In terms of revenue, Apple is following the money. iOS has made Apple the wealthy powerhouse that it is today, not OS X. They don't want to lose the installed base or be perceived as just a phone company; OS X gets them mindshare and stickiness in certain quarters that matter (i.e. education and youth) for future iOS revenue.
But they don't actually want to invest much in it; it's increasingly the sort of necessary evil that is overhead, so it makes sense for them to shift to an iOS-led company. In the phone space, where the consumer upgrade cycle is tied to carrier contracts and upgrade cycles, it's important to have "new and shiny" every single year; consumers standing in AT&T shops are fickle people that are easily swayed by displays and sales drones that may or may not know anything about anything.
So the marketing rationale at Apple is (1) follow the revenue, which is mobile and iOS, (2) do what is necessary to stay dominant there, which means annual release cycles at least, and (3) reduce the cost of needed other business wings as much as possible so as to focus on core revenue competencies without creating risk, which means making OS X follow iOS.
It makes perfect business sense in the short and medium terms. In the long term, it's hard to see what effect it will have. It's entirely possible that they could wind down the OS X business entirely and remain dominant and very profitable as a result of their other product lines. It's also possible that poor OS X experiences and the loss of the "high end" could create a perception problem that affects one of their key value propositions, that of being "high end," and that will ultimately also influence their mobile sales down the road in negative ways as a result.
I'm a Linux switcher (just over five years ago now) that was tremendously frustrated with desktop Linux (and still dubious about its prospects) after using Linux from 1993-2009, but that has also in the last couple of months considered switching back. I switched to OS X largely for the quality of the high-end applications and for the more tightly integrated user experience. Now the applications business is struggling (the FCP problem, the Aperture events, the joke that is the iOS-synchronized iWork suite) and third-party applications have declined in quality (see: MS Office on OS X these days) as other developers have ceded the central applications ground to Apple. Meanwhile, the user experience on iOS remains sound but on OS X it has become rather less so as a result of the iOS-centricity of the company.
What to do? I've considered a switch back to Linux, but the Linux distros I've tried out in virtual machines have been underwhelming to me; the Linux desktop continues, so far as I can tell, to be in a worse state for my purposes than it was in 2008. I have no interest in Windows (I have Win7 and Win8 installations in VMs for specific applications, and even in a VM window they make me cringe; just complete usability nightmares).
It's a frustrating time for desktop users in general, I think; the consumer computing world has shifted to mobile/embedded devices and taken most of the labor, attention, and R&D with it. The desktop, needed by those of us that do productive computing work, has been left to languish on all fronts. It's completely rational in many ways at the macroeconomic level, but at the microeconomic level of individual workers and economic sectors, it's been a disaster.