Designing for touch UI is a tricky business. The Apple iPad HIG emphasizes minimalism: fewer buttons, less noise. When executed well, the result is a clean, elegant and intuitive UI but a potential side-effect is reduced feature discoverability.
Karel Vrendenburg describes a discoverability problem he encountered in the Safari browser:
…several parts of Facebook simply didn’t work… the problem was scrolling… when I tried to swipe down on the list, the way you scroll down everywhere else on the device, the page would move slightly but the pop-up layer would move with it but the list of friends would stay put.
Karel’s perfectly reasonable conclusion was that the site wasn’t fully compatible with the iPad. It wasn’t until a friend told him about the two-finger scrolling required to navigate iframes that he was able to use all the features on the site. I’ve had similar experiences with other Web site features, such as mouseover states on buttons and menus (you have to tap to get the mouseover, then tap again to take the action).
Jacob Nielsen chimed in on the “low feature discoverability” of the iPad in his less than flattering take on iPad usability:
iPad apps are inconsistent and have low feature discoverability, with frequent user errors due to accidental gestures. An overly strong print metaphor and weird interaction styles cause further usability problems.
In a post titled GESTURAL INTERFACES: A STEP BACKWARDS IN USABILITY, Don Norman and Jacob Nielsen state:
The usability crisis is upon us, once again. We suspect most of you thought it was over… Well you are wrong… Discoverability is another important principle that has now disappeared… Bold explorations should remain inside the company and university research laboratories and not be inflicted on any customers until those recruited to participate in user research have validated the approach.
A crisis! Sounds pretty bad, doesn’t it? And yet, the iPad sold 2 million units in 60 days and continues strong sales. It also touts the highest customer satisfaction index ever tracked. The market literaly can’t seem to get enough of them. Clearly, these “experiments” with touch UI are not hurting sales or customer satisfaction.
How big a deal is this discoverability issue?
Time for a little perspective. It’s still very early days for touch UIs and clearly, not all touch gestures are intuitive across all contexts. Is the answer to this problem simply to place more affordances in the UI or, as Nielsen suggests, make touch UIs more Web-like, with home pages and back links? Probably not.
Adding more affordances ultimately leads to more clutter and complexity. Simplicity, beauty, direct manipulation and delight are key reasons users are so enamored with touch UIs and they are just as important to the user experience as discoverability. Making sure everything is easily discoverable will almost certainly have a negative impact for these aspects of the user experience.
Scott Berkun has a great post about the myth of discoverability:
The trap, and the myth, of discoverability is that in any design, not everything can be discoverable.
What’s a designer to do?
A natural tension exists between discoverability and simplicity. Nielsen would have us focus on discoverability to the exclusion of simplicity, but I’d argue that’s an overly simplistic approach that doesn’t take into account the whole user experience. So, where is the line between simplicity and discoverability?
Stack rank your features
What is the primary purpose of your mobile app? It should outweigh other features and functionality by a wide margin. Everything else takes a back seat.
Editing is key. You should trim out anything you don’t think is absolutely necessary. Does your app have more than one primary purpose? You may want to rethink your design or create two apps.
Use gestures to get to less frequently accessed features
Should you use the swipe to delete gesture or place a delete button in your UI? If delete is an infrequent task (and it usually is) then providing swipe alone might be just fine.
Leverage Physicality & Realism
I’ve talked before about the importance of iPad’s physicality and realism guideline which helps users understand how to interact with your application. Taken to the level of kitsch, it can be more noise than signal, but applied thoughtfully, the physicality of your touch UI can help users intuit how to interact with it.
The iPad Photo app depicts collections of photos as a stack, inviting the user to manipulate the object as if it’s a real stack photos.
But make sure you don’t misuse physicality and realism. In another Apple example, the iPad Contacts app is depicted as a book of pages, yet swiping the page does nothing.
Pay attention to the little things. Invite exploration by making more items in your UI interactive. Again, these items should not be mission critical to the core value of your app, but should provide opportunity for surprise and delight.
Touch UI design is evolving
I’m certainly not saying that discoverability problems don’t exist. There are a lot of touch UIs out there that are poorly executed and could be improved. But again, touch UI is still in its infancy. When you think about it, it’s pretty amazing that people have responded so positively to touch UI despite the fact that it’s still a maturing interface. Imagine what the future holds once we really figure out what we’re doing. Touch UI is evolving.
Usability evaluation needs to evolve too
Traditional usability testing is great at pointing out discoverability problems in a time-compressed and artificial setting. In fact, discoverability issues are probably the most common output of a usability test. They are easy to identify and, in the abstract, easy to solve, especially when you’ve got “all that screen real estate” to add links, buttons and descriptive text.
Nielsen’s method of handing someone a tablet and saying “go” isn’t very realistic. Users are often more persistent and resilient than a traditional usability test indicates. If users were as tempermental in a real-life setting as they are in the lab, Apple would not have the success it has today. Before we decide that touch UI needs to conform to Web conventions, let’s spend some time determining if there is a more effective way to evaluate them.
How do we measure happiness?
How do we measure happiness in design? Avoiding frustration is no longer a differentiator, it’s a minimum bar. What are effective methods for determining how seductive and pleasurable a UI is? Simplicity has a large part to play, as does curiosity and delight. Done well, discovering a feature you didn’t initially know was there can be an enjoyable experience. Just look at all the games that have you unlock functionality as you progress through them. From a purist usability perspective, these hidden features would be a discoverability issue that needs to be dealt with.
It’s not the end of the world, I promise
Okay, so Touch UI isn’t perfect and neither are our methods for evaluating them. But it’s not the end of the world. With a healthy dose of user-centered design, a dash of common sense and some good, old-fashioned fun, we can create applications that change the world. Experiment, tweak and, for the love of God, release! Evaluate, but don’t keep your products in the lab forever and don’t be afraid to take risks. Touch UI is here to stay and I for one think that is a great thing.