Re: Homebrew PCs are simply becoming more common
Exactly. PC sales have long been driven by the obsolescence cycle. There are not many categories of durable goods that have to be replaced every three years even when they're working perfectly well, but that was the PC market through the 90s and the early part of the 2000s. That short product life cycle came to be accepted as the norm for PCs (much as it is accepted as the norm for smart phones now), and now that PCs are typically kept until they cease to function (like most things), we're hearing about how they're dead now. Oddly, no one says toasters are dead, even though they usually are only replaced when they stop working.
Now that a six year old PC performs perfectly well for most things, there is no longer a compelling reason to upgrade if your PC is functioning as intended. It used to be that you had to upgrade or the world would pass you by, with your old machine failing to run the software everyone else uses. No matter how spacious hard drives get, how fast CPUs become, or how large a system's RAM, the developers of software have always been more than willing to use all of those resources up to give the user a few more bells and whistles.
That progression has slowed greatly, and it might be argued that it has gone into reverse in some ways. For years, everyone had to have more power, more memory, more storage. Now that we have Windows 10 promising a new software paradigm wherein a single "app" will run equally well on a mobile device and your desktop PC, the performance bar has been lowered significantly. Mobiles are far behind desktops in every performance category... they have far less powerful CPUs, much more limited RAM, and several orders of magnitude less onboard storage.
What that means, of course, is that an "app" that is capable of running on a smart phone (even a high-end model) or tablet will only be using a tiny fraction of a PC's power-- and with the Microsoft app market in its nascent stage, it is doubtful that any new apps developed for the platform will require anything more than a mid-range mobile device.
If that is the new development target, there's even less reason to upgrade. A high-end smartphone or tablet of today is roughly equivalent to a midrange desktop from about eight years ago, so that's all you would need to run an app for a modern mobile device. That's where the "mobile first" strategy of Microsoft inevitably leads us on the PC desktop-- eight years into the past.
Of course, this is all predicated upon the idea that Microsoft's "neither fish nor fowl" OS actually does work in one of its primary purposes, which is to force a well-stocked Windows app store into being (at the cost of an inferior experience for PC users). I seriously doubt it will work, personally. I think expecting PC users to settle for uglified versions of the programs that would have been state of the art more than eight years ago (with the added bonus of being designed around tiny touch enabled displays, with all of the goofy, ridiculously oversized controls that entails) is doomed to fail, particularly when a robust market of native Windows programs already exists. Absent the free built-in market of desktop Windows 10 PCs promised to them by Microsoft, app devs are left once again to ponder whether it is worth it to produce Windows 10 mobile apps when the market is so small and future so uncertain.