As always, everyone forgets to mention the poor, slow and by now outdated Android WebView - which by the way will finally be getting a very important upgrade in the form of WebViewFactory. This may allow using alternative WebViews, such as a chromium-based webview - another important step towards defragmentation.<p>Some more info is available at <a href="http://danosipov.com/?p=572" rel="nofollow">http://danosipov.com/?p=572</a>
It's worth noting how closely Android's apparent path going forward mirror's Apple's strategy with both OS X and iOS: closed end-user components built on top of an open source core. I guess Google's realized that "openness" isn't a meaningful market differentiator for the average consumer.
This is an interesting move on Google's part. It is a lot more mature (from the perspective of running a software business) to have this level of control over your destiny. It is also another step away from the engineering driven focus toward the business goals focus. One of the more difficult things to do is have the conversation why something is correct for the business, even if it is an inferior engineering solution. I hope they manage that transition well.<p>The other part of this story seems to be that if you move to this as your "api" then you can move the underlying OS tech to something else as well. (away from Linux to something perhaps smaller for a budget phone.) Should be easier than it was for Microsoft to move everyone to the WinAPI and off DOS.
Smart move by google. Potentially they've regained control of their platform past, present and future...<p>In one fell swoop they can resolve fragmentation, keep everyone up to date on functionality/APIs, bypass OEM/carrier update issues (be it carrier QA, OEM lack of interest, or just crapware). Most importantly google can keep any OEM's (ie samsung) in check and prevent them branching away from android.<p>The downsides: what happens when google break an entire range of phones/functionality with an update? Are they going to QA every android product in existence?
Will older models slow down and become unusable as the baseline resource requirements increase? (I'm looking at you IOS7)...
So Rubin departs leadership of Android just before Android becomes "not open". Seems like there's a backstory here that I'd love to hear about.
So, the ultimate nail in the coffin for open, fragmented services. If you want to build a good OS, you ultimately have to take almost complete control. (Google is basically following Apple's route.)<p>The reason I think the Google system is worse is that theoretically <i>any</i> application can achieve this level of control. So while I might trust Google, it seems all too easy to trick users into installing a different process which can take complete control of their phone. With iOS, only Apple is capable of this level of control, thus decreasing the attack vectors.<p>EDIT: Admittedly it's highly unlikely that Google would ever let such an application into the Play Store, thus limiting its distribution potential.
I think this is great for Android as a whole, users will be better served.<p>It does cut into the openness of Android though. Unless Amazon bends to Google's way of doing things a little bit their platform will stay even further behind, for instance.
Only boring updates in 4.3? Have you seen app ops [1] functionality? Hands down the greatest thing to happen to a mobile OS in a long time. I know there's Xprivacy, but having this thing built in is awesome.<p>[1] <a href="http://www.androidpolice.com/2013/07/25/app-ops-android-4-3s-hidden-app-permission-manager-control-permissions-for-individual-apps/" rel="nofollow">http://www.androidpolice.com/2013/07/25/app-ops-android-4-3s...</a>
I don't understand what the all negativity over these changes is about. Has a significant amount of mainstream Android features being incorporated into play been from the ecosystem, rather than Google? If so, I can see how this could result in a worse end-user experience. But if not, how does providing timely updates and a less fragmented ecosystem such a horrible outcome? Feel free to enlighten me if I'm missing something.
It'd be interesting to see how it pans out for OEMs that don't, for various reasons, bundle Play Framework/Services with their phones. This could be fixed by having the user install a GApps zip through recovery, but it'd be a pain to the average user. Also, are there, or will there be, any manufacturers forbidden to use the Play Store and its framework? It seems for the vast majority of people, this move is great, but for a tiny minority, i'm not so sure.
The elephant in the room here is security.<p>A lot of security updates can be pushed out for the apps and Play Services. However, the underlying OS still needs regular, frequent updates to stay secure, and that's just not happening for the vast majority of Android devices. Who needs 0-day exploits when you have millions of Android devices out there that are vulnerable to exploits that are one, two, or even three years old!?
"It has its own silent, automatic update mechanism that the user has no control over. In fact, most of the time the user never even knows an update has happened."<p>Great. So, if it was difficult to trust your phone/tablet/tv <i>before</i> it is now impossible. I'm looking more forward to Ubuntu and FirefoxOS than ever before - I want to be able to control what software runs on my devices!
A closed source app that can update itself without notifying the user, and has full permissions? How is this any different from the 'loss of control' to American institutes, like the Windows 8.1 / TPM 2.0 warning from the German government. Isn't this exactly the same thing?
>>This is how you beat software fragmentation.<p>This does help defragment Android from a Google (developed) app perspective.<p>However, it doesn't do anything for 3rd-party developers. It doesn't do anything for regular consumers who wish to upgrade their 2.3.3 phones or 4.1 phones to the latest version of Android. So Google has a long way to go before they really defragment Android.
i think this will ultimately prove to be a terrible move for google. android's success is largely a function of how attractive it is to device manufacturers and carriers. stripping control from them is going to force them into a corner where forking android or moving to another platform (amazon's fork of android, firefox os, windows phone 8 and ubuntu mobile are all potentially technically competitive even if they are not currently feasible).
It seems that this article doesn't really separate between user concerns (having the latest google apps) and developper concerns (access to latest APIs).<p>From what I understand, this means developpers will be able to rely on android "services" being updated more often and more consistently available.<p>However, what about things like using new UI components ? For this, you still need the latest OS version, right ? If that's the case, then maybe this would mean users caring even LESS about updating their OS, and thus an even worst situation.<p>i suppose Google thought about it, so there's probably something false in my statement, could anyone confirm ?
Hackers need to do something about the big companies abusing the app store system. Google, Apple, Amazon are taking far too much fees from independent developers.<p>- Apple's store especially needs to be killed fast for innovation to thrive, Apple is turning out to be the new Microsoft, Tim Cook should fear the wrath of developers<p>- and Google needs to be reminded that it's business could swiftly vanish if it isn't pro-developer and pro-consumer, a 30 percent cut for what amounts to hosting a file is way too much.
I'm sorry but this makes it sound like Google has come up with this magic way of beating fragmentation. It shouldn't have happened in the first place.<p>Contrast this with iOS upgrade path or what webOS had been doing. Seamless (for the most part) and a great UX.<p>Google is able to do this; see Chrome. What is strange is why they did not adopt the same strategy for Android as for Chrome.
Aside from fighting over whether Android is closed or not this is an interesting solution to dealing with the inconsistencies present in the Android universe. The freedom everyone wants to love that comes with a fully open source Android platform is, in reality, a real pain when you're trying to develop apps. There is at best a plurality of different operating system versions. Phone manufactureres spontaneously stop supporting devcies, only pushing out updates to the newest of the new. OS improvements are lost on those devices left behind until the users can afford a new device or reach the point where they can upgrade.<p>Unlike Apple, where the iOS landscape is 100% consistent (minus physical hardware variations) Android is a mess. Before going to town on Google for "closing" the operating system, let's at least look at the problem they are trying to solve and how this attempt - I'm not saying it's correct, or good - addresses those issues.
Google/Android really can't win now that this "fragmentation" frame has stuck.<p>Google's been updating core parts of the system like this from day one. The Android Market/Play Store, Maps, Youtube and many other things have received massive updates and been pushed out to millions of people. However, the dominant storyline at the time was that "Android devices don't get upgrades, because fragmentation".<p>Then suddenly, for no obvious reason, the tech bloggers all noticed at once that this was happening (possibly because they had started actually using Android?) and it was portrayed as sudden U-turn by Google. A weapon in their "war" against Samsung (don't get me started on that one).
Notably, Play services is totally closed. So Google has abandoned the open source part of Android and is now developing the operating system as a completely closed product.<p>Edit: downvotes don't change the truth of the observation. Android is no longer meaningfully open, other than a years old core of basic functions. Just like OSX and Darwin.