Even if OS-wars don't matter as much as they used to, Android vs iOS is shaping up to be an interesting one. Android has taken the lead in features. Every new release of iOS is packed with things that Android has had for years. Android is still lagging behind in polish, and iOS is utterly dominant when it comes to gaming.
Google spends more than twice what Apple does on research (that's after Apple bumped up their R&D budget too!). Apple is going to be playing catch-up in terms of features for the foreseeable future. Google has taken some steps to improve their polish, such as project butter, but that bright future is most unevenly distributed. Google's attempts at polish will likely be thwarted by the same thing that scares off game-writers: fragmentation.
In the near future, I think Android will take the lead and become the better OS, except for gaming. Google is putting too much effort into polish not to catch up quick (with their premier devices at least) while Apple isn't doing enough to close the feature-gap and is actually falling further behind (e.g. NFC, wireless charging, etc.). If Google ever manages to solve their fragmentation woes that'll basically be it for the iPhone.
Windows 8 phones are, of course, a bit of a dark horse. WM8 appears to be a nice mobile OS that is going to come on some very nice hardware. It might be WM7 all over again, or it might actually catch on. I'd love to see another front opened up in this war!
The fact that Android has more features than iOS doesn't really mean anything. There were existing smartphones that had more features than the iPhone at it's launch. It was the quality of the implementation of these features that made it stand out.
Now clearly Android is a far more capable OS than the smart phones OS's of 2007, but in some ways it still lags iOS. One that matters to me for example is the lack of a good low latency audio API. Audio production apps are very popular on iOS and some of them are ground breaking. This matters far more to me than NFC or wireless charging.
In addition, although Google plans to add (or maybe already has) a better audio API, there will be a huge lag before there are enough devices with the new API available, so most developers won't bother porting.
On the other hand, it's easy for Apple to add wireless charging or NFC since they're mostly hardware.
To address the original post, I don't think this is like Mac vs Windows. Apple is making a large share of the industry profits, and unless Android developers start suddenly making considerably more money than iOS developers it still has greater developer mindshare. Also, the premium Apple is charging for it's devices is less than it was for the Mac vs PCs duing the 80s/90s, and most of it is swallowed by the carriers.
I think it's clear Apple isn't going dominate the sales numbers, but I don't see any signs of iOS dwindling to near obscurity like MacOS did during the 90s.
Low latency audio is actually really hard to do on a mobile device because low latency and low power use are literally the exact opposite. I wonder if Apple dynamically adjusts things to only have low latency when requested or if they just burn power here and make it up elsewhere.
In any case, Google just needs to swap out AudioFlinger for PulseAudio and they'd have this taken care of. Collabora have even done most of the work for them already.
I'd sooner bet my money on FirefoxOS than Windows 8. Can't wait to develop natively with little more than what gets me on browsers. And I'd similarly sooner bet on FOS going head-to-head with iOS with a non-fragmented system than on Google getting his act together.
Especially since Android is expanding to so many other kinds of devices, it's fragmentation is it's biggest trait and I think it's a pretty hard trend to revert. That's like asking Internet Explorer to become the zeitgeist standards-compliant browser--they have taken too long, much as they try to revert (and they are) other browsers have already taken over it's share and will continue to (especially now that Microsoft guaranteed the non-longevity of it's desktop OS going forward).
To be clear, I don't mean they can't solve the fragmentation problem, only that they have left this breach open long enough for a competitor to become the new cool kid before they could hope to pivot their behemoth operation. That's my prediction at least.