Mobile vs. Desktop: Apple iPad Pro and Microsoft Surface

By Norman Chan

What we talk about when we compare mobile and desktop platforms.

I've been testing the iPad Pro for the past week and a half now, using it not only as a go-to tablet, but also as an alternative to a notebook for as many day-to-day tasks as possible. I strapped it inside a Logitech Create keyboard and brought it as my sole computer for a weekend work trip to LA. There's a lot more testing to do--my Apple Pencil hasn't even shipped yet--but I wanted to share with you my thoughts on how the device performs, and where it fits and doesn't fit into my work and home use. Specifically, I want to discuss how it, along with other devices, are changing the conversation and role of what are typically classified as mobile and desktop-class computers.

The release of Microsoft's new Surface devices (Surface Book and Surface Pro 4), along with the release of the iPad Pro has renewed the idea of mobile vs. desktop. You can find many reviews that boil their evaluation down to whether the iPad Pro can replace a laptop, or whether Microsoft's Surface laptops can replace the need for a tablet. I'm not interested in that head-to-head comparison--the products are set at different price points, and in my mind serve different purposes. Their hardware and software design illustrate different priorities for Microsoft and Apple for their respective families of computing devices. It's those priorities and design approaches that are really interesting; I want to compare what the iPad and Surface lines stand for: a future that's mobile first vs. one that's desktop first.

To do that, we should first define our terms. So much of this discussion can get muddled in pointless semantic disagreements. When talking about the iPad and Surface, what categorizes one as mobile, and what categorizes the other as a desktop device? Is it the physical formfactor and size? Having a built-in keyboard? Long battery life? Processor architecture? Touchscreen? App selection? All of the above are important to varying degrees, but I think the difference currently boils down to windowed applications and input models, and how those implementations affect how you can use those machines.

Windows and a Desktop: Multitasking for Productivity

For me, the biggest difference in the way you currently use a desktop-class device (eg. a notebook) and a mobile device (eg. smartphone and tablet) depends on whether the operating system employs a desktop model of running programs and file management. As opposed to runnings apps full-screen, Desktop OSes allow for windowed applications to run alongside each other, on top of a virtual and visualized desktop surface. It's a really simple concept to understand, and yet there are grey areas. For example, the home screen on iOS doesn't count as a desktop--it's just an application list, like the Start Menu in Windows. Simple. But on Android OS, being able to arrange files and shortcuts around a launcher screen and run apps in windows makes those devices more akin to desktop OSes, even though Android is typically classified as a mobile OS.

In Windows, Mac OS, and Linux, the desktop paradigm isn't just about running apps in tiled windows, it's about the shared intermediate space--the "desktop"--for data to move between programs, and more importantly, for you to be able to visualize that information exchange in a spatial metaphor. File management is perhaps the best use of this, but the instancing of programs is another--being able to create multiple instances of applications and have them run side by side. I strongly believe the tiled-window-on-a-desktop model is innately more conducive to productivity because of how our brains group ideas and comprehend the spatial relationships between objects, whether physical or conceptual. That's why I gravitate toward desktop-class computers.

Is a desktop and the ability to run programs in windows essential to productivity? Of course not. We could have "desktop-class" apps on OS platforms that only allow full-screen experiences--some iPad apps are really powerful. But thus far, applications made for mobile apps (eg. ones that run at full-screen) have on average been less robust than their desktop counterparts. Lightroom for Desktop vs. Lightroom for mobile, for example.

But I contend that full-screen mobile apps aren't any less powerful than desktop apps because being full-screen requires them to be. Full-screen apps aren't designed that way simply to lock down user focus to a single task, but were necessitated by the limited mobile processing power and memory of early smartphones and tablets. Conservation of processing power and memory informed the design of mobile OSes and their apps, but is less of a factor now that ARM chips are more powerful without sacrificing battery life. As processing power and memory become more available to the platform, those OSes are going to give developers more leeway to make more robust apps. Multitasking with the iPad Air 2 and iPad Pro is one step toward that, but it really does come down to app support and developers taking advantage of that power. (And based on current app store prices, developers may not be incentivized to build those apps.)

Mobile and desktop OSes may be fundamentally different in their models for application interaction, but the productivity potential of each is determined by the programs developers can and choose to make for each platform.

Input Models: Cursor vs. Touch

The other fundamental difference between a desktop and mobile platform is the input model--which today is often reduced to the difference between an OS designed primarily for cursor or touch-based interaction. There are incredible advantages to each, but neither are the end all be all model for interfacing with computers. A cursor-based model isn't necessarily the most efficient for desktops, and a touch model isn't always the best for mobile. In fact, neither operate alone on the platforms they're most associated with. On desktop systems, the mouse works alongside a keyboard, and there are some tasks like video editing and programming for which the keyboard is the most powerful and efficient input method. On mobile, touch and text input is increasingly being complemented by voice.

Once again, the model that has worked best for the platform was determined by the hardware design and limitations of early devices. Mobile apps and OSes are designed for touch because touch is more intuitive for a mass market and works well on small screens. Precision pointing wasn't viable on mobile without the technology to support it, and so app developers created interfaces with big touch targets and novel ways to interact with swipes and taps. But as the hardware platforms mature and technology improves, we're seeing cursor models--which is synonymous with precision--being incorporated into mobile. iOS has a cursor, even with no mouse support. You can force-press on the on-screen keyboard to move a cursor around text. The stylus and the precision it brings is just another way of incorporating the cursor model to app interaction.

Of course, Windows has embraced touch on the desktop, finally getting to a point with Windows 10 where touch feels every bit as natural to the daily use of the OS as the mouse and keyboard. Input models that historically have differentiated desktop and mobile platforms are cross pollinating, in ways that embrace new technologies and user experience ideas, and ideally in ways that make sense and are intuitive. I'm a huge fan of this. The convergence of desktop and mobile isn't about creating one platform that does everything best, but a race toward the middle from two sides. The future isn't about choosing between desktop or mobile, it's about choosing a platform that's desktop-first or one that's mobile-first.

The 80-20 Use Cases

When Microsoft says that its Surface is a "tablet that can replace a laptop", what they actually mean is the opposite: here's a desktop-first machine that can serve the functions you would typically associate with a laptop. Sure, it's not as elegant a tag line, but it's not misleading, either. And the Surface Book's design really illustrates that desktop-first prioritization. Microsoft is very clear about how it envisions its customers using the Surface Book: 80% of the time as a desktop device, 20% of the time in the tablet form-factor. And in practice, my testing showed that this usage division was accurate.

The Surface Book and Surface Pro are do that 80% part really well. They're great at multi-window applications, and incorporate touch and the stylus in intuitive ways that complement the desktop experience. It executes that 20%--the mobile experience--not as well. Full-screen apps are serviceable, but not optimal. Watching Netflix on full-screen is just fine, but full-screen drawing with OneNote is cumbersome. Windows 10's "tablet mode" setting isn't good enough to adapt apps designed for the desktop for mobile use. Menus don't fade away or adapt properly. The presence of the window frame is felt, even when you can't see the actual window.

On the iPad Pro, even though Apple doesn't say it explicitly, I'm finding that it's very good as a "80-20 device" as well. Except here, I'm using 80% as a tablet, 20% as a laptop replacement. This really is the "tablet that can replace a some cases". And that's how a lot of people have been using the iPad for a long time--as a replacement for their laptops. My parents use their iPad as their primary computer. Fast Company's Harry McCracken has used an iPad instead of a laptop at every product briefing I've shared with him. The iPad Pro's proposition isn't that it's going to replace your desktop, but that it's improving the quality of that 20% time people may be using it instead of a laptop. I don't think it's succeeded, yet. The processing power may be there, but the application support and consistency of experience isn't. Typing in Google Docs felt slow, and keyboard shortcuts like for copy and paste didn't work everywhere I expected them to. The apps aren't there yet, and iOS needs work to better support multi-tasking and navigating with just the keyboard.

A Race for the Middle Ground

I see the convergence of desktop and mobile devices as a race to the middle, with Microsoft and Apple choosing different paths to get there. Microsoft is betting their 80-20 platform on the Surface, adapting Windows to stay really great as a desktop-first platform and slowly improving the mobile use cases. Its Surface hardware and Windows 10 design choices reflect that progress. Microsoft's disadvantage is that it doesn't yet have a 80-20 device that's tablet-first. For people who prioritize mobile, iPad Pro is currently their best bet. Apple, on the other hand, doesn't have a desktop-first device that can also serve mobile needs. Mac OS and iOS won't be collapsed into one platform. The iPad Pro shows that Apple acknowledges the value of breaking down the barriers between desktop and mobile, but its implementation of that 20% desktop use case on iOS is only serviceable, not great. Improving that 20% experience on these hybrid devices should be a huge priority and can be a huge opportunity. But it'll depend on improvements to Windows and iOS, as well as developers willing to make the software to support them.