Nokia N900 experiences and what’s in the iPhone usability fairy dust
In June this year I wrote down some thoughts about the single task design of the iPhone OS and hardware, and after having played with the new Nokia N900 properly for a couple of weeks I think the argument is even stronger - the single tasking design, no matter how much the geeks/power users have hated it and how the iPhone is slowly growing out of it, has been core and key to creating a whole new category of “mobile computing devices” that are actually used by people who are not geeks.
The N900 is an extremetly cool device, basically the capabilities of a netbook in a mobile phone format. It does true multitasking, several windows open at the same time, music playing, two SMS bus tickets on the same screen to show the bus driver. Conference backchannel in one window, switching to email and twitter in others. It’s very powerful, and I’m sure the device will have enormous appeal to pro users who want this functionality. It really is more of a computer than a smartphone.
But playing with this device as well as the Palm Pre and some Android devices, I see even more clearly the conceptual and mental overhead that the multitasking creates, both for app developers and users. I’m not saying this is “bad”, just that it has a profound impact on the appeal and usability of the device for users who are not geeks or pro users.
In a lot of discussions about usability on smartphones, there seems to be a consensus that the iPhone is very user friendly and easy to use, but I haven’t heard many people talk about why. There seems to be this vague notion that it has to do with the design, that the Apple stuff is just friendlier and looks better, and thus – by way of this design fairy dust – easier to use.
What I think is that the design choice to have the iPhone do one thing at a time is a huge ingredient in that usability fairy dust, and below I try to reason a bit about why and how.
So the situation I want to talk about is one that happens all the time when interacting with anything, really. It’s the question from the you the user looking at the interface and wondering “What do I do now?” The user wants to do something, has some agenda, or has gotten some notification from the device, and needs to take some action. If at any point you don’t know what to do now - you get this little pause in your interaction where you have to stop and think about what to do, rather than getting on with what you wanted to do. The more of these pauses you get when using something, the more lost you feel, and the less of what you actually wanted to do gets done.
When having one window active on the N900, I don’t remember what else I have open. Switching tasks on the N900 first takes you to the window manager and shows you what’s already running - another tap takes you to the desktop or the screen with all the apps. If the task I wanted to switch to is not already running, it takes me at least three taps to change to what I wanted to do: Close the current task/go to overview, tap to get to the app list, tap the right app. In between that, I have to scan the currently open windows to see if what I wanted to switch to is already running.
To show you what I mean, let’s say I’m looking at a contact. If that’s not what I want to do anymore, there is the back arrow to navigate within the contacts application, or the “window manager” corner to get me out of the current app.
Pressing the corner takes me to the window overview of all the stuff that’s open right now.
Here I need to figure out if the thing I want is already running, or if I should tap the area outside the windows to get to the desktop (where the thing I was looking for might be on the desktop) or the app overview corner that takes me to the application list. For a pro user this is welcome flexibility and a bunch of great features in a powerful mobile OS. But I can’t teach this to someone in 10 seconds, and it took me several days of use before I remembered exactly what each tap would lead to and how to navigate between tasks. I still get it wrong, stuttering in my usage of the device, constantly stopping and having to think about what to do now, what action to take to get where I want.
So how does this compare to the iPhone? There is nothing magically user friendly about the initial iPhone user experience - a lot of people who don’t own one and borrow one to make a call or check something on the web are confused at first and don’t know what to do. But as soon as you teach them the one thing they need to know - what the home button does - they are good to go. The key here is that in the vast majority of user interactions, the answer to the user question of “What do I do now?” is always one of two things:
1) It’s on the screen, tap the feature you want. (Tap an item you want to select, tap a button for a feature, tap a tab to switch screen in the current app)
2) It’s not in the current app, press the home button to select another app.
There is nothing to remember except the home button action.
If you are in the Phone app for example, all the main things you can do are visible in the tab bar at the bottom of the screen, and if you drill down to an individual contact for example, the actions you can perform on that item are displayed as tappable things on the screen. There is nothing to remember, and so people who don’t take naturally to keeping a more complex system model in their working memory at all times are saved from feeling lost. It’s such a simple answer to the user question “what do I do now?” that you basically forget that there is some action between you and your next task here once you figured it out.
Multitasking creates some serious mental and conceptual overhead on a mobile device where the screen is basically too small to have several things visible at once. On a computer, you have status bars and docks and task bars, and the screen is big enough to give you an “at a glance” overview of what’s running. On a mobile device, the current task usually needs to use the full screen to be useful, and this places the mental task of keeping track of what else is running on the user, without immediate clues to trigger the memory. This simply makes the device harder to use, and users more often find themselves in that pause position where they are not exactly sure what to do, and get “stuck” on figuring out how to navigate the UI rather than getting on with the task they were aiming for. You can actually draw a little parallell to command line interfaces - they are extremely powerful, but the knowledge about what to type on them has to be in your head since there are no clues in the interface itself.
There are of course lots and lots of people who like the multitasking complexity and the power it gives them, which is why I think Android is doing better and better, and the N900 will be in the pockets of many a big or small geek.
But I think we generally underestimate the complexity added by multitasking on a mobile device, and overestimate what and how we need things to run in the background, modelling it on standards from desktop computers.
A major reason in my view that the iPhone is such a huge success is because Apple found a way to force conceptual simplicity on the right level - basic usage navigating between applications with complex and powerful features is always the same, and something you can teach a person in less than 10 seconds. Behind that, each app can provide as much complexity as it needs or likes, but for the user there is just one basic rule of operation to learn. The rest most people don’t even know about, and it doesn’t matter. They can still use the device, and thus discover in time more and more things they can do. The overall system model you need to understand to have agency and feel in control of your experience with the device is small, and always the same. That empowers users, and makes the technology invisible.
As a final note to this observation, I do think the iPhone is slowly growing out of this simplicity. Apple has educated a whole lot of users into expecting a lot from their mobile device, and they become more and more confident in understanding what these mobile computers can do. Frustration with the powerfully simple interaction model that launched a whole new “smartphone” market might pave the way for sleek, multitasking little netbooks like the N900 or the new Droid. But I don’t think they could exist or grab users without the trail the iPhone has blazed, and I think there is still a lot of work to do in making those devices more invisible.
