UNIX Fragmentation 2.0: Android
First of all, let me introduce you briefly to the UNIX fragmentation: Before Linux was around (1991 Linus released the version 0.01 of the Linux kernel) there was a huge problem on the UNIX-based operating systems, since they were all proprietary and each OEM had their own version of the OS. It was common that some softwares would not be able to run on all the UNIX variations which caused a lot of trouble at the time since if you were a company buying computers you had to know if the computer you were buying was compatible with the software you had to run. You can read more about the history of UNIX if you want. I’ve learned about the history of Linux and the open source movement from two books: Just for Fun and Rebel Code.
Now, if you look carefully and you’ll notice that the same is happening with Android, maybe less intensive since software may run on different "distributions of Android" (aka different Android versions made for a particular device) but it’s happening. In this article I’ll be talking about this new era of UNIX/Linux fragmentation in two main areas: UI fragmentation (skins), version fragmentation (OEMs holding the source code) and how Google doesn’t care about their app developers.
One of the most visible fragmentation issues of Android is the UI fragmentation. OEMs are (still) thinking that Android is just the smartphone version of the old Java ME-based OS they used on their feature phones. They are skinning Android just like they skinned their feature phone OS.
These skins are creating a huge issue among consumers because since they are not geeks, like us, they think that Android is something like a platform that manufactures use to run smartphone apps, most of them don’t even know its built by Google. These customers also get a little confused when they switch to a newer smartphone, from a different OEM, and the skin of the OS is completely different from the one they were previously used to.
Skins are terrible for the Android ecosystem, but most of them added some really great features, for example the “slide to call/text” from TouchWiz. These improvements to Android UX in my opinion shouldn’t be embedded into these crap bloatwares, but instead should be pushed to Android’s tree (remember the Open Handset Alliance?) to be used on all Android-based devices since they’ll improve the end user experience with the OS.
I don’t know which one is worst: UI fragmentation or version fragmentation, but I think it’s more likely to be the version one.
This problem is completely Google’s fault, even Google itself had problems with Android’s version fragmentation: Google Chrome for Android only runs on Android 4.0+.
Time passed and Google hasn’t come with any solution to Android’s biggest issue, and they know that the only way to correct this is by making pressure on OEMs to only announce new devices if they are running the latest version, and impose tight timelines to update all the older devices that are capable of running the stock version of the latest major release if they want to continue having access to Google Play on their phones. Don’t come to me saying that there is also the carrier problem because there isn’t any carrier problem, Apple updates their devices to the latest version from day one. You know why this is possible? Simple: Apple told the carriers to fuck off. They imposed to the carriers that their OS shouldn’t be bloated, so carriers couldn’t include their crapware with it. As I said previously: Pure stock crapwareless Android is always the solution.
Version fragmentation is extremely frustrating since all those awesome features (and UI) that Ice Cream Sandwich and Jelly Bean have are currently available (stock experience) in one single mass-market device: The Galaxy Nexus.
This is a very serious issue on Android because its making app developers stuck with their innovations, which doesn’t happen on the iOS side. Everyday we see awesome iOS apps, with innovations in UI and UX, getting released for iOS, but we don’t see the same happening with Android mostly for one reason: Developers need to make their app compatible with older Android versions.
The only company that I see that is doing a great job on this area is RIM. They care about their app developers because they know that users want not just awesome smartphones, but awesome smartphones with a lot of awesome apps they can download. A great example of this is how they support natively Java, Flex/Flash, HTML5 and C/C++. They even created a awesome mobile UI/UX framework for HTML5 apps called bbUI.js.
Everyday Android feels a lot more like vaporware for me. I’ve stopped using my Android phones on a daily basis and I don’t think I’ll ever start using them daily again. I’ve stopped developing for Android and now I’m almost fully committed to the platform I’m loving most, and felling more comfortable, to develop to: BlackBerry.