Originally Posted by Michael Scrip
But what about non
For instance... Android "L" will introduce 5,000 completely new APIs. (some Google... some not)
So in order for a developer to use the new non-Google Service APIs... the user must have Android "L"
Instead... developers will forgo most of those new "L" features until enough people can use them... right?
I'm just trying to wrap my head around this.
iOS 7 is installed on 91% of its devices... which allows tons of people to take advantage of the new features introduced in iOS 7
But then I've seen comments saying "85% of Android users are using 4.X so it's not that bad"
That's fine if you only use the features introduced in Android 4.0. But if there is some amazing feature available in Android 4.4... your audience isn't so big after all.
That's the fragmentation problem with Android. There are API features that won't be utilized by many new apps until maybe a year after they land, when the number of phones running with support for a specific API version reaches a critical mass. Of course, legacy apps can maintain versions for different API levels, supporting both 2.2 and 4.4.4, for example, but with different .apk installations. Not many developers are set up to maintain 10 or 15 app versions for various APIs, deciding what new features to try to port backwards like that, and what to just leave out. iOS has a clear advantage here, because there aren't as many API levels out there in popular use.
That said, the new Android and iOS versions bring so much more than just the APIs. The look and feel, process scheduler, memory management, etc. On top of that, the Google apps themselves are my primary apps, and they keep versions around for all devices, especially the latest and greatest. So I can get a new phone with "L" in Oct or Nov this year, and be confident that all the Google apps will make full use of the new OS, and that my other primary apps (Strava, Waze, and Netflix) are probably already there as well.