Google’s Android operating system has changed quite a bit since the first Android-powered phones began shipping in 2008. And maintaining support for old versions of Android gets increasingly difficult over time.
So it’s unsurprising that a few years ago Google announced it would be largely ending support for devices running Android 2.3.7 Gingerbread or lower by requiring users to have Android 4.0 or newer devices in order to access the Google Play Store and other Play Services. But now the company is taking things a step further – if you are still using an Android 2.3 device, you won’t be able to use it to sign into Google apps and services after September 27, 2021.
That means you’ll likely get a username or password error when you try to log into Google apps and services including:
- Gmail
- Google Maps
- YouTube
You may be able to continue using some of those services indefinitely on devices that already have you logged in. But if you sign out and try to sign back in again, attempt to add a new account, perform a factory reset, or change your password (on any device), then you’ll be locked out.
Google’s Zak Pollack explains that the move is part of the company’s “ongoing efforts to keep our users safe.” That’s pretty much the end of the explanation, but the implication is that since Google no longer provides software or security updates for Android 2.3 Gingerbread, logging into your account on a device running an 11-year-old version of Android could be insecure. So the company recommends moving to a device running Android 2.3 or later
There is one workaround though: use the web browser on your phone. You’ll still be able to login to your Google account to check your email, access maps, watch videos, or use other Google services through their web apps that way.
Alternately, you could see if there are custom ROMs available for your older Android devices that allow you to update to unofficial third-party builds of newer operating systems. The xda-developers forum is often a good starting point for that search. Or if you’re interested in exploring alternatives to Android altogether, you might want to see if there’s a port of a Linux distribution like postmarketOS or Ubuntu Touch that can run on your device.
via Bleeping Computer and Google (1)(2)
If Google is concerned about security and Android fragmentation, this is the time to set out clear rules. While putting Gingerbread to bed, they could also announce that Honeycomb will be retired at the end of the year, and Ice Cream Sandwich six months later, and so on.
That doesn’t achieve much, if anything.
It’s a much more complex and tougher problem. I can’t figure if Google was a genius, naive or greedy, when it comes to the Android ecosystem. They were naive to make it open-source and think competitors won’t take advantage of the goodwill (BBK, Huawei, Xiaomi) and make their own forks, or act not in the best interest of the company, consumers, or law. Or greedy that they specifically made it good for everyone to have rapid-deployment and almost monopoly of the computing marketshare. Or geniuses because they knew they wouldn’t get in trouble as a monopoly if they’re open-source, or that these teething issues are pointless if they can merely get consumers adapted to upgrading their device once every year or two.
Solutions?
Simplest: Google can bring Android back to close-source if they wish, that is one option, where they can regulate software and services much tighter like Apple.
Most complex: Or Google can stop working on Android10-versions, and throw all their developers, money, and years into crafting a new, clean, much more advanced version (eg Android v20) but it requires licensing by OEMs and strict limitations. They can let all the “losers” flounder with the old AOSP. So they essentially have two branches of the software, the paid/licensed version which is now the market standard/expectance, and the lower quality software in the free open-sourced version.
Least likely to succeed: Google can put monetary incentives for OEMs to sell phones on the latest version, having fewer bugs/bloat, faster security updates, and mandatory platform updates for a set number of years. These incentives sound great, and is ethically the best practice, but practically won’t achieve much if most OEMs get their profit margins from hardware sales and not reduction of licensing costs.
Simpler solution: take the goals from your incentives plan but take out the incentives and replace it with a big stick, namely licensing for their closed-source Play Services. Nearly every manufacturer licenses these for devices sold outside China. A user can try to sideload them if they’re not present, but if they don’t know how to do it, they’ll see lots of apps break including many of Google’s own. That’s why this has become a nice income stream for Google. They already are able to control manufacturers by restricting the licensing term on Play Services, E.G. not allowing them to simultaneously sell Googly Android devices and plain AOSP Android devices. It would be easy for them to start adding terms to that contract. It would have been similarly easy five years ago. They chose not to, and they still won’t, but if they did, that would be a faster and cheaper solution.
There’s over a Billion population in China, and they’re not poor. That’s a huge demographic to throw away.
So what this allows is for another company to piggyback off Google’s work. They basically take the massive amount of work they put into Android 4.0.3, or worse Android 5.1, or worse Android12 (the latest), and fork it.
Google’s Services? Don’t need em.
You have decent competitors in the West, and better competitors in Mainland China. So what can/will happen is these powerful competitors (eg Baidu) will simply buy up or get into an agreement with a Western competitor… and now you have a problem. And by the time Google tries to do something about it, it is simply TOO LATE. They now have momentum and market share, and they can continue development of their fork, and continue development of their services. Thanks Google you just created your rivals, with the largest caveat being that they do not share your values and beliefs (eg Do No Evil). This could end up very bad for governments, and very bad for people (information war -> actual war).
It was “don’t be evil”, they replaced it with the more nebulous “do the right thing” after people kept pointing out the ethical problems with their datamining.
Just the datamining though, as anyone who disagrees with google on anything else is a monster.
People have been predicting that for years. It doesn’t happen. It also won’t happen. Huawei has a set of services they’re trying to sell right now. Unsurprisingly, they haven’t gotten other manufacturers taking them up on that.
People don’t want to do the work of changing services to something that won’t work as well when they have the option to give people Google’s and make that Google’s problem. They comply with the restrictions that currently exist and they would put up with some more before it would make sense to look for an alternative. China, meanwhile, has at least five competing versions because companies don’t want to hand major advantages to their rivals, so they’re not about to agree on a single standard and come for the rest of the world.
postmarketOS is based on Alpine, which is decidedly not GNU/Linux. It’s just Linux.
Interesting. Having been called out for just writing “Linux distributions” a few years ago, I’ve taken to throwing in “GNU/Linux” when writing about most desktop Linux distros. But a little digging suggests you’re right, postmarketOS and Alpine might be examples that really are best described as Linux distributions. I’ll update the article. Thanks!