Android fragmentation is a huge problem. Or at least, that’s what we all think after listening to technology experts, web developers, game developers, and simple technology geeks, all talking about fragmentation as if were a nuclear war or a shattering earthquake.
Pretty much every discussion and debate regarding Android’s strong and weak points right now starts with what most people think is the operating system’s first downside, fragmentation. But is fragmentation such a dramatic issue in reality, at least from the app developers’ point of view? Isn’t Android’s own F-word just a nuisance that gets blown to enormous proportions due to exaggerations and opinions based on second-hand experience or no experience at all?
Todd Hooper, a well-known and recognized developer and entrepreneur, posted a very interesting article on TheNextWeb yesterday, surprisingly stating that Android fragmentation is not a very dramatic issue. Zipline Games’ CEO told TNW’s readers the story behind the making and developing of Wolf Toss, a very popular mobile game, currently doing very well, both on Apple’s Appstore and on Google’s Play Store.
Todd Hooper starts his “confessions” by admitting that he was, at first, terrified about offering Android support, due to all the horror stories he heard about fragmentation. The scenario seemed to be confirmed by reality immediately after Wolf Toss’ Android release, when many users complained about bugs and resolution issues.
However, what initially looked like a problem caused by Android’s very diverse, complex, and “fragmented” world, turned out to be an issue simply caused by… insufficient QA (quality assurance) testing.
There were indeed a bunch of device-specific bugs, but the vast majority of issues were, in fact, caused by “classic software engineering” flaws. After fixing bugs that really had nothing to do with the fragmentation of Android or the diversity of Android devices, the guys at Zipline Games reached the conclusion that they simply didn’t pay enough attention to these gadgets.
Just to pause Todd Hooper’s confession for a moment, and to shortly give my two cents on the issue, I think that this is, in fact, one of the issues that most mobile development companies encounter right now. The Android world is clearly very vast and extremely diverse, but you don’t need to test each device separately in order to make a game run on it smoothly. What you have to do, though, is target a specific set of devices, each of these being similar or identical in most terms with a bunch of other gadgets. More specifically, you don’t have to test an app on two, three, or ten smartphones that run stock Gingerbread and sport the same exact screen, you only have to test it (and thoroughly, if I might add) on one of them.
That being said, I would like to end this article with a quote from Todd Hooper’s post, which, I feel, perfectly summarizes the “issue”: “The idea that Android fragmentation is an insurmountable issue for developers is overblown (…) A little extra complexity for Android is worth your time as a developer.” Or, in other words, Android is tougher than iOS in terms of offering support for apps, but in the long-haul, it pays off to take some time for more thorough testing.
What do you guys think, is Todd Hooper right and has fragmentation been simply blown out of proportions lately? Hit us with a comment below and tell us your opinion!
Like this post? Share it!
Ive been saying this forever man. iFanBoys scream this and poor devs that dont properly QA their code have issues. The PlayStore’s biggest problem is to get users to buy apps to keep devs developing.
I’ve been ranting in various articles about how fragmentation isn’t a problem for Android, especially when compared to the REAL fragmentation from back in the J2ME days. Back then you TRIED to group devices, but even that was questionable, so companies like mine (now dead due to the LACK of fragmentation, by the way) purchased hundreds or even thousands of devices to find handset specific issues and build porting solutions. Even better about Android, you really only need a few physical devices, because most of the more rare configurations can be lightly tested using COMPLETELY EMULATED DEVICES. Back in the day, only RIM (BlackBerry) and maybe the odd Nokia device had this feature, but everything else was a simulated device that may or may not behave like the real thing. It’s actually the lack of fragmentation that’s helped push the industry from J2ME to Android Java. I’m glad someone with a bit more clout is finally speaking out.
Sorry, *ranting in response to various articles. As of yet, nobody has quoted me, to my knowledge.
I have to agree. I think the fragmentation issue is a lot of FUD spewed by iOS devs and Android n00bs who don’t understand the platform.
You can check out Wolf Toss at http://www.wolftoss.com – send your bug reports straight to me. :)
Discover the latest apps and games. FREE and enjoy instantly on your Android phone or tablet.
Way overblown. Our company does mobile health solutions on iOS and Android and our Android team is far more efficient. These guys are roughly comparable in talent as developers. Its just a lot easier to build reusable components in Android, easier to wire UIs in iOS, easier to deal with memory management in Android, etc. There’s pros and cons to both. At the end of the day, the Android applications are getting out the door faster. That’s our experience anyway.
Can anyone give me an example of an app that doesn’t use the Android equivalent of JNI (Java Native Interface – used for direct access to the hardware) that has any problem running on a device with the same (or higher) level of Android? There is normally no need for developers to use the JNI, and the only other issue I can think of is the variation in screen sizes.