A great mobile Linux device starts with a set of requirements.
Those are how questions. They're about the developer.
Let's look at it instead from the user's point of view. My point of view.
Here's what I've been seeking for the last 25 years, ever since I saw my first "portable" PC, a Kaypro II I once lugged up Peachtree ridge to do some work at Newsweek (just to prove I could).
It's a reporter's notebook. (Judy Gnant says this one belonged to Dan Dahler of ABC News.)
It should do everything a reporter needs done on the job. It should take notes. It should take pictures, even moving pictures. I should be able to write on it and do my research. Then WiFi for the research and to get the story out.
Some of this is easy, simple to visualize, and the parts are all available. A camera lens in the back, the front mostly a screen. A plastic flip-back that lets you set it upright, on its side, and a light button at the "bottom" (the left side when it's held upright) which can display a working keyboard on any flat surface. A pen interface acting as a mouse, with a browser, calendar, and Open Office installed. A phone.
Now here's where we set it apart. I want IBM ViaVoice. Not so I can talk to the thing, but so I can record interviews and have my source's words translated into text, text which I can edit or cut-and-paste into a story, or save as voice files to accompany it.
That's what I want. I want something I can whip out of a suit pocket which does what I need done on the job. Something whose size and shape are defined by the work I've done all my life.
Give me something that will collect my notes, let me research and write my stories, and oh, would it hurt if the rechargeable battery lasted more than an hour, and were easy to replace?
Those are my personal system requirements. My guess is you have your own. What are they?
That's how we start toward mobile Linux dominance. Start with users, not designers, and you're on the right path.