26 Comments

Summary:

Google’s Android platform is gaining smartphone market share, but future growth is at risk due to fragmentation. Fortunately, however, the company appears to have a plan for addressing such an issue that reportedly involves “decoupling many of Android’s standard applications and components from the platform’s core.”

Google continues to gobble up large chunks of smartphone market share, but offering four different versions of the same operating system will eventually stunt that growth by derailing customer purchase plans. After all, why buy a new device with an earlier version of Android — say, version 1.6 — if the better hardware and software is devoted to version 2.1?

Such fragmentation is running rampant on the platform, which has only been shipping products since October 2008, leaving the owners of older handsets pining for apps that their friends with newer phones can run, or wishing for advanced native functions like multitouch capability. Developers, meanwhile, are challenged by having to build different Android apps for different versions. Luckily, Google appears to have a strategic plan to address these problems, Engadget reports today.

Thanks to conversations at last week’s CTIA, as well as some follow-up information, the site says it has “reason to believe that the company will start by decoupling many of Android’s standard applications and components from the platform’s core and making them downloadable and updatable through the Market.” In other words, only the base Android functionality would be in the hands of carriers and handset makers, while third-party developers — and Google itself — would expand Android functionality through downloadable software.

The expectation is that this effort will take place over the next two Android updates, codenamed Froyo and Gingerbread respectively. By managing the fragmentation in-house and divesting core apps from base Android functionality, Google regains control: It can move key applications to its Android Market, and reduce the carrier influence over what apps can or can’t be on the phone. The approach fits nicely with the Google Nexus One strategy Colin Gibbs outlined at GigaOM Pro (subscription required) as Google attempts to wrest control from the cellular network providers.

Back in February, I noticed subtle signs of a shift to address the Android fragmentation issue, and rumors at the time were indicating that Google could try to migrate all existing handsets to Android 2.1. That’s a tough road to hoe because here in the U.S., carriers decide what software is pushed to handsets on their network. A notable exception to that practice is Google’s own Nexus One which accepts software updates directly from Google — cutting the carrier out of such a role. Still, the first whisperings of a solution were heard, so I kept my eyes open.

Three weeks later, my watching paid off — I noticed that Google’s software strategy had shifted over time. Instead of the latest and greatest native apps making their way to Android 2.1 only, functions were filtering into older versions of Android not long after release. Google’s new Gesture Search, for example, appeared on Android 1.6 devices only two weeks after debuting on Android 2.x phones.

From a consumer standpoint, separating core handset functionality from applications can reduce buyers’ remorse as Android matures. That doesn’t mean that every Android application in the future will run on the handset you just bought, but the functional base between various Android devices should be much more similar. And if Google can get a more standardized version of Android across its handsets, developers won’t be as challenged to port code between various SDKs and feature sets. Happy developers ought to make for happy customers and help continue Google’s path towards mobile dominance in the smartphone market.

  1. Dameon Welch-Abernathy Monday, March 29, 2010

    This goes some ways towards solving the platform fragmentation problem, but it still does not address other aspects of platform fragmentation including:

    1. Different operating system versions between handsets

    2. “Customizations” operators make to their version of Android that might prevent some apps from running (like the Motorola Backflip not allowing non-Android Store apps)

    3. Differences in handset form factor that application makers don’t take into account.

    Platform Fragmentation is not only an Android problem, of course (Nokia is the undisputed king of this). Unfortunately, the “solution” to this problem is something along the lines of what Apple does, and that it’s own flaws.

  2. This is a band-aid and will not end fragmentation. Instead it makes the problem even worse as now the consumer needs to go fishing to find the right components – fine for geeks, but not your average user. Further, for the carriers, part of the attractiveness of Android is that they can “customize” to make their own unique versions (hardware/software) to differentiate themselves from their competitors, the described approach takes some of that control away from them and leads them to being even more of a commodity. Ouch. Google is really whoring up the market.

    1. In my experience of using Android, I haven’t had to fish around for updates, but they’ve been pushed out to my handset (this was for the Milestone 2.0 update to the 2.0.1).

      Updating apps on the marketplace weren’t as obvious, as I had to navigate to ‘downloaded apps’ and then see if there were any updates, but, again- after finding that option it was quite straightforward.

  3. I see fragmentation in the frequency of app updates – it seems like each of my 20 added apps gets a new version every 3-4 weeks, so I average about 6 updates per week, often with no added functions for my phone. It’s annoying, must be tough on Android developers, and I hope it doesn’t accelerate as more new Android phones hit the market.

  4. What some would call fragmentation, alot more would see as customer choice. A big reason why Android will soon challenged Symbian as the most widely used smartphone OS on the planet, is that Android gives it users choices, its not a locked-down, closed toy phone OS.

    1. Choice. Thats a good one. Thats all you got? Choice between an older more lame version of Android or a newer slighyly less lame version?

      1. Choice, freedom. For example, for Apple cultist, there is no choice is when All Knowing Leader Steve Jobs dictates what is good or not good for the group.

    2. Dameon Welch-Abernathy Tim Tuesday, March 30, 2010

      Obviously you haven’t seen the Motorola Backflip, which AT&T has configured to prevent people from installing apps that didn’t come from the Android Marketplace.

      Choice even within a platform is a good thing, but too much tinkering is also bad. An example of a diverse, but not fragmented platform is Microsoft Windows: generally speaking, you can buy a piece of software that says it runs on Windows and provided you’re running on a Windows PC bought in the last 5 years or so, it will work.

      The problem is, unlike with Windows where Microsoft has complete control over the core OS, all mobile phones operating systems (except the iPhone) are “tinkered” with by third parties (i.e. the operators). This single practice is what is responsible for the most damaging fragmentation within any given platform.

      If we could somehow prevent the carriers from touching the core OS of the phones (or if they took a far lighter touch than they do), it would go a long way towards solving the platform fragmentation problem.

  5. This is bound to fail.

    Carriers and cell phone manufacturers want to differentiate their products so that their branding will stand out. This “choice” instantly creates platform fragmentation. Since the carriers and cell phone manufacturers provide different hardware capabilities and since they do not upgrade their core software often, this causes enormous fragmentation problems. Note that Apple updated its core software 18 times in the first year alone. Carriers and cell phone manufacturers will not do this.

    Google’s solution will only create more fragmentation since consumers won’t be lock-step in upgrading their phones like Apple’s customers are. This means there will be MICRO-FRAGMENTATION as different handsets within the same model, same manufacturer, and same carrier will have different functionality. Thus developers will be driven even more batty. They can’t make assumptions other than the lowest common denominator – which is getting smaller and smaller and smaller. This will create a situation like Windows where developers create software for the lowest common denominator – leaving the fancy integration of applications and the possibilities of wonderful collaborations between multitasking applications – off the table.

    Too bad.

    Choice creates Micro-fragmentation and hundreds or thousands of permutations among the cell-phones available to consumers. Pity the developers.

  6. The device fragmentation problem persists | The Equity Kicker Tuesday, March 30, 2010

    [...] Has Google Solved Its Android Fragmentation Problem? (gigaom.com) [...]

  7. I disagree that fragmentation will impact customer growth because 90+% of customers don’t even known there’s a problem. What it does mean is fewer developers expending more effort to maintain compatibility with more flavors of Android resulting in fewer interesting apps. So customers notice their friends iPhones have some cool app which they discover they don’t have for their HTC SenseUI’d Android x.y handset; This is what will impact customer growth.

    Google needs to solve the fragmentation issue sooner instead of later before the Android brand becomes synonymous with something cheap and unpleasant.

    1. People do know its a problem, because the competitive platform (ie the iPhone) pushes out updates to all its devices. People are used to the new features they get every time its updated (like voice recording when you update to os 3.0) Why cannot a G1 run the latest and greatest Android when the iPhone edge can? Sounds like a problem to me.

  8. This also makes sense for the Android platform in general – not just smartphones. The platform is being used in all sorts of consumer products, for example washing machines. I’m sure the devs of these products don’t want the native Android apps on the platform as standard (though I’m sure its pretty easy to remove them).

    @James Katt – I’m no expert on this but surely Google controls the OS updates on Android, so you wouldn’t need to wait for the phone manufacturers to update.

  9. Go Android !

  10. Google’s Android Fragmentation Problem Persists: AdMob Tuesday, April 27, 2010

    [...] them available for download on both old and new Android handsets. As I pointed out last month, such an effort helps reduce fragmentation on existing handsets because “only the base Android functionality would be in the hands of carriers and handset [...]

Comments have been disabled for this post