And it will be as long as keyboards are the fastest way to type. So you connect a bunch of junk to it and you have what's kinda a laptop but less neat. Different devices do different things well. The biggest misconception in technology is that devices are converging. Ask yourself do you own more or less devices now than 5 years ago?
I know you've clarified your point, but I feel the need to interject on one point here.
You are right, the number of available devices is growing, and the amount people own may be growing as well. This is primarily because, despite convergence of features, it's more to add a feature to the box and advertising than approached from an actual use case scenario.
I mean, "smart TV" seems to be the big talking point right now in that sector, but people aren't actually using them, because they already have a cable box off to the side that is their primary interaction with the device itself. So all of that converging technology just simply goes unused except to illicit a "wow" factor when you take it out of the box.
The features are all there, but there's a certain... discomfort having to rely on 2 different devices and interfaces. For smart TVs to be truly convergent, TV makers should have been dropping cable box functions into the TVs themselves and allowing cable companies to drop some firmware in to get them running, or a hardware module you can plug into the back, as they already do for certain models as I had seen on TV. THAT is true convergence, because it doesn't just merge features, it merges use cases into a cohesive whole.
This is why there is still fragmentation: tech companies simply don't understand that features without a use case are useless.
And this includes Apple products, as well. I have long held the belief that the Apple TV should become a full-on set-top box with all the Apple TV's features overlaid onto the standard set-top PVR experience. THAT would drive adoption, much like integrating the iPod/iTunes ecosystem into phones with the iPhone was a primary driving motivation for early adoption. People never left home without both their phones and their iPods, so the use case was there, and Apple was able to capitalize. That is an example of convergence done right.
And this is where Microsoft is struggling.
The use case for the iPad was very simple: people wanted something in between the powerful PC and the handy but limited iPhone. PCs and Macs both are complex by layers of abstraction, and that is entirely by design and their nature. You can do ANYTHING with them, and that's part of the boundary to adoption and part of the consumer stigma applied to them.
Before the iPad, people who wanted a Mac or PC would say "it's just too much computer for me, when all I mostly do is check my email and browse the internet."
Netbooks were a step in the right direction but didn't gain traction for this reason: despite being "less computer", it still had all the layers of abstraction that made it appear to be the same as any other computer.
However, the very user-friendly iPhone/iPod touch/etc. are clearly designed as on-the-go devices, not something to enjoy in one place.
There was a use case that had been waiting to be filled since the PC first became a big must-have thing back in the 90s and became more apparent when there was another marker denoting where the middle ground that needed to be achieved was.
Meanwhile, for the sake of convergence and so as not to give up on the "power user" stereotype of Windows product users, we have Surface, which suffers from the same problem that netbooks did: despite a lot of ease-of-use interface wrapping, any time you use it as advertised, the wrapping comes off and the layers of abstraction show through again and make it seem like something more than what people are looking for, or conversely, not enough.
It's not enough computer for people who need that and it's not enough tablet for people who need that.
They made a convergence device without a use case. Currently there simply is not enough of a use case for a quarter-measure intermediary between a tablet and a PC (or not enough to build a market around it, anyways). Most people are quite fine to go to either extreme, which is hardly the same extreme as it used to be when it was just full-on computer to a smartphone.
And make no mistake, consumers have set smartphones and computers as the borderlands for their multipurpose work/entertainment devices, and anything too far in either direction outside of it is simply not of enough interest to the modern consumer to build a market around it. Apple TV as a $99 TV relay for all your other devices is a great thing... but the use case for another box attached to the TV with ANOTHER user interface you access separately is never going to be mass market. It needs to cannibalize another use case or find a new one that's in need of exploiting.
Microsoft tried to find a new use case, and in turn made the Surface the answer to a question no one asked.