Column

Handheld Computing Evolution

Network Computing Mobile Observer, June 13, 2007

By Peter Rysavy

We think of our mobile telephones as phones or PDAs, and sometimes as
music players (including the Apple iPhone that launches this month) or
cameras, but let's be clear: These are computers. We just don't think of
them as general-purpose computers, since not even the smartphones are as
"general" as our laptops and desktops. However, there are huge
implications in the increasing computing capabilities within handheld
form factors, combined with broadband capability over Wi-Fi and 3G
networks and huge storage capabilities. I've written on this topic in
the past, but a number of new developments warrant further discussion.
First, I've just completed a large private handset evolution study
working with Datacomm Research, where we interviewed 44 organizations
that provide all the various components and capabilities of mobile
handsets, including operating systems, chipsets, Wi-Fi, ultra wideband,
3G, user input, storage, power, fixed mobile convergence and so on.
Second, there have been some recent industry developments--including
those by Apple and Palm--that portend interesting trends.

The one by Apple was CEO Steve Jobs recently indicating that the company
is putting serious thought into how to allow third-party application
development for the iPhone. You'd figure that Apple, with all its
innovation in computing, would be well positioned to examine the
handheld market with a fresh perspective. So here is Jobs, realizing
that the company's iPhone, which runs Apple's Unix-based OS, potentially
will be able to run a vast number of applications. But at the same time,
he realizes that for the platform to be successful, it has to provide a
superb user experience with respect to stability, ease of use and
enjoyment--something sadly lacking in many smartphone systems today.

What's of interest is that there isn't a clear-cut answer. Do you target
devices like this with a fixed set of smoothly interacting applications,
or do you open up the system as more of a general-purpose platform? My
view is that a prepackaged set of applications may better address the
consumer market in the short term, but that over time people will expect
a range of applications that can only be supported through third-party
development.

One of the key conclusions of our handset study was that while
general-purpose smartphone operating systems like Symbian and Windows
Mobile (and soon mobile Linux) are the base for today's high-end
smartphones, they will increasingly power the mid-tier phones of
tomorrow. There are a number of reasons for this. One is that the
hardware costs to support these more powerful OSs are going down
quickly. Another is that as the applications become more sophisticated,
more complex and more diverse, the ecosystem of software component
suppliers will get larger, simply because no one company will have the
engineering resources to do it all.

For example, sooner or later our handhelds will need security defenses
as thorough as those on our desktops, meaning that solutions will need
to come from companies that specialize in security, such as Symantec.
And that can only happen in systems that have well-defined interfaces,
and it can only happen for a handful of platforms. It's no different
than the forces that have resulted in only a small number of operating
systems on our desktops and our servers. As more applications are
crammed into phones, the proprietary real-time operating systems that
power most phones today simply won't be able to keep up, except in lower
tiers of the market.

Greater handheld capabilities raises the following question: If handheld
devices have all the computing power users need to run applications that
matter to them, why would they want to maintain a separate laptop
computer? The answer is an extension of the trend of people having
separate desktops and laptops when laptops first came out to the now
common use of laptops for all of people's computing needs. The "handheld
for everything" approach is exactly what Palm just announced with its
Palm Foleo. Palm describes the Foleo as a mobile companion device with a
10-inch display and full-size keyboard that is designed to work with
smartphones. It turns on instantly, connects to the phone by Bluetooth,
has Wi-Fi and is expected to cost $500 when available this summer.

I think that we're a little early in the game for this kind of solution,
simply because the number of available applications with this particular
system is quite limited compared to what's available for laptops, which
don't cost much more. However, I do think it's a sign of things to come,
namely that the handheld has all the computing, storage, communications
and entertainment users need, and all that's missing is a larger display
and keyboard for some of the applications. So kudos to Palm for being
the first to suggest this approach, especially since it addresses the
obnoxious problem of having to maintain multiple computing
environments--no easy task as smartphones get increasingly more
complicated.

Finally, there is the ultra-mobile PC trend, another example of the
potential for miniaturizing computers. The issue here is that user
interfaces like Microsoft Windows don't scale well to tiny displays.

Where all this is heading is not completely clear, but I do think one
inevitable consequence will be more powerful general-purpose operating
systems in an increasing percentage of phones.

Return to column index