OK, this is just plain nuts, but occasionally such are the thoughts of analysts who, after all, are at least partially in the business of synthesizing such scenarios from pieces of disparate data. And this isn't a prediction, at least not yet. I'm a Mac user at least in part because of the stability of the MacOS platform, and, while I won't ever go back to Windows (although the nightmares do seem to be much less severe over the past few years), Linux- and purely cloud-based offerings don't yet have the stability and maturity to meet all of my IT needs. The Mac does.
But considering the effort Apple is putting into iOS and their own Cloud services, and, since it is all about the apps, suppose Apple did decide to emphasize iOS over MacOS (I still think the two will converge eventually regardless) and moved all of their own apps to the cloud (like Google)? Why would anyone need a traditional computer in that case? Sure, there's video gaming, but that works in the cloud model, at least to some degree, and activities like video editing, that are highly data- and screen-intensive. These might be best addressed by a transparent local server that caches not just data but the execution of apps as well, and provides local processing horsepower purely under the direction of the Cloud. Think about that - all of your data and applications in the cloud, transparent local caching of both if speed is important, but otherwise everything is resident and managed in the cloud. You could use any subscriber device (well, any Apple subscriber device, although the model here doesn't purely require that) you like, and all of your context, not just your content, is available anytime, anywhere - the mantra of mobility. BTW - don't even think about patenting this. I'm intentionally putting the idea into the public domain here.
All of this, of course, fits nicely into my vision of infocentricity - computers, networks, and applications disappearing into the Ether, with an emphasis on convenience, low cost, and ease-of-use. Apple, of course, has never been about low cost, but they can make a very good case for the other two elements and at least cost-effectiveness. So if any major firm could move us in this direction, Apple would get my vote. And, as their restless shareholders would attest, Apple really does need to do something big - while having a family of handsets rather than simply serial introductions of minor improvements to a premium standalone product, a cheaper iPhone alone won't do the trick. Apple's making a ton of money from the Cloud already, and I'll bet that future product offerings across the board exploit that capability and position. Oh, and yes, subject to a transition period, I'd absolutely move my Mac infrastructure to what I've described here. In fact, assuming iOS becomes more MacOS-like, I can't wait.