Android and iOS have won the first world, it's done.
Some contenders (Microsoft, Blackberry) try to grapple some market share, but unless Google or Apple do something very stupid it seems doomed to be a very hard & slow progression.
Even with Microsoft throwing moneys at every major app makers, the windows marketshare make it very hard for windows phone development to be worth it.<p>If I had to make a gamble, I would say that Google is going to take over the remaining 5 billion people too. Apple and emerging markets don't go together at all. Microsoft & Firefox are targeting these markets but I feel that with Android One & the changes made in 4.4 in order to have acceptable performances on low end devices, Google has the formula to win. Nothing is done yet though.<p>Google has also decided to conquer other form factors. In order to do not frighten its partners, these products are called Android (and it is indeed the Android tech stack behind them). Google Wear is the first wearable device that makes sense to me. That does not mean that it will be a big hit, but it is a good start.<p>All these new hardware initiative (car, health, wearable, home, ...) from Apple & Google are very impressive. They have many ideas in common and both only work with mobile devices made by their company. There are probably at least some technical reasons behind that move but it is still very disappointing. I don't think that anybody in their right mind could be entirely satisfied with a car or home automation system that dictate which brand of smartphone you are going to be able to use.<p>As an Android dev, I am confident in Google's capacity to provide the tools I need. Even though I don't think that Google made ALL the right choices (lockscreen widgets still look like a very weird choice to me, they seem to be disappearing with L though), my most critical needs were fulfilled with the last i/o round of announcements with improved performances, first class design ('Holo' was really starting to look bleak compared to the other OSes) and the replacement of the terrible ListView by RecyclerView.<p>There are a couple of parts where Android is still weak from a dev point of view :
-Even though Google is visually making many efforts in that area, the company does not have the tool making expertise that Apple or Microsoft have. Using IntelliJ was a very good move, the JetBrains guys know how to create an IDE. Right now we are in a weird phase where Google is slowly giving the finger to Eclipse and forcing us to migrate to Android Studio but it is still a bit immature. Staying with Eclipse means not benefiting from many new things, while migrating to AS means dealing with its bugs..<p>-Java is not as bad as its reputation can make you believe it is but there are real pain points (in the context of Android development at least, I don't really know the other sides of Java development to be honest). For example, Enums in java are syntactically awesome but for big codebases, they can contribute to bloating up your app if you use too much of them (each time you use an enum for the first time, all its members need to be instantiated and enum also generates some big methods). The alternative is to use bitfield flags or simple ints. It is better from a performance perspective, but it is absolutely ugly to use and does not offer any type safety. Some of Java weaknesses can be overcome with things like EventBus or RXJava but some are too deeply engrained in the language in order to be solved.<p>-Again, Google does not have enough tool-making expertise. I should be able to override the default widgets colors with a simple line in my layouts. This has just been added with the L release and I am not even sure that it is a back-compatible change, this should have been an API lvl 1 feature. There are many other examples where hundred of lines of code are necessary whereas it should be doable in one line with a good framework.<p>-Fragmentation is also not that bad but it is a reality you will have to work with. Crashlytics, bugsense and other crash reporting solutions are your best friends in order to detect and fix major fragmentation issues. Anecdotally, so far Samsung has been the biggest source of such issues for me.<p>-According to Android engineers, Java is here to stay. Since they don't communicate on their future plans, it is impossible to know whether they are working on alternatives or not but it would be welcome.