The 75% to 15% is only for the last quarter of sales, but if the sales can be sustained, and they probably can (they kept growing so far), then eventually the userbases will look like 5:1, too (for phones at least, considering iPhones are only about half of iOS sales).
And Samsung will not fork. They will lose all the apps, and considering they are nowhere near starting their own app store, it's impossible for them to do that within 18 months. I don't know why this idea comes up every now and then. I guess it's the tech press' love for the melodramatic, and from their point of view it would make for great intrigue if Samsung tried that, but when you look at things objectively, that outcome is nowhere near close to a possible reality right now.
For iPhones, the vast majority of people I know with them use then constantly and have a wide range of apps on there.
Of course, Android still has its power users, who are probably even more passionate about Android than the average iOS user is about iOS, but like you say, they are still in the minority at the moment.
As an example, my company has a mobile app out there for Android and iOS. Over two thirds of the downloads we've had so far have been on iOS, and they typically use the app far more frequently that the Android users.
Microsoft, Apple and Google have this in their DNA. Phone manufacturers, like Nokia and Samsung don't. If they could have build their own Android / their own Google Play, they would have done so already.
Also, companies can mutate 'their DNA'; it is not frozen at birth. For example, the Tabulating Machines Company (nowadays called IBM) has reinvented itself several times, and Nintendo started as a manufacturer of playing cards, and DSM (http://en.wikipedia.org/wiki/DSM_(company)) started life as 'The State Mines' and mined coal, but now is, quoting their site "a global science-based company active in health, nutrition and materials".
Here's another way of looking at it: There is product competition, and Apple is losing.
Samsung is neither low end or high end. They are selling whats suitable strategy for each country and its working damn well.
I am curious how much of an advantage Android has by allowing pirated apps. In price sensitive markets the cost of apps has to be a key factor. If it is a big factor could be ominous signs for Nokia in particular.
To respond to this will require care. They can go with the razor-margins approach, and have their margins eaten away to virtually 0 (i.e. Dell etc). However, this only props up an ecosystem that benefits Google. They could, I suppose, find a way to out-build the entire rest of the market and get decent margins at the very low end of price (but I find that unlikely). Alternatively, they find a way to differentiate, which is most cheaply done at the software level.
Anyone picking path #1 is basically going to have to own the market to make any money in the long term, but will keep decent balance sheets in the short until commoditisation is completed. Anyone picking path #2 has more risk in the short term, but a lot more to gain in the long (e.g. as Apple has).
The majority of Samsung's customers are Samsung customers. They've bought into the brand. They're not about to jump ship to LG.
iPhone users (and "high end" Android users), on the other hand, probably know exactly what they're buying into. These are the users that will determine developer targeting. Two ways I see for Android to decisively win this: either these "low-end" users start to participate in the eco-system, or the rise of low income countries (much of Asia, Africa etc) makes the weight of numbers so far in Android's favour that even with a much lower engagement rate, the money's there. Apple could counter this with "iPhone Nano", I guess, but it would require them to sacrifice a lot of margin.
Outside tip: Samsung are becoming the Apple of Android smartphones, selling the preponderance of the high end handsets. Within 18 months, I suspect we'll see them fork Android to differentiate themselves from the "sell at near cost, make it up on ecosystem" strategy that Google's pursuing now with the Nexus (that reduce Samsung to basically a hardware supplier on razor-thin margins). Either that, or they'll use carrier relationships to ensure these cheap, high-end handsets don't ever really get in front of non-geeky consumers.