I had been waiting for Bruce Tognazzini to update his AskTOG column with some juicy contribution. When I realised that his most recent article talked about solutions to improve the interface of the iPhone / iPod touch, I immediately started reading, with much trepidation. But I have to admit that I was rather disappointed by his analysis, and I find none of his proposals particularly elegant or persuasive. His analysis is interesting and well thought out and deserves to be read in full, of course, he’s the interface and usability guru, not me. In the introductory paragraph, What’s wrong with today’s Springboard 1.0? (Springboard being the formal name of the home screen on the iPhone / iPod touch), Tognazzini makes an interesting premise:
Unlike the Finder or Desktop, rather than giving access to as many apps as you could possibly want, the current Springboard limits you to 180 apps. Paradoxically, this would not be a bad upper limit on a Mac or PC, as apps tend to equal trouble and the more you have, the more trouble you’ll encounter. On the iPhone/iPod Touch, however, 180 apps is terribly limiting as iPhone/iPod Touch apps translate to fun, not trouble, and the more apps you have, the more fun you can have.
The big limitation of the current system of screens on the iPhone is that you cannot sort the applications in a very efficient way:
Yes, Apple does give you the ability to sort out your apps, but that quickly breaks down when you have, for example, one and a half pages-worth of travel apps, a quarter page of medical, 7/8ths pages of scientific instruments, etc. With a fixed upper limit of 11 pages, with no way to label pages, and without sufficient space on pages to hold all one’s apps for that category, things begin to break down. As one approaches the 180-app upper limit, the pages descend into chaos, as new apps randomly place themselves in any available spaces, with nowhere logical to move them.
Once you hit the maximum number of apps, apps just start falling off the edge. This is apparently already happening in sufficient numbers that Apple, in 3.0, released an all-too-typical programmer hack: They enabled users to have invisible apps they can call up using Search as long as they can remember the app’s exact Name. For example, if you have the American Automobile Association app, you have to type in “AAA”. Oh, wait! It’s not called “AAA”, it’s called “Roadside”! What are the chances you’re going to remember that two years from now when your car breaks down?
Tognazzini suggests five alternatives for improving the situation.
The first is to enable users to have identification labels associated with pages:
True, Tognazzini suggests that “Apple could initially show no labels at all so that the new user would encounter no added complexity,” but for me this solution is less effective than it looks. Example: what do you do when applications that fall under a certain category (or label) are more than 16? Suppose we have 22 applications under the label “Dictionaries and reference”: the first screen of 16 applications will have that label at the top, but what about the following screen, containing the other 6 apps? What label will it have? Still “Dictionaries and reference”? Or “Dictionaries and reference 2”? It does seem impractical to me.
And for those with 500 applications this is still a patched-up solution to say the least, as the only form of shortcut is to hold the finger on the label to display a drop-down menu with all the labels; this way the user can reach the screen with the desired application slightly more quickly. However, this creates two further problems, or rather two sides of the same problem: how to handle applications that do not belong to any label? (Because if you have only two IM apps, for instance, it seems pointless to devote one full screen to them). To better take advantage of the functionality the user should necessarily break applications into categories, a rather tedious task. Not to mention the interface itself, which becomes unnecessarily crowded from a visual standpoint, and unnecessarily complicated from an operational standpoint.
The second proposal (perhaps keeping in mind the partial ineffectiveness of the first) is vertical scrolling.
Vertical scrolling would not move page-at-a-time the way horizontal scrolling does, Tognazzini writes, Users, instead, could scroll row-by-row. They could also “throw” the page vertically and have the window scroll down rapidly, in the manner of the address book. This way you would circumvent the limit of the previous proposal (that is, how to handle more than 16 applications under the same label across more than one screen). With vertical scrolling you could have, say, 27 travel apps on a single labelled screen, for example. You keep the most used on top and scroll down to reach those you use less frequently.
It’s more appealing a proposal for sure, but I get the impression that being able to scroll in both directions, horizontally and vertically, will end up confusing users. Then I’d also like to know whether Tognazzini has thought about setting a limit to the vertical scroll or not. How many applications could be included in a vertically extended screen? I also think that by extending the app placement over the two axes requires more steps to reach an app. Imagine this scenario while you’re on the go: to find app XYZ you have to flip three screens horizontally, then scroll ‘a bit’ vertically, but since vertical scrolling doesn’t move one page-at-a-time but row-by-row as a list, finding the application you are looking for isn’t so straightforward because it is more difficult to memorise its position in your muscle memory, so to speak.
The third proposal: User-Controlled Icon Positioning; that is, letting the user place application icons anywhere in a page in some characteristic way, to be able to recognise a certain page at a glance:
Uhm, no, I can’t find this convincing enough. I admit I have no rational objection to it, just a gut reaction: I do not find this improvement particularly elegant or effective. As you keep buying and adding apps, it’s difficult to maintain these distinctive layouts.
The fourth proposal: Containers. A lot of people talk about the problem of organising applications on the iPhone, and the idea of enclosing them in containers (or folders, drawers, stacks, what you will) is usually the first that comes to mind, especially to tech-savvy users who are used to the metaphor of the desktop computer. I have already addressed part of this argument in the past, and I still think that the introduction of folders would be a step backwards with regard to the immediacy and simplicity of the iPhone’s interface. It would be necessary to devise a way (or command or gesture) to create a new folder, then another to visually differentiate each folder, then another to insert / change the name, then another to move items in and out of a folder, to manage a folder hierarchy, and so on.
Yes, I have my own proposal, which I’ll explain in a moment. Now comes Tog’s fifth proposal: multiple links, namely the possibility to have aliases to reach a certain application from different points.
No, no, absolutely not. It’s complicating and confusing things to the highest level. It is, once again, looking at the iPhone and thinking that its interface is extensible in the same way as a computer interface is. Or considered as such. On iPhone spaces and paradigms are different, and one must take these differences into account when navigating its interface. Any addition should consider how much it affects the whole interface, how much complexity it adds, how many new gestures it introduces and whether they’re worth it, and so on.
My modest proposal
I, too, have been reflecting on how to tackle the problem and limitations of the iPhone springboard mentioned by Tognazzini. My proposal takes into account something I find essential: the problem of managing more than 180 applications installed on your iPhone or iPod touch directly affects only a minority of users (we are talking about a lot of people for sure, but considering the global amount of iPhones out there, those who manage more than 180 apps are still a minority). Such a proposal for improving the iPhone springboard must therefore be as non-intrusive as possible for the majority of users, who should be able to keep using their iPhone as usual. The proposal must not complicate the iPhone UI and must not introduce visual clutter or new commands or gestures that are difficult to memorise.
My proposal is the possibility to view the contents of the iPhone in an alternative, optional, transparent way. It doesn’t move or change anything permanently — it’s the ability to switch to a different view that hopefully can help those who have a disproportionate number of applications. My proposal is to create on the iPhone OS the equivalent of this application (or widget):
Say you have an application that is called “Manage Apps” or “App Manager”. Tap the icon and you’re presented with a MobileMail-like UI: apps are displayed in ‘List view’. Want to put all weather apps in a specific folder? Tap on [+] and you create that folder. You can even choose between a set of default folders, possibly with distinctive icons (e.g. reflecting the App Store categories). Folders always appear at the top of the list, followed by any application you don’t want to put into a folder. Folders and apps are sorted alphabetically, and to browse them quickly there’s the familiar A‑Z column on the right. There might even be Smart folders, like in the Finder or in iTunes. User-definable Smart folders, of course, but also with presets to automatically organise the applications following the App Store categories and criteria (a quick solution for those who can’t be bothered to create ad hoc categories).
To manage the movement of large numbers of applications inside folders you may select them in advance and then move them, just like email messages within MobileMail. Or at worst you could resort to iTunes and do it from your Mac.
All this sorting and organisation would happen only within the “Manage Apps” application, and would be an easier, more powerful option for those with large volumes of applications. You would return to the familiar screens of icons and classic iPhone view by quitting the app at any time. So it would still be easy to have access to the apps we use most (the ones we generally place on the first 2–3 screens) and there wouldn’t be any unnecessary complexity.
Those who have a lot of apps would benefit from having the ability to set alternative views within a sort of ‘app browser’. Those who only have four or five screens of apps can easily ignore this new app browser/manager and continue to use the iPhone or iPod touch without any visible change in the UI they know and love. I think it’s an elegant and potentially effective way to manage large amounts of apps, but obviously it’s not for me to say.