So, what's next? Do phones become our personal go-everywhere pocket computers, and will displays, keyboards, "vacuum" laptops etc. just become a way to improve the interface capabilities of our personal pocket-computers? Cloud computing, if the NSA doesn't manage to kill it completely, offers a way to at least have our documents follow us around (e.g. Dropbox). There's no reason why a lot more of our computing environment couldn't follow us around via the cloud too. It's easily doable to store everything about your current session on the cloud so that, when you leave work, your computer at home has everything waiting for you just the way it was at work. Having the exact same computer following us around without transmitting everything to Big Brother would have a certain amount of appeal if not for the greater danger of losing everything with our phone! So, what's next?
MS isn't as far behind as Cringely thinks. The surface pro is basically a desktop that's meant to follow you around. It's not quite phone sized, but MS is definitely showing a strong interest in unifying the mobile and desktop user experience. They've done more so far than Apple has, since Cringely's speculations on iOS are still entirely future tense. Still, I'm not entirely convinced the one-computer following you everywhere approach is going to beat the cloud. A little back-tracking by governments on their right to invade user privacy or a credible open-source secure-cloud environment actually taking off could lift cloud computing's chances. Yes, I'm aware it will largely be chance which one wins, and user privacy will likely be compromised no matter which approach wins, either by cloud-data storage or mobile-device backups.
No matter how it happens, things are evolving towards a world where we won't have separate, discrete devices, but just multiple ways of accessing the same persistent computing space. When it comes right down to it, new ways to interface with our computing environment are probably just as important as unifying our discrete computing spaces. Google's work on Glass indicates another possible way to change the game. Free both the user's hands and you have the potential to deliver richer user-input capability. Still, it's not entirely clear to me that this is Google's intention with glass. They may be getting too hung up on the idea of augmented reality when improving user input capability would bring far greater benefits. Augmented reality is nice, but if that's all Google sees in Glass they're going to get their butts handed to them.