“Don’t you love it? One OS that runs on ALL THOSE PLATFORMS!”
I’m an app developer (some Android, mostly iOS).
Calling it “one OS” is irritating, because there are significant differences between each major version of that OS, differences that make it difficult to write apps that support more than 1-2 versions. People see new OS features, and expect those capabilities be supported in our app, but if they’re not running that latest version then we either have to write those capabilities ourselves for prior OS versions or tell users “TS, upgrade already”. Android suffers severe platform fragmentation, with _lots_ of people running old OS versions, and surprisingly few running the latest version (can’t or won’t upgrade) - Android may “run on all those platforms”, but our apps can only run on a smallish subset thereof, ticking off a very large segment of Android users. (And that’s not addressing all the hardware variations that the software may have a tough time either knowing about or keeping up with.) iOS on the other hand, has a small & highly-predictable hardware base, and _huge_ adoption of the latest OS version (usually >90% within a few weeks, IIRC) - this makes it easy for us developers to deliver new features fast & reliably.
Now, you might understandably express disinterest in developers’ problems. Well, there’s only so much we can do to get the app & updates out the door, and supporting increasingly aged & obsolete/fringe platforms is a matter of diminishing returns (more work for less payoff). And on top of special-case support among for “all those platforms” for “one OS” (overlooking dozens of versions & forks), working under the hood is just plain less pleasant for Android than iOS. You may like the idea of “one OS to rule them all”, but making it happen may or may not be worth it.
More: “all those platforms” don’t pay nearly as much. Android may be on an overwhelming number of platforms, but we have way more (3:1 at least) actual _users_ on iOS. As the food on my table is paid for mostly by iOS users, they’re going to get the priority development - so those on a few platforms are going to get better apps sooner than “all those platforms”.
.
Seems a good point for a tangent Swordmaker & a few others might appreciate:
The iOS toolset (OS library calls, IDE, etc) seems robustly built to implement given features.
The Android toolset seems very much a “me too” follow-up, trying to implement those same features BUT (important part here!) is trying to do it differently just to do it differently.
This is significant to a developer because there is a natural way which some processes flow, which iOS seems to follow ... but Android is trying to reverse those processes just so it doesn’t end up too much like iOS, making it harder & counter-intuitive to develop apps.
Example: if you want to put a marker on a map...
- with iOS, you have the map own the marker (map contains a list of markers)
- with Android, you have the marker own the map (marker contains a list of maps)
This may seem a pointless distinction to most users, but to developers is the difference between a natural vs counter-intuitive flow of information & handling.
So far, I have yet to find a developer who actually prefers Android over iOS.
I’ve developed my share of software (including some well-regarded FEA packages for magnetics design). Writing fairly portable code allows you to stay relatively current without having to rewrite every year or so... In fact, most things that ran on Android Honeycomb will run on the current Marshmallow as well.
So? What’s the big problem? Write with an eye towards extensibility and future optimizations and it’s not that big of a deal. Seriously, I get it - I’ve supported OS changes that big (and bigger)... Architect it right, and it’s not much of an issue.