I've been working on mobile devices and systems for more than 25 years, with a short break in the middle to spend a little time with workstations and supercomputers. My first project was the Grid Systems Compass Computer, which was also the first laptop computer.

This was an incredibly tough engineering project. Every morning, there'd be a new challenge waiting for us. The power supply wouldn't fit. The modem cost too much and was taking too long. With no vents or fans, cooling was a constant issue. Nonetheless, we solved every problem, but always with numerous compromises along the way that had to be made.

Designing and engineering any mobile product is an exercise in compromise, even to this day. Consider the range and number of issues that face product designers and marketers -- size, weight, cost, shape (PDA, candy bar, flip, slide, etc.), display, user input, processor, memory, time to market, cost, ruggedness, software environment and did I mention cost? I like to draw an analogy to trying to solve a mathematical problem with more variables than equations - there will be many possible solutions, and it's often hard to tell which is optimal in any given case.

To make the problem of designing and marketing mobile products even more difficult, the shelf lives and overall lifespans of these products tend to be short. Styling is a major concern and is difficult to evaluate (with a major impact on cost). The competition can be fierce, and the market is filled with a large number of products that are poorly differentiated and have poor documentation and support. Since mobile products often have poor returns on investment, due to competition driving down prices, vendors and carriers don't want to invest in documentation and support, which are often poor as a result.

A good counterexample to all of this is Research In Motion's BlackBerry product family, which has been around for more than 10 years and has a loyal following. The BlackBerry does just one thing very well - mobile email - and is largely unsuitable for broader applications. The BlackBerry browser, for example, is just awful, as are the other applications that ship with the device. The browser is not very up to date with regard to modern web standards, so it doesn't do a very good job with many web sites. Its ancillary applications, like address book and calendar, are primitive when compared with those on other smart phones.

The single device paradox

Some time ago, I coined the phrase "the single-device paradox" to describe a core challenge facing mobile computing and communications, and it's still with us today. In a nutshell, it's likely impossible to build a single device that can do everything because, as noted above, there are too many competing interests in designing, building and using mobile communicators.

Let's consider a few examples. If we're just building a phone, the challenges are relatively simple. There are a lot of good cell-phone handsets with excellent usability, battery life and convenience, especially in terms of a small footprint. But suppose we want to add web access to the phone. Sure, one can get the web, sort of, in services such as Verizon Wireless' Mobile Web 2.0. But that's not really the web. It's the mobile web with its restricted functionality, which is often desirable and necessary on small cell phones, because most Web sites won't work well on tiny screens. For usability, we need a much larger screen and probably a micro-keyboard for more effective messaging and email. Now not only has the functionality increased, but also the size, weight and cost. Then battery life becomes a challenge, and on and on.

You get the picture. We're going to continue to see innovation, but always with a degree of compromise attached. You're likely to try lots of mobile devices, and all of them will prove unsatisfactory in at least one dimension. Everyone, for example, is lusting after Apple's iPhone [Not me - Editor]. But it has no physical keyboard, and touch-screen keyboards have been notoriously fickle. You should be able to attach a Bluetooth keyboard, but that's another accessory to carry, and convenience would take a hit. I'm already getting anecdotal concerns about battery life. Plus, at US$499 for a 4 Gbyte model and $599 for an 8Gbyte model, it ain't cheap.

So, while I don't think we'll ever get to a single device that can really do it all, you should expect continual progress in mobile devices. Competition is fierce, and there's a lot of incentive for new ideas: The total available market for all handheld communicators is in excess of 1 billion units per year. At least one design, by definition, will work for you. While all mobile devices will always represent compromise, such a compromise is always better than nothing at all. Happy shopping!

Craig J. Mathias is a principal at Farpoint Group, an advisory firm specialising in wireless networking and mobile computing. This article appeared in Computerworld.