conference indicated that mobile devices are becoming increasingly useful and are poised to move beyond early adopters and enter the mainstream. The conference highlighted several key changes and industry trends:
iPaq is now the mobile device of choice
and was the platform for almost all new services. Last year, most start-ups based their systems on WAP phones, but virtually all presenters now see WAP as a doomed technology. Think of the hundreds of millions of dollars that could have been saved last year if the VCs had bothered running a
WAP usability study
still had a presence at the conference, though it was used by dramatically fewer services than last year.
The PC is emerging as a personal server
that supports a user's mobile devices, often through its wired Internet connection. For example,
downloads audio content to the PC based on user preferences. It then transmits the audio files wirelessly to the user's car when it is within range. Although it cannot support real-time news, it does offer a virtual broadband connection to the car and lets users specify desired audio content through the PC's full-screen user interface.
Humans add value to the network
both presented ingenious ways of injecting full intelligence into a mobile system. Users simply speak their information request; the system then compresses the audio recording into a data file and transmits it through the Internet, to locations where highly qualified labor is virtually free. For example, a human expert at web searching could research the user's question and transmit the answer back for less than a dollar. Using text-to-speech synthesis for the response reduces intercontinental transmission costs even further and lets the call center expert compose replies by cutting and pasting information from Web pages.
WiFi (802.11) is now the wireless connectivity of choice
and was used by almost everybody at DEMOmobile'2001. Last year, Bluetooth was prominent, but this year it was almost gone.
DEMOmobile also indicated that venders are increasingly
aware of transmission costs
and the need for a smooth hand-off from one network to another. For example, NewsTakes bills users by the packet and lets them select the richness (and thus the expense) of their multimedia object versions. If you're on an expensive wireless connection, maybe you don't need full-motion streaming video after all.
iConverse gives users a seamless service even when they're moving between network connections. For example, you could connect through a cheap, high-speed WiFi connection while within range of an access point, then hand off to an expensive, low-speed telephone company wireless service when you're out of WiFi range, then jump back to WiFi when you approach another access point. This is great for users, but will further hurt telephone companies as their expensive 3G licenses will pay off only when users are in between WiFi access. Most high-value mobile service will be provided by WiFi services at hotels, airports, espresso bars, office buildings, and so on.
However, we're still missing a WiFi charging model that doesn't force users to subscribe to each access point. I predict that we'll soon see national or international roaming services that let users have a single subscription and pick up service from any access point, while payments are handled invisibly on the back-end.
Danger Device: Double Dualities
The most eagerly awaited product announcement was
's new mobile device. The company was trying to promote the overly cool name
(though it's hand held), but everybody else called it the Danger Device.
The Danger Device is a two-in-one design, in two different ways:
First, it's both
a personal digital assistant and a cellular phone
(through a headset).
No more cell phone
to carry in addition to your data device. Voice is data.
the form factor
has two different configurations. Initially, you see only the screen and the device is the size of a fat deck of cards. As I have often requested, the entire surface area is devoted to screen space, with the exception of a few thumb-operated buttons. This base form factor works fine for checking appointments or incoming email. If you want to
to email or interact with a data service,
the device twists open like a Russian snuffbox
and reveals a keyboard under the screen. In this configuration, the device is obviously twice as big and doesn't fit in the pocket any more, but that's okay: You twist it back to its compact size when you're done typing.
Danger Research has done many things right. Even though the keyboard is small, it works fine for two-thumb typing. It is also slightly bigger than the Blackberry keyboard and, based on a five-minute trial, I think Danger may support slightly faster typing. Of course, I'll reserve my final judgment on this until I've run a controlled typing study.
Even though I like having an A-Z keyboard on my mobile device, I am not convinced that
are the best long-term solution. Better handwriting recognition is coming, and Pocket PC 2002 already has reasonably good recognition that doesn't require users to learn a new alphabet. Also, with Danger, Blackberry, and most data phones, you have to use a track wheel to move around the screen. This feels very unnatural and constraining compared to using a direct manipulation device like a pen or a mouse, or simply poking your finger at the screen.
A more certain Danger Device drawback is that it does not use the default screen to display a
home base view
of the user's current data.
, please. Or, better yet: show prioritized lists of upcoming appointments, important later appointments, and the most significant email, with brief summaries of urgent items. When you pull out your PDA, you should not have to drill down into several separate application views to get an overview of your status.
The design has a few other weaknesses as well. For example, it treats SMS and Instant Messaging as two separate applications, even though they obviously serve the same purpose. All previous studies have shown the benefits of unified messaging. Splitting communications according to signal-carrying technology is not helpful to users.
Automotive Devices: Complexity = Death
Many of the new mobile services are intended for use in cars, mainly by drivers. This changes the equation for usability. Typically, we are concerned with mundane matters like increasing website profits, improving employee's intranet productivity, or reducing the training and support costs for software applications. For in-car interface design,
complexity will lead to deaths
By current estimates, driver distractions cause at least 10,000 deaths per year in the U.S. alone, and mobile user interfaces may increase this number substantially unless usability is given high priority in device development.
Some people naively think that accidents can be avoided simply by using voice-activated user interfaces. However, studies of cell-phone-related accidents seem to indicate that hands-free is
than manually operated phones. In one
, simulated use of a hands-free cell phone increased the driver's probability of
missing a traffic signal by 150%
. Even when drivers did see the signal, cell-phone users had a
10% slower reaction time
than a control group who were listening to the radio.
Speaking to other people by phone places a sufficiently dangerous cognitive load on drivers. Trying to fathom an interaction design, track voice menus, and remember state transitions could be substantially more distracting.
That said, bringing Internet content into cars can be wonderful.
transmits audio content to the user's car overnight. One day, your car may read the latest Alertbox to you the morning after it comes out. Or you could listen to your favorite columnist from any of the world's newspapers or your favorite NPR show that airs at that inconvenient, non-drive-time hour. Such systems will eliminate traditional local radio channels that never play what you most want to hear anyway, since they are broadcasting to broad, generic audiences.
I'm not opposed to mobile technology and content in cars. The key is to design such services with the utmost simplicity so they don't exact a price in terms of increased accidents.
Update Added March 2002
I saw a updated version of the Danger Device at Esther Dyson's
conference, March 24-26, 2002.
Danger has fixed several of the usability problems I noted in the above review.
For example, the home screen now displays an overview of new email so that you don't have to go into the email app to find out what messages you have.
Even though I like the usability improvements, the most important conclusion from these changes is that the company takes the user experience seriously. They actually bother changing their design based on what they are told by usability. This bodes very well for the future quality of the device.