Pointer Control

I’d like to share a few thoughts about pointer control on the iPad. My perspective is narrow and specific: I’ve been using an iPad as my full-time computer for three and a half years, and my job is writing books. The iPad does everything I need it to, and I’ve written a lot already about why it’s the right device for me.

In particular, I wrote a piece about the weird disconnect of pointer-driven interfaces, which only really becomes apparent when you move away from the familiarity of pointing devices. Switching entirely to the iPad meant committing to many new ways of doing things, to the extent that I now find traditional computers a bit awkward to use. They feel like machines with interfaces designed to let you use the pointer, rather than with a pointer designed to let you use the interface. And of course that’s exactly what they are, because it’s logical to make the software capitalise on the nature of the hardware.

Pointing devices were a brilliant innovation, both because of their pseudo-intuitiveness (once you got past the ninety-degree mental flip where moving your hand horizontally away from you means up, and vice versa), and the fact that they did it with modest hardware: a wheel in a box. The intersection of kind-of intuitiveness and technical pragmatism led to a feeling of inevitability, and thus well-deserved ubiquitousness.

Apple is very, very good at designed inevitability. Those things that seem obvious, so much so that they become the reference model within a year. They’ve done it dozens of times. It’s one of two major reasons why I find it difficult to imagine myself choosing to leave the Apple ecosystem. The second reason is that they love to throw stuff away.

When you write books, you get accustomed to throwing stuff away. Not quite comfortable; just accustomed. You have to. Some of what you write is either garbage outright, or garbage contextually. It has to go, whether that’s to the bin, or just somewhere other than where you wrote it. Killing your darlings is one of the premier hardnesses of fiction, so it’s best to get used to it as soon as you can. And Apple does it constantly, with an apparent gleefulness that I’ve learned is actually the hard-won feeling of liberation that’s borne of ideological purity.

For people of my background and generation, it was a bit of milestone when Apple got rid of the floppy drive. Switching to USB was another, but the floppy drive’s dismissal felt like what’s now quintessentially Apple: both a statement and a prognostication. I’ve learned to give those decisions the benefit of the doubt, because I invariably discover their wisdom on my own afterwards.

But it’s so tempting to second-guess. To cling to what we know, and see it as minimum viability, instead of the status quo. It’s so easy to frame things in terms of what’s different, instead of what’s better. The problem is exacerbated by how a lot of people don’t really think about things, at least beyond a superficial level. That’s necessary for survival; we can’t spend our time navel-gazing. But it vexes me when people talk about things as if they’ve thought carefully, then get surprised or defensive when you point out something that even a little more consideration would have rendered obvious.

Here’s an example: pointer control on the iPad isn’t a mouse; it’s a keyboard.

Most people don’t, and won’t, understand that. But it’s evident and simple. Traditional computers in the style of Windows and macOS are pointer-driven devices: their primary interface is made for a pointing device (external to the display) with a pixel-precise cursor. But they can also, optionally, be controlled via a keyboard. You can launch apps, trigger menus, perform commands and so on, all via the keyboard — and we all know this, because we see keyboard shortcuts displayed in those menus we pop open with the pointer. The whole interface is made for the pointer, and optionally controllable via the keyboard if and when you find it advantageous to do so.

On the iPad, the pointer is a keyboard, in that sense. This time, the system is designed for your fingers, but you can also optionally control it via a traditional pointing device (or indeed the keyboard too, as a third option).

There are people — a lot of people — who heralded the advent of iPad pointer control with a collection of phrases involving words like “productivity” and “laptop” and “serious” and “work”. In their thinking, which is very close to non-thinking, the addition of pointer control finally fulfils a manifest destiny — which, implicitly, is for the iPad to be exactly the kind of computer that it isn’t. Those people are wrong. Transparently, trivially, category-error wrong.

Pointer control is actually another entry in the set that includes hardware keyboards, Apple Pencil, and now trackpads and mice. That set comprises optional additional input methods which are secondary to the device’s designed interaction paradigm. This, in turn, is transparently and trivially correct.

It’s like buying different lenses and accessories for a camera, isn’t it? Or, I don’t know, a hole saw. Or one of those screwdrivers for the star-headed screws. Some people need those things in their toolbox (or camera bag). Some people don’t. The important thing is to know which kind of person you are, and acknowledge that other people aren’t necessarily the same as you.

Some people encounter contexts where sub-finger accuracy would be helpful. Accuracy is a function of two things: the size ratio between the pointer and its smallest targets, and the degree of control precision available when using the pointing device. Control precision, in turn, is a factor of minimum reproducible movement distance, and the required sustained muscle tension before control degrades. Fingers on a screen don’t exceed at either of those things, especially if the screen isn’t bearing all (or any) of the hand’s weight. So there’s a genuine need — for some people. There’s a functionality-suitability gap, for given use cases.

As for Apple’s specific implementation of pointer control, well… it’s new. Of course it is. I don’t love the pronounced pointer affinity for big objects like app icons, giving a slightly tacky (in the adhesive sense) or gravitational feel. It’s a lovely thing to look at, and a delightful demonstration of intentionality of engineering. I’m just not convinced that it’s wholly aligned with pointer control’s raison d’être. But I’m willing to be convinced. And in small-target-size scenarios, it feels like what they want it to be: aim assist. I applaud their continued willingness to question everything.

There are two main ways to design something: make it optimal, or make it familiar. You can almost always incorporate elements of both, but you get to (and should) choose which one is your guiding philosophy. For Apple, it’s optimality. The iPad was created according to a philosophy, and whatever expansions of functionality it has undergone since then, they’ve been in the service of broadening its utility — but always within the context of its designed identity.

What I like about pointer control on the iPad isn’t that it’s now available, but rather that I only have to use it when I want to.