At first glance, that approach makes sense-after all, why carry two pieces of personal electronics when one will do?

Or will it? There has a been a broad range of approaches to building all-in-ones since Eo's groundbreaking products in 1993, followed by Motorola's Envoy and Marco, and too many more to count. None have done well in the market, perhaps because of something I call the single-device paradox: It seems as if we should be able to build the ideal mobile communicator, but no one has, and it may be a long time, if ever, before we get there. Why? Because designing and building any mobile product is an exercise in trade-offs and compromise.

Consider just the big problems. We want a screen that's as large as possible, but bigger screens mean less portability. We need an input metaphor that works while mobile, but keyboards are too small, keypads too limited, and character and voice recognition too imprecise (and, for voice, perhaps too irritating to those nearby). Power and batteries remain a huge challenge, with no good solution in sight.

Human factors also bedevil designers. Most PDA phones can be held up to the head and used like an ordinary handset, but how can one see the screen and access all that exciting new functionality while talking? Headsets are a good solution, but who wants to fumble with a cord when the phone rings? And, of course, wireless networks, with the combination of coverage, throughput and pricing to create mass-market demand, remain a ways off. Checked the prices for GPRS lately?

Still, new designs keep coming. Check out Danger's hip-top model or the PC-Ephone, both sophisticated entries. And the range of products will continue to expand. But for now, the paradox remains.

Got a clever solution? Write and tell me your ideas. But don't forget to file that patent application first.

Craig Mathias is an Analyst with the Farpoint Group (Ashland, Mass.).