a mouse or touch-pad, one hand has to move completely off the keys. We are back to about a second of wasted time and effort just to execute a modified click.

Zoom

This is application-specific but typically involves a modifier key combined with the use of the pointer device or a gesture on the touch-pad.

This category is going to end up with a result similar to the scrolling and clicking results above.

Pointer movement distance

How much waste occurs when we move the pointer in the same direction for a long distance.

All three technologies support acceleration. Let’s consider problems that are not solved by that technology. E.g., moving the pointers across multiple monitors or (as in a game) moving the pointer in the same direction for a long time.

Moving the mouse while it is lifted is waste because it is time and effort that is spent only because of the limits of the device. We have to do this with a mouse because the mouse is not placed on an infinite surface and our arms can only reach so far comfortably. When the mouse gets near the edge of its surface, or to a position that is not ergonomic, it must be lifted and moved. Any pointer movement that occurs during this process is unintentional and that time and effort are waste.

Gamers will be painfully aware of this limitation of a mouse.

Lifting the finger off the touch-pad and repositioning the finger to another spot on the pad is waste because it has no effect on the position of the pointer. We have to do this with a touch-pad because its size is not infinite and our fingers are not infinitely long. When a finger reaches the edge of the pad, the finger must be lifted and repositioned. This is waste and this is why you wouldn’t even consider playing many games with a touch-pad.

Pointer movement speed

All three technologies support adaptive acceleration. In the case of the mouse or touch-pad, you can simply move your hand or finger faster. With the TrackPoint, you just press harder or softer for speed.

I’m going to give this to the mouse with the touch-pad coming in a very close second and the TrackPoint coming in a very close third. If we count the time it takes to move to and from the home row, then the scene changes in favor of the TrackPoint.

Pointer movement accuracy

This is about getting the pointer exactly where you want it to be and no further.

For me, this is a win for the mouse.

I don’t know how much practice would be required to get mouse-like accuracy with a touch-pad or TrackPoint but I haven’t seen it done.

Productivity Conclusion

By almost every measure, the TrackPoint wins for productivity. The chief exception being the accuracy of the mouse.

So why do people use anything else?

For gamers, accuracy and speed are paramount. Not only that, typically, the mouse hand never needs to move to the keys.

But we are focused on software engineering. Why do software engineers use anything but a TrackPoint? I think there are several reasons. None of them are technical.

  1. “Design.” (See below.)
  2. Social — all the cool kids speak vehemently about their disdain for the TrackPoint (no technical reasons given.) Agreeing with them helps to form social community. You can find blog posts from fanboys (I won’t link to them) that say things like “If you’ve ever used it, you know it’s pretty much the worst thing ever. The nipple mouse isn’t just incredibly ugly — it’s also very hard to use.” and “It’s old technology, it’s incredibly ugly and nobody wants it.” I’ve never seen technical reasons given.
  3. Market — only high-end laptops have TrackPoints built in. By the time you (or your company) make enough to afford these, developers already have their habits, opinions, and social networks formed.

Design

I promised not to offer opinions on subjects of taste. I leave that to the experts. The best design is whatever Apple or (shudder) Microsoft say is best at the moment. Apple make the most noise about being design experts and they say that one big, unlabeled button that doesn’t look like it is clickable is best for everything.

Apple TrackPad 2 and Apple Magic Mouse 2 win this easily. (Obey)

Recommendation

I use the ThinkPad Wired USB Keyboard with TrackPoint or the ThinkPad Wireless Keyboard with TrackPoint. It works with any laptop or computer (or phone).

I remap the CapsLock key to act as another control/Ctrl key.

I use or configure applications to use keyboard shortcuts that don’t involve function keys and I learn those shortcuts.

Conclusion

What criteria did I miss?

Can you think of productivity measures where the touch-pad wins?

Disagree with my “measurements?”

Comment below but please keep it productive, constructive, and technical if possible.

Additional References

--

--

Benji Shults

Staff Software Engineer at SmartThings with a PhD in Mathematics and Artificial Intelligence from UT-Austin