I personally believe we are heading towards a convergence of devices and whoever gets their first with the right execution will disrupt the market.

What I talking about is the pluggable computing concept, i.e. Jolla's The Other Half or the PuzzlePhone. Both of these examples have been poorly executed in my opinion. I love The Other Half concept and for me that was Jolla's USP but they've done little more than release a few ambience changing back plates to showcase it. Without community support from dirkvl there probably wouldn't be a TOHKBD. The PuzzlePhone concept is also nice but right now it seems to be limited to nothing more than phone customisation.

A few months ago I bought a $200 Toshiba Chromebook 2, formatted it and installed regular Linux on it. I've done very little to optimise the power usage but the battery life is fantastic. I tend to get 7-10 hours on a single charge with fairly normal usage patterns. It's x86-64 based so it's pretty much a fully functional laptop. I've experienced no performance problems at all. It's obviously not great with heavy loads. Full kernel compilation takes several hours (having said that, you often only need to compile what's been changed so it's still quite usable for this purpose). It's replaced my both my i7 desktop and my HP TouchPad for 99% of my tasks.

The point I'm trying to make is that the technology is already there and affordable for many people. If not, the technology is very close. Battery life is not as much of a problem as it used to be and there are potential solutions such as the 10000mAh battery or Huawei's recently announced fast charging batteries. A single pluggable device that can be used as a smartphone, a laptop, a Psion/TOHKBD style PDA and many more kinds of imaginable form-factors is a very feasible prospect.

Now what about the OS for this pluggable device? I see no reason why it can't be based on Linux, just not based on any mobile or desktop distro that currently exists. I have some ideas about how a desktop/server distro could be adapted to be suitable for use in such a device.

Who would be interested in buying one of these devices? That really depends on the form-factors that would be available. Use your imagination. There's a huge handheld gaming market that might be interested in a gaming dock/module. A laptop dock/module could potentially appeal to desktop Linux users. I say Linux users as it starts there but once ordinary (non-Linux) users see what such a device is capable of, there is a potential for market expansion.