/* ---------------------------- HTML STARTS HERE----------------------------------------------------------------------------------------------------- */

Mobile Community Design
Research and design information for mobile community developers.

Saturday, June 09, 2007

The problem with the iPhone

I've been meaning to write an article on the much hyped iPhone, but now that Engadget and Dvorak have taken a pass at it, I figured it was time to for a more detailed design analysis.

Disclaimer: I increasingly think designers don't sympathize enough with other designers during design reviews. It is often possible to think up an excellent design: but given the realities of limited research time, small budgets, marketing pressure, development politics, technical limitations and power within the organization, it is a wonder we ever get anything that is truly designed well. So take this with a grain of salt.

The iPhone appears to have done a lot of things right. They have turned the phone into a small tablet computer. Objects on its screen can be intuitively touched and manipulated. Objects and tasks that inherently have complexity or larger size are reasonably accommodated with zoom and pan with minimal delay. They are integrating local features (e.g. your voice calls) with remote features (e.g. a google map phone number) and making it seamless. They have also integrated the music with the phone so that you don't have to carry two devices (this has been done before, but not in a format that is actually very useful).

This would all be great if they hadn't missed the main future growth area of mobile devices: non-voice communication. Every great killer app on the Internet involved a communication medium (e.g. e-mail, web, Instant Messaging). Voice is old-school. It has many usability and sociability problems. It is slow. It is difficult to use in many situations. It annoys people. It is synchronous (with the exception of voice mail which is expensive, slow, and poorly designed). The new generation of communication and collaboration tools (wikis, blogs, forums, instant messaging, tagging systems) are all about text. So shouldn't a new "visionary" mobile device handle text input well?

As the above image depicts, the iPhone has a lot features requiring extensive, detailed, text entry. It is also missing instant messaging which is heavily used by both teens and in business environments. How are we supposed to be entering all of this text? Via a small cramped (not even horizontal) touch keyboard, without haptic feedback, using one finger at time.

Are these Apple designers insane? Is there such a thing as carpal tunnel in the pointer finger? I think they're going to invent it.

If you watch Job's keynote he demonstrates typing while finding a Starbuck's via search query. He actually prepares the audience for if he mistypes the name of the restaurant; however, he does manage to get it right and you can practically hear his sigh of relief.

Now, to be fair, input devices for mobile devices have long been a major problem. Apple has at least removed the tired 9-key number keypad that is rarely used for anything anymore (who actually remembers and dials phone *numbers* in this day and age?). But touch screens aren't ideal for text entry. Other companies (such as Helio and I-Mate) have been thinking about this problem and have come up with a somewhat obvious, and mildly clunky, although workable, solution of offering two kinds of hardware keyboards in the same device.

A number keypad slides out of the Helio Ocean for the rare times when you need to enter a phone number (This really points to the large usability problem of not being able to easily exchange phone numbers. This is often done by initiating an incomplete call or sending a text message, but it should be easier). This keyboard could be safely removed and replaced with an on-screen touch keypad for it's infrequent use.

The Ocean also has another QWERTY keyboard that slides out the side which supports reasonably usable longer text entry. This is also similar to Danger's Sidekick 3, which was designed with instant messaging real-time conversations in mind. I have used this phone for a year and the keyboard is passable, but definitely not ideal for typing longer e-mails or IM conversations.

So where are we going with mobile devices?

Danger is actually ahead of the game, but not just because of their text-communication-capable phone which has been on the market for at least 3 years. It's because they are designing a 'mobile experience', not a mobile device. They even have a web page discussing the strategy. They don't just offer a phone, they offer services for data backup, system updates, online ordering and viewing of phone data on a remote server via the web.

Apple is headed the same direction with partnering with ATT to improve the usability of voicemail (not worth the time in my opinion, it is fundamentally flawed). But Apple isn't marketing this phone as an experience outside of the phone features themselves. The phone needs to be designed to integrate with your life: all of the activities that you do and the people you communicate with. Helio has direct integration with MySpace, an online community site. They are taking steps toward integrating you with your social network via text.

Text and voice both have their problems. As many sci-fi writers have envisioned, ideally you would be able to think and have it navigate a device, or enter the text that you dictate. We're not there yet (but it is in the works). And for the record, that will still result in a lot of usability problems and design challenges. But until we get to that, what would the ideal mobile device support?
  • It would have a multi-touch screen. One model is small and pocket size. The other is A5 paper size and is similar to a small tablet computer. The larger one supports certain complex tasks in a more efficient manner.
  • It would support longer text entry. Initially a slide-out QWERTY is reasonable. Plug-n-play integration with a keyboard and external monitor should be supported for more complex tasks when a stationary device is available. Wireless integration would be better.
  • Support for group collaboration and communication via software specifically designed for this task. (Get away from the idea of synchronous calls between two people.) Support community participation and notifications.
  • Support sketching and conversion between this and text and voice. Notepads still get used because there is nothing better for taking notes and drawing ideas.
  • Design the whole customer experience and then let it evolve with close attention to user needs. Phone manufacturers and phone companies and data backup third-parties and online community operators can't approach this piecemeal. The user of a phone doesn't know who provides a specific feature (or who caused a specific problem) and doesn't care.
So in summary, Apple is as-usual providing disruptive technology. This will be good for the market. It's a beautiful machine too. But they've missed a huge mobile growth area: text entry and communication mediums enabled by it. iPhone users will rapidly realize how tedious a small on-screen keyboard is to use. Hopefully this will prod the UMPC and tablet manufacturers to reduce the size of their tablet computers and start thinking about them as mobile communication/collaboration devices, instead of desktop computers crammed into a touch screen.

Update: 6/28
Apple has posted a video explaining a bit more about the on-screen keyboard and corresponding usability. It turns out to be reasonably sophisticated, particularly in how it resizes buttons on the fly based on statistical probability of certain letters coming next in words. However, the fact that they have actually added this as a feature indicates that hitting the wrong letter is a significant problem. They state several times that it is "faster than most other small keyboards", but they neglect to say what an upper-bound on WPM is using two fingers, or how it compares with wider hardware keyboard (e.g. Sidekick and Ocean) WPM entry rates. My guess is that it is slower and has higher error rates. They could have improved text entry speeds by widening the keyboard; this would be most easily accomplished by remapping the keyboard to horizontal mode. Since the iPhone already senses it's orientation I wouldn't be surprised to hear about a hack or custom application that allows re-orienting the keyboard for faster text entry (at the expense of reduced room for seeing entered text). It is however nice to see that Apple did put some thought into innovating the touch keyboard, which wasn't obvious in previous presentations.

Update: 7/16
I've had a chance to use the iPhone. Overall it's impressive in many areas, but my predictions about the keyboard are accurate. It's better than an old number keypad with t-9 but it is cramped, slower to type with, and easily and frequently results in errors. This will make the device difficult to use for many social Internet applications.

Here's a usability study of it by User Centric.
"Participants uniformly found text entry SMS and email to be difficult. They were frustrated by the forced use the vertical keyboard and the lack of visibility for editing the middle of a word or sentence."
Finally, participants were surprised (and somewhat annoyed) to discover that horizontal text entry was available only in in the Safari browser."