The Verge reports that Google is changing how they report on Android numbers:
Google routinely gives monthly numbers breaking out the adoption of various versions of Android, but the company has now changed the way it calculates those numbers — providing a distinctly different portrait of the Android ecosystem in the process. As outlined on the Android Developers site, Google now uses the data collected when users visit the Google Play Store; under the previous system, any check-in to the store by the device would have been incorporated into the results, user-generated or not. The new system went into effect starting with this month’s results.
I don’t see how this is any better than going off activation numbers. Google is choosing to ignore a substantial number of Android users – the Mum and Dad users that just got the phone for free and who don’t care about apps.
It’s a double-edged sword, on the one hand it would show a reduction in the number of Android users but on the other hand it cleverly hides the fact that those on low-cost Android devices are running ridiculously updated software.
I wouldn’t be surprised if this was a precursor to Google eventually restricting access to the Google Play to those running more recent software, say Ice Cream Sandwich onwards. That way, only newer versions of Android would be shown, giving the impression that fragmentation is decreasing, and since most Google Play users are likely those who try and stay fairly up to date with Android, there’ll be minimal outcry.