Apr 03 AT 3:52 PM Dustin Earley 53 Comments

Is fragmentation really that big of a deal?


When I think of the nastiest word I can associate with Android, I think of “fragmentation.” Fragmentation is like a big, dark cloud hanging over Android’s head. Those controlling the weather don’t want to admit they’re the ones who accidentally made it, and those gazing on from the distance have no problem telling you how nasty it looks. But how does it really feel under the dark cloud of fragmentation? Is it really as bad as it seems?

First, let’s address what fragmentation really is, and how it happens. As Google works to release new versions of Android, it’s up to the manufacturers of older phones to update those phones to the latest version of Android. Before a phone can be updated to the newest version of Android, like 4.0 for example, the update has to be reworked and customized for every individual phone a manufacturer plans on updating. After manufacturers rework the newest version of Android for every individual phone, carriers then have to add custom apps, preform tests and work to get rid of all the bugs to create a decent user experience.

When Motorola only had the Droid and HTC only had the G1, myTouch and Hero, the process seemed fine enough. I don’t think Google anticipated Android taking off the way it did. Now, there are more manufacturers with more handsets than I care to name, with less employees than they need to get everything done. Manufacturers release so many phones at a consistent enough rate that they can’t possibly always upgrade every single one. Especially because as soon as a phone hits retail, they’re already working on the sequel to the phone.

So new versions of Android come out, and manufacturers can’t or won’t update their older phones. And most consumers keep a phone for two years. Which means that in some cases, there are consumers out there who are actually still running versions of Android that are years upon years old.

Take the latest Android version tracker updated by Google. A total of 2.9% of Android users are running a device with the newest version of Android, 4.0. Android 4.0 has been around for about five months now. Meanwhile, 6% of Android users are still stuck on Android 2.1, which first came out in 2009.

Comparing Android 4.0 users to Android 2.1 users is a quick way to make the situation seem really, really bad. In all reality, 87% of Android users are using a device running 2.2 or 2.3. Which is pretty good.

Whatever older version of Android you’re running, the outcome is still the same. I could go into the details of every single aspect of Android updates and what technical aspects those updates bring, but what I really want to take a look at is whether or not those things heavily impact the user experience of your average non-Android enthusiast. More often than not, the answer is no.

The fact of the matter is, whether your phone is on 2.1, 2.2, 2.3, or 4.0, you can still make the same calls, send the same texts and run the same apps. Of course, there are exceptions to apps, but they go both ways. Some of the newest apps won’t work on older Android devices, and some older apps won’t work on newer devices. For the most part though, every single app in the Google Play store will work on at least 90% of the Android devices still being used today.

That is the main reason why Google doesn’t think fragmentation is a big deal. And it’s the main reason why your average consumer has no idea what fragmentation even is. For the average user, it’s simply not that big of a deal. They don’t even know they don’t have the latest version of Android. So long as their phone works and can use all the latest apps, they will be fine.

Of course, if you were to ask anyone using a device running an older version of Android if they’d like their phone to be faster or something similar, they’d say yes. But the same goes for anyone using a device running a newer version of Android as well. There’s a lot of factors involved when it comes to a phone’s performance and usually the version of Android you’re running is the least of your worries.

When it comes to how the user interface looks, how many people out there really buy a device thinking, “hopefully, the operating system on this phone looks different some day.”? I’ve personally seen people get upset over their phone being updated and something looking different. Again, so long as everything functions as it should, most people could care less.

Now before anyone loses it in the comments, it’s not like I support fragmentation. I throughly believe that Google and all of the manufacturers using Android need to get together and figure something out. Google needs to share the development process with manufacturers so they can get a head start on updates, and manufacturers need to release fewer devices. And service updates are totally different from major OS updates. I’ll save that for another post.

Still, no matter how you look at it, running an older version of Android is not as big of a deal as you may have been led to believe.

I know fragmentation is real hot topic issue among Android enthusiasts, so I’d love to know what you think. Is it going to be the death of Android? Am I out of touch here, does the average consumer really care about being stuck on Android 2.3, or do they have no idea? Let it all out in the comments.

Dustin Earley: Tech enthusiast; avid gamer; all around jolly guy.

    Most Tweeted This Week

  • http://dylandersen.me/ Dylan Andersen

    For most of the enthusiast users that complain on a daily basis about fragmentation, they should’ve bought a Nexus device or at least been aware of the issue before they complained. Nexus S 4G users on Sprint actually are still stuck on Gingerbread, officially, when they should rightfully so be on ICS.

    We’re just too addicted to updates as a society, lately.

    • http://mihai.discuta-liber.com/ tmihai20

      Most of the users commenting here are not your typical Android users, so fragmentation is an issue for them. I have a friend that uses a Motorola XT720 with Android 2.1, he is a power user and he is not bothered with the fact that he uses such an old Android version. I started using Android only after Gingerbread builds became available for the HD2 I used back then (mainly because I thought that Gingerbread was really worth using every day and because Android in the HD2 NAND was only possible after Gingerbread was launched). Fragmentation is also caused by the manufacturers wanting to release and sell more and more devices (I am referring to the major manufacturers). I don’t agree with carriers meddling with Android, I guess this is mainly an issue in the US, because in Europe updates are released rather quickly after their official release, at least Romanian carriers (Orange, Vodafone and Cosmote) don’t pollute Android with their own bloatware apps (yet). One thing that was taken into account was fragmentation caused by geographical placement, I’ve encountered countless apps limited by my IP (I’m not talking about US specific apps).

      • rijah

        Sometimes, even Europe update is bizarre. I have a friend who has a Samsung S2 Unlocked Edition not updated to ICS yet and most of his friends on the carrier network updated to ICS..!! As an android user, according to me, we should only promote the vendor who updates the OS for 2 years minimum for any phone it releases, otherwise should be boycotted. I remember in the early days, during version 1.6, when Sony was not updating its android devices and the users getting furious at Sony for not updating them. Sony did have a bad time after that, if I am not wrong, till version 2.3 and have started updating all its phones to the next level since. Hope all the vendors take a leaf from that book.
        From the other side, Its the customers who force the mobile giants to release new device every year or half year as we tend to buy them. Google should force only 3 models per year from each manufacturer – high, medium and low level model.

    • Kevin

      goodness, its true. In korea ics for the nexus s has yet to begin, all the while samsung, released ics for the galaxy s2.

      they actually outpaced google at their own game :)

  • thekaz

    Honestly, I think part of this is growing pains. Once they get to a stable version containing all the functionality or the groundwork for the functionality and the groundwork for the visions they have, we will see the impact of releases become less. This in turn will hopefully allow hardware manufacturers to catch up. There will be old devices which will force some users to only be able to upgrade via purchasing a new device, but eventually, we will get to a release which will work across hardware, and upgrading will be easier.

    That’s my two cents…

    • dcds

      I also think this is the most probable outcome. Great comment.

      To that I add the benefits of having an open source system: we are already seeing the efforts of Google and others to backport a lot of API 11+ functionality to devices as old as API 4. This on the developer front.

      We are already starting to see the quality of general and low-profile apps improving dramatically after Ice Cream Sandwich was launched, and after people started backporting stuff. I’ve seen many apps being updated to ice-cream-sandwich-tasty UIs lately. Even on older phones. Things like Actionbars, seamless data loading and others are effortlessly done right now.

      A huge change from 1 to 2 years ago.

  • McLovin

    I do think Google should “step up to the plate” a bit here. I’m not asking for a full tilt Apple Nazi style locked down app store. But they could be more of a gatekeeper on what gets into the play store as far as will it run on more than one device. or adhear to some standard.

    • http://www.nexsoftware.net Justin Shapcott

      Erm… Every app in the Play Store (that name still doesn’t sound right to me) adheres to a standard… It’s called the Android SDK. Some developers choose to use additional non-public APIs, or include libraries that may or may not play nice with every version of the SDK, but they are very much in the minority. The fact is, Dustin is right in that the vast majority of apps and games in the store will work on every relevant version of Android.

      • Mike

        Except even if you only develop to public APIs in the SDK, you still can run into problems. I’ve run into a few myself. Off the top of my head I remember a known problem with how the first Samsung Galaxy had a problem with some image animations. I don’t remember what the specifics where, I’d have to go look in the comments of my code, but with the varied hardware and SDK versions out there, there is a high possibility of different phones having bugs that affect the level of effort a developer has to go through to get an app to properly run on all phones. Something I have less of a problem with when I’m developing for iOS.

        • http://www.nexsoftware.net Justin Shapcott

          That’s a device problem/bug, not a fragmentation problem. But I get what you are saying and can agree that sometimes manufacturers ship buggy devices that can have problems even with the public APIs. I’m not sure that is part of this particular discussion though.

          • esper256

            Device bugs are a fragmentation problem. Two devices that are supposed to be Androids that have different behaviors when there’s only supposed to be one behavior.

            Ultimately Google is battling these types of problems in the form of adding CTS tests. Every device that gets called an Android device has to pass these tests to help weed out these problems. As time goes on, more and more of these tests are created to help make sure OEMs don’t produce phones with glitches that will cause apps to not work. It’s really a growing pains issue. Given Android’s explosive growth, it didn’t really have time to mature the CTS before becoming super popular and spread out to hundreds of devices. But as the versions become more mature, and the changes between OS versions less drastic, these issues will really become few and far between.

  • Adryan maldonado

    No dustin you are not out of touch you are practically spot on. To the average consumer it doesn’t matter. In fact your consumer is generally going to use very common apps which usually work on 95% of devices. My mom doesn’t care that she’s not on ICS. My brother barely cares but doesn’t really pay that much attention to it like my self. Like you said google and the OEMS need to figure something out but then at the same time if less consumers care then how much are they going to care?

  • Dan

    Fragmentation isn’t a huge deal when it comes to software, but it is when dealing with hardware.

    Most apps aren’t compatible or run into problems because of the device they are using, not which version of Android they have. This is why we are seeing developers abandoning android projects.

    Sure, it’s great to have 100,000 different phones to choose from, but that’s 100,000 problems for developers to worry about.

    • dcds

      Sorry, except for the first paragraph, I can’t agree.

      What you’re saying is that it’s not developer’s fault if said devs are lazy to implement graceful degradation. In short, you’re implying that everyone should have fewer devices to choose from to save devs most work hours.

      Think again: the world is exploding in mobile usage. The staggering we’re seeing are nowhere near what they will be in the next years. Customers don’t like to choose from only one car, one house layout, one bike, one job, one shirt, one CPU, one GPU, one computer, one laptop. People are not going to lose weight to save designers hours to think for clothes just for thins, or learn to drive manual to save GM et al from having to manufacture more complicated, very model-specific automatic gears.

      Do your research and see which sensors work for your application. Specifications can be found on the Internet.

      Wake up man, that’s how the world turns. It’s up to us to cater to users needs.

  • mkstvns

    It doesn’t matter what Android’s problems are, be they technical or commercial, genuine or imagined, the end point is this: so long as developers remain disinterested or deterred, for whatever reason, it is Google’s fault.

    It. Is. Google’s. Fault.

    If handset manufacturers are screwing things up in someway, it’s Google’s fault for allowing that environment to develop, and for not making a more proactive effort to ‘fix’ it.

    I don’t care what ‘fix it’ even means here; Google is loaded with money, and with smart people, and it’s not my job to imagine a solution for them.

    It’s not even my job to spell out exactly what’s wrong with Android. All I need to know is that there are hundreds of incredibly interesting, exciting, compelling apps on iOS, and the only thing keeping them from Android is whatever the developers think is wrong with the platform.

    And as I said above, whether it’s imagined or a genuine technical problem, Google needs to get these heads in a room together and nut it out.

    Developers aren’t staying away because it’s not hip to be on Android, they’re staying away because at any given moment, they can’t tell which way The Wind Named Android is going to blow.

    For whatever reason (again, I don’t care to investigate it, and I damn well don’t need to), too damn many developers are worried about going within sniffing distance of Android.

    Fragmentation, screen sizes, hardware conflicts, whatever. It doesn’t matter. Google hasn’t built it right (the environment, the business model, the relationship, the image), and they (the most exciting apps) aren’t coming.

    On the plus side, the past year or so has seen some pretty exciting apps come, and obviously more are coming. But Google needs to do something to make developers froth at the mouth so that their apps are here immediately, not a year after the iOS version.

    • Alex

      I completely disagree. Aside from a few niggles, developing for Android is really easy. If you’re a good developer, fragmentation, screen sizes etc aren’t a particularly big issue (unless your app uses the NDK or a game). Android is designed so that, in almost all cases, an app which works on 1.6 will work perfectly on 4.0.

      I bet most of the developer “issues” are iOS developers looking at it from an iOS perspective. Let me tell you, the platforms are radically different and stuff that works on iOS doesn’t mesh with Android.

      • Mike

        In my experience, it has been more than a few niggles (and I was an Android dev before I had any iOS under my belt)

        It’s not the 1.6 vs 4.0 that gives me fits, it’s the one manufacturers hardware vs another’s. More hardware manufacturers/SDK version combinations means more chances to introduce a bug in the process. I’ve run into a lot more of “I can’t get your app to run on my phone” in Android than I do in iOS.

        I still like the platform, but fragmentation does cost me more dev time. I do agree though, that consumers rarely care.

      • mkstvns

        That’s not the point I was making, Alex. I was saying that whether it’s easy or not, there is obviously something keeping many developers away from Android. Not all, and certainly we’re seeing some great new apps make the leap to Android, but there’s certainly no flood either.

        All I’m saying is that whatever the perceived problem is, Google needs to make a bigger, more obvious effort to get developers excited and on-board.

        Has Google made an effort at all? Sure it has. ICS is great, and it’s doing its bit. But is Google doing ENOUGH to entice the best developers? That’s my question, and I think the answer, for now, is no.

      • Dan

        I have a family member who works for an international software developing company. Believe me, the problems of having to design apps and games for so many pieces of hardware is very real. As a business decision, many developers just skip android altogether.

  • Alex

    this is exactly what I said earlier on another forum. The people who are on 2.1 *aren’t* going to care they’re on 2.1. If I ask my dad what version of Android his Galaxy Ace runs, he’d tell me “the Google version”. For most people, a phone calls people, sends texts and maybe checks a few websites or sends a couple of emails. They don’t give two hoots if they don’t have Ice Cream Sandwich.

  • Lee Swanson

    I’m probably somewhere in the middle of the average user that’s fine with what my phone does now and the person waiting for the update to the next Android version. I really want some of the battery/performance improvements of ICS and Google Talk with video chat, but I could care less about face unlock and the interface changes.

  • Haloruler64

    I agree completely. Most users will never know what version of Android they run. Updates bring very little to the average consumer. Does ICS TouchWiz 4 look much different than GB TW4 on the Galaxy SII? No. Do I love the ICS update? YES! Will an average consumer care? Definitely not. Differences to the core experience are minor. Nothing big has changed except broken compatibility with some older apps. When you buy your laptop, you are never guaranteed updates. You can BUY them, hope they run, but you aren’t guaranteed. If you buy standard parts, like an Intel motherboard and an i7, and a fairly common GTX 560, you know it’ll be compatible because drivers will be released and NVIDIA and Intel support their hardware. Updates should not be guaranteed for Android devices IMO, you buy what you bought that day, and updates shouldn’t be GUARANTEED. However, Qualcomm and others should be a lot more open with drivers and proprietary modules so people can get newer versions of Android running on older hardware at their own disclosure.

    I had to RMA an HP laptop. It came with XP (netbook, XP was cheaper than 7), and I had 7 on it. They told me they wouldn’t repair it unless it had XP. So I installed XP and shipped it off. Same with Android. If you mod software, you can’t send it off. But public drivers are a MUST.

  • Bon

    I believe It’s a bigger deal to the developers, but not to the end users. the users with enough knowledge know what they are getting when they buy a new phone. the others probably don’t know or care about what version they are running.

  • Bryan Stoner

    First off I love your article~!! I absolutely agree with it 100%. It truly does not bother me one bit and I’ve noticed too that end users don’t really know the difference anyway.

    Even if you can’t dl an app there probably is an alternative that is just as good or even better.

    I think a great thing for manufacturers to realize, if they want to get the ball rolling, is that end user developers are a secret key to success! How great would it be if ROM and Kernel developers could have the spotlight. Manufacturers could have a bounty for their device in order for an end user development team to swing by and modify android for that device. That way those developers could make a living doing something they love, manufacturers could save a ton of money, and the said android devices would get the latest update possible. Of course, manufacturers should get the hell over their obsession with skins and custom apps. If that happens I’m sure we could see super speedy updates.

    That’s just an idea. I think there may be some nasty side effects from that like stealing work, dirty intentions, and probably some other stuff. But overall it is a wonderful idea.

  • Peter

    I think that you need to compare Android to Apple and count how many iPhones are running old versions of iOS? It would be significant because the older iPhones can’t run the newer OS. So really we need to see how many phones that could run a newer OS aren’t because they have not been upgraded by the manufacturer as opposed to older phones not having fast enough hardware to run it. I could upgrade my two year old phone to ICS (it is rooted) but I choose not to as I don’t think it has the performance.

    One last point is that a lot of Android phones could run later Android due to rooting the phone. So although Android is bad for fragmentation, it also gives people the option to upgrade through rooting.

    • Greg

      Almost all iOS devices are on iOS5 (the percentage is around 80%).

  • Walt

    I think fragmentation is made into a bigger deal than it really is.

    Fragmentation is a hassle for the techie types like myself who want the lastest and greatest, people who stay up on Android news, people who read blogs like this.

    For people like family and friends who use their Android smartphone for Facebook and Angry Birds, if you asked them what ICS is, they would have no ideas its a version of the Android OS and could care less about it on their smartphone as long as the apps they use function properly.

  • Aaron

    I’m so sick of grammatical errors by “writers.” “There‚Äôs a lot of factors involved when it comes…” I think you mean “there are.” It’s not “could care less” but rather, “couldn’t care less.”

  • Mark

    “Again, so long as everything functions as it should, most people could[n't] care less.”


    that is all!

    • Aaron

      Brilliant. I’ve posted it to my idiot friends on Facebook. I can’t stand when people say it like that.

  • spazby

    it’s a big deal for me and i would think a big deal for most people on android. the reason i am on android is because i care about the guts and possibilities. if i did not care about that, i would have an iphone

  • Ishiken

    The average customer doesn’t understand what happened to Android 3.0. They look at their phone and see v. 2.6.35 or whatnot and then see the GNex or the other devices getting 4.0 and a quizzical look falls over their face wondering why they never got 3.0.

    Fragmentation is a figment of the imagination of those who obsess over something they should just relax on. Mobile hardware is advancing faster than it can be supported. it is not economically viable for an OEM to produce devices, which is all it does, and then to support its older devices because people want the latest and greatest. Apple gets away with it because it only manufactures ONE (phone) device per year. ONE. If I made only one (phone) device a year, I would have to keep up on its old hardware too. It would be detrimental to business not to do so.

    Do you get pissed off when your laptop isn’t upgraded as often? These devices have a finite shelf life. Expecting the OEM to maintain it for more than a year when their business model requires you to purchase more than one phone per year or every two years Stop bringing this up as if it is an issue that matters. Android is an OS, one that goes on a lot of devices (like Windows!) and that OS is updated regularly by Google and those within the AOSP and AOKP. It is upgraded for free by the way (Unlike Windows… Damn you Redmond!).

    So enjoy your Android version, tweak it so it works smooth as butter and when your update comes around or you decided to root and upgrade the device yourself enjoy that too and be good. Don’t let all the fragmentation talk get to you. It’s nonsense and really is beneath most of the people who “report” on it.

  • Schmidty850

    To me, fragmentation is really not that big of a deal. It also helps that I’m rooted :P

  • pman

    I think manufacturers should start working on their customizations separately from their devices, then offer it on the Play Store as a paid app for devices not from them, and as a free app for their devices (to make it clear, here’s an example: suppose HTC offers Sense as an app, and I have an SGS2. To have it, I have to pay, say $1. While my friend with an HTC Sensation can get it free)

    That way, there are several upsides:

    1. one consistent version of the skin (so we won’t be seeing Sense 3.0, 3.5, 3.6 and 4.0 at the same time)

    2. additional revenue for OEMs (I think there are a lot of users out there who have non-HTC devices still attracted to the Sense skin)

    3. Android hardware will by default be running stock. Google can showcase their homebrew UI, OS upgrades will be faster, consumers will be happier people.

    Though most if not all manufacturers won’t be buying this idea (it will take out their “differentiation” to some extent), it will actually be beneficial to all if they will be competing with hardware. They will be racing to provide DSLR-quality camera sensors, uber-long battery lives, slimmer form factors, and the list goes on…

    So to cut my long story short: GOOGLE AND OEMS, Y U NO GO FOR A WIN-WIN SOLUTION?

  • http://droidsamurai.blogspot.com DroidSamurai

    The fragmentation problem is not just about app compatibility. It’s also about whether Google can guide the platform in an efficient manner, and the answer is a “NO”. Everything Google wants to deploy will take at least an year (or more) to reach most consumers. Its rivals will have a lot of time playing catch up.

  • Nathan D.

    If there wasn’t such heavy amounts of phone produced this wouldn’t be such a problem with some people. Either way I really don’t care about it that much since it doesn’t concern me that much , yet.

  • redraider133

    I do not think it is that big of a deal. I think getting bug fixes are more important than necessarily having the latest version of android. Most people will not complain if the phone is pretty bug free and gets updates to fix any issues it does have over having the latest version of software. Most people who do not visit sites like these do not know the difference between versions anyway.

  • BigCiX

    it wouldn’t be so bad if updates were getting to devices 3 months after release.

  • skugern

    A good portion of the blame can be laid at carriers’ feet. Just for an example, look at Engaget’s recent review of the Samsung Galaxy S Blaze 4G from T-mo: one of their biggest complaints was all the bloatware Big Magenta put on it. When a new OS version comes out, the carriers have to ensure all their carrier-specific apps are compatible with both the new OS and with the manufacturer skin, which takes even more time.

    Also, I can see fragmentation as a possible issue for the average user when they’re trying to make sense of all the software versions and instead turn to a fruit phone because of the perception that “it just works.” Otherwise, I don’t think fragmentation is a huge deal to most users.

  • Michael

    Absolutely I am concerned with having the newest version. Apps receive. Updates based upon the version of andy you are running ie. The new google maps update for ics. If not for that factor, I could care less.

  • cb2000a

    My phone uses 2.3.4 and I am very happy with that version. If I never get an update it will be fine. I don’t want a lot of apps on it anyways.

  • Chuxter

    I believe the biggest issue or problem with fragmentation has less to do with the user perspective. This article is correct in stating that it’s not THAT big of a deal if your not using the absolute latest version of Android.

    I think the most detrimental thing about fragmentation is that it makes it difficult for developers to create apps that are supported across all devices. Too often an app is released that is not compatible with certain devices OR installs only to crash on said devices. This needs to stop.

    The real value of an ecosystem is largely dependant on the apps which are available in that ecosystem. Making it harder for developers to create smooth running apps doesn’t just harm the developers, it harms all android users.

    • brklynmind

      I think this is exactly it! Users dont care about ‘core’ android – because it does everything they were expecting (calls, texts, menues, etc…) and a few new options (many of which are available as separate apps are usually insignificant. Only people who have (mostly a psychological) need for the ‘latest’ version care. BUT people do care when someone says – ‘hey get this new app/game’ and it wont run on their device.
      By way of example most high end games will not run on HTC Rezound – why? I have no clue but it is frustrating.

  • JB

    The version of the OS is usually the least of the issue. As a developer we see the real problem with fragmentation is the different screen sizes and screen resolutions. Yes, we can write apps in a device independent way. But a phone with 3.2 inch screen and a tablet with 10″ screen have significantly different screen areas and apps should be written to take advantage of the differing screen sizes. Just stretching the same app as the 3.2″ to run on an 10″ screen is not a good solution.

    This means, a good app’s user interface may need to be written multiple times (at least 3 times) to put the various screen sizes and resolutions to good use. That is an expensive and time consuming process.

    To me, the differing screen sizes and resolutions is the real fragmentation problem. And I don’t think that issue will go away anytime soon.

  • Gold D Roger

    Hey…..Fragmentation is really a problem. Makes it hard to develop for android..

  • Wellred

    Excellent article. Very well written, balanced and well explained. I now have a better understanding of the elements in this subject area. I will tag this to my friends,

    Great comments too.

    I think a wise manufacturer could develop excellent devices and stick with it and plan on long term sales, support and upgrades. Great devices have more than a 3 year hardware shelf life. So often manufacturers pat the cash cow and are very tactical to try and stay on the bleeding edge; when the edge (new devices; or versions of a device) doesn’t really provide the mid level user ‘major’ new experiences.

    The Sony PS3 is probably the best example of this strategy. There was some long term thinking. When did they first get released?
    I wasn’t going to say it but the Iphone could have been another, but they still have a shorter release cycle.

    On a side note; there are many people, definitely in certain age groups, classes and cultures; that are addicted to new technology and platforms that will continue to drive the cash cow that is; to get a new phone/device every time I renew my data/phone contract because there’s usually a package deal. Why not? It’s a great deal all calls and new phone included.
    The constant proliferation of this market the manufacturers . Anyone been to Japan?
    There are many people in the world who could easily count over 10 mobile phones they have had in their lifetime so far.

    I live in Australia and most people get a new phone every time you renew your contract, which is usually two years. Why not? The cost of the phone is, most commonly, incorporated into the free calls and minimum spend level. The other option is to spread the cost out over the 2 year contract.

    So which manufactures are going to start producing a device ahead of it’s time for the right money? Is it possible?

  • Lukas

    The fragmentation is not a really big problem for android developers. The standard api is docummented well and in most (but not all) cases works as expected. What is a REALLY big problem is all those custom ROMs and the fact that those ROMs have access to the Android Market App. When we released our last product, and it started gaining users very quickly (n7player BTW) about 80% of crashes happened on CM or other mods.
    We had many equalizer issues, for example devices raporting that equalizer is supported but every band is 0khz. If you ask me, mods should not be allowed on Android Market because they are lowering the user experience. It doesnt matter if the user is using mod or not – when app crashes he will put a 1* in the review.

  • GiqueGEAR_Todd

    FINALLY. Responsible treatment of this topic. Stats of install rates are generally meaningless.

  • RevSpaminator

    Great article about the reality of fragmentation.

    Fragmentation seems to have the greatest impact on those who don’t even own or like Android devices. :)

    # Android FUD Machine

    while (1=1) {
    echo “FRAGMENTATION !”;

  • graywolf323

    fragmentation is the one major appeal of iOS to me over Android still

    as much as I love my Galaxy S II I’m not sure I wouldn’t switch if T-Mobile ever officially got the iPhone (I do already have an iPod Touch and am getting a hand me down iPad from family as a birthday present)

  • Thomas Taylor

    I just got back from the Nonprofit Technology Conference, where I facilitated a discussion among a group of IT directors at mostly mid- to larger nonprofits (100+ users) about managing mobile devices. As an Android enthusiast, it quickly became clear to me that adoption in business is one of the places that fragmentation is killing Android. Many orgs got so frustrated with the variety in details and quality of carrier- and manufacturer-built Exchange ActiveSync implementations before Android’s built-in support rolled out in 2.1 or 2.2, that they made a policy of not supporting Android at all. And the current (I’m on v2.3) isn’t even that great either. Along the way, there have been capricious and unnecessary renaming of UI elements, so that documenting how to get your phone connected to the company’s Exchange server is needlessly difficult. Even with OS-supported ActiveSync now available, the variety of UI skins and other manufacturer & carrier variations makes Android devices much more difficult and costly to support in a business environment than a Blackberry or iOS device. If Blackberries didn’t suck so hard for browsing and other uses of smartphones, a lot of IT Directors would love to stick with them.

    The other major issue around fragmentation that this article doesn’t even touch is the fact that even within a given OS version, there are many apps that run poorly or not at all on certain devices. This also has a very negative impact on business adoption, where BYOD is an increasingly popular and cost-effective way of handling mobile devices. But if you have to get into the weeds of mandating specific devices from different carriers because you’ve tested them thoroughly with all your business apps, then you’re losing a lot of the benefit of BYOD.

    Fragmentation goes way beyond multiple versions of the OS in the wild.

  • Num

    Fragmentarion IS a bug deal for the developer. End user not so much