As most of you know if you care about the subject at all, Apple’s iOS migration rate is many times that of Google’s Android. So less than eight months after iOS 9 was released, Apple lists an 84% adoption rate; it’s over 90% at Mixpanel Trends.
These numbers are somewhat better than iOS 8, but you can’t say the upgrade rate wasn’t high for that release too.
On the other hand, users of Android gear don’t seem so inclined to use the latest and greatest operating system. So six months after Android 6.0 Marshmallow arrived, only 9% are using it. But that’s nothing unusual. Normal first-year upgrade rates for Android tend to be in the single digits.
One article I read claims that, ‘The problem is that given Android’s current update model, no one wants or cares anymore about yearly Android releases.”
The article goes on to explain that the newest releases tend to be installed on high-end Android gear, but not always immediately. It may take a few months for the new release to be deployed, and it would tend to be installed on newly released gear, not existing products, and only a small portion of those.
The article goes on to claim that, “Android users don’t care about updates (even security updates),” but provides no evidence that this is so. It may be more about the fact that they aren’t being made available. I cannot imagine many customers saying they’d rather not install a security update that made their experience in Android-land safer. Are there such surveys? When I went online to check, I found an blog post calling the claim “dumb,” and I would tend to agree.
The real issue is that the process of pushing Android updates is sadly broken. Despite promises to do better, Google isn’t really doing what’s necessary to make it happen, to simplify the current schemes.
So the updates are first sent to the handset makers. They will have to merge and test it with their own customized version of Android, which includes lots of changes and junkware. If and when that is done — and it’s rarely a given — the update would have to be submitted to the carrier. In turn, the carrier will test it and integrate it with their own customizations. After this process they may push it to the customers, but not too often.
This is the sort of multistep process that’s rarely done, which means users of Android gear have a fairly low chance of being able to upgrade, even if it’s to fix a critical security problem that could result in their hardware or data being compromised.
Notice, that the FBI isn’t going after Google or Samsung, or LG or HTC or any other Android handset maker to demand that they unlock a handset in connection with a criminal investigation. More than likely, they’d have no problem breaking in.
In any case, clearly operating system upgrades are important to Google, since they hold special developer conferences to show off the new features. They must want developers to deliver apps that exploit those features.
But that would be a waste of time. If over 90% of Android users don’t have an upgrade six months after it’s released, it makes more sense to support prior releases, which may have a larger user base. So more than 45% of Android users have handsets on which 5.0/5.1 Lollipop, released in 2014, is installed. Or the previous version, Android 4.4 Kit Kat, which has a reported 23% of the user base. Both are based on Mixpanel Trends figures.
So maybe Google should be telling its developers that they love to see them support the new OS, but maybe next year or the year after.
Apple’s key advantage was in obtaining permission from carriers to push its own updates. So wherever you buy your iPhone, you will receive support and OS upgrades direct from Apple. Sure, you may sometimes see a carrier update being fed to your device, but it’s mostly about the OS. Apple relies on its own servers to send updates to customers. If something goes wrong with any of these releases, and iOS 8.0.1 is an example, it can be quickly withdrawn and replaced with a fixed version.
Imagine that happening on Android. Even if you are able to upgrade, if something goes wrong, it may take weeks or months to get a fixed version, assuming it’s ever made available.
Part of the problem results from allowing manufacturers and carriers to customize the OS. That puts them in the middle of a process that ought to be the province of Google. It might be better to install the same core OS on current handsets, and put the customizations in a separate package as a collection of apps. That would separate them in a way that would make it easier to get the needed updates to customers.
More to the point, Google should be allowed to push its own updates as they are made available. That would no doubt require renegotiating agreements with carriers and manufacturers. But it would benefit customers, and aren’t the customers Google’s main product? Otherwise, nothing changes, despite Google’s empty promises to do better next time.
| Print This Article