Android Problems and Their Solution
In my opinion, +Android
is the best mobile OS currently available, but it has grown to such a size that the same approach that was previously used to grow adoption is becoming problematic now. Here are the problems I see and the solution I think +Google
should adopt to fix them:The Problems
1. MADA: http://goo.gl/xmhV6S
In an attempt to reduce fragmentation somewhat, Google has strict control over what apps get installed, and where those apps are located on the home screen in order for the device to get the +Google Play
Store/Services. Ultimately, while this did ensure that users could pretty much always be assured that they could find whatever Google app they were looking for on their device and always had the most up-to-date anti-virus possible (the specific type of fragmentation that this is targeted at), it also resulted in a PR and legal nightmare (often being compared to Microsoft's monopolistic behavior with IE).
2. Carrier/OEM motivation to add spyware (a la +Lenovo/ #Superfish ): http://goo.gl/sIw2vD
As Ben Thompson (from www.stratechery.com
) argues in the video at the link above, while Lenovo is most certainly in the wrong for what they did, they were also financially motivated to do what they did. The same is true of Android in its current form. OEMs make little off the hardware they sell, which results in them having a motivation to do things that compromises their user's security (i.e. not provide OS updates and include spammy/questionable software).
3. Bloatware: http://goo.gl/XN6mAO
As the link above explains, bloatware is any unwanted app that comes pre-installed on the device and cannot be removed without root privileges. This includes different apps for different users (yes, even Google apps in some cases), but ultimately, whichever apps they are, they are widely despised among users. The reasons for these apps being included are diverse. Most OEMs include duplicates of many of Google's apps as well as a few unique apps because they the more people use and enjoy their apps, the more likely they will be to buy their next phone from the same place. Google wants their apps included because that is how they monetize the Android platform, and if they were not able to monetize it, then it would not exist in the first place. But ultimately, both companies want users to use their apps and services over the other pre-installed alternative because that behavior ultimately makes them more money in one way or another. The problem is that these apps take up valuable space on the device (+Samsung Mobile
is especially notorious for using up as much as half the device's memory with bloatware), and cause the device to be sluggish and a horrible experience. +Motorola Mobility
's recent practice of having the minimum amount of bloatware on their devices (as they describe here: http://goo.gl/6gRJbj
) has resulted in a rapid growth in the sales of their devices. This is great, but clearly, the other OEMs either haven't noticed or want their devices to be differentiated so much that they are willing to sacrifice the speed of their devices to get it.
4. UI Fragmentation: http://goo.gl/B4MEAh
The vast majority of Android users either don't even know they are using a device with an OS made and designed by Google or they blame the faults of parts designed by the OEM on Google. They can hardly be blamed for making these mistakes. The relationship between Google and the OEMs is complex, and the UI across various Android devices is often inconsistent. A person may expect that moving from one Android phone to will result in a similar experience and as a result either be unpleasantly surprised (due to having a good experience on one Android phone and moving to a bad one on a different Android phone) or give up and just move to an #iPhone
(due to having a bad Android experience and thinking all Android phones are the same).The Analogy
I think it's helpful to make an analogy between phones and cars. AOSP makes up the drivetrain, frame, and a few body panels of the car. The OEMs skin and Google's apps make up the rest of the body and interior. The car also comes with additional "body panels" and "interior pieces" in the trunk, but while you may be able to trade out some of the default body panels and interior pieces, you can only add to what you have. You can't really throw anything away. Google has recently been paring down the body panels and interior pieces that come included in AOSP and moving them to a place where a whole new car doesn't need to be replaced to get the newer version of one particular one. This helps some but ultimately, only fixes part of the problem.The Solution
I think the solution is to give people the option of what body and interior pieces they want to have on their car when they first buy it (i.e. when they set up the phone). This would mean that the OS would come with only a "drivetrain" and "frame," and the user would choose everything else that went on it.
In order to do this, a few things would have to change:
1. The Play Store:
Google would need to set up a means for apps (whether from Google, and OEM, a carrier, or any other developer) to be installed as a bundle all at once. This appears to be already being done/worked on, at least in part, with users being able to install apps from previous devices in mass, and employers soon being able to install a collection of apps on their employee's devices all at once. It just needs to be extended such that any developer could create their own bundle of apps for users to install. An alternative implementation could allow a user to install all the apps from a particular developer at once. Obviously, either way, the user should have the option to select or deselect apps from the bundle that they do or don't want to install.
Google needs to allow more parts of the UI to be customized with apps from the Play Store. This can already be done to a decent extent with third-party lockscreens, launchers, keyboards, and other similar things, but they need to finish the job. Currently, the following things cannot be customized with an app from the Play Store: the notification shade, the app switcher, the settings UI, and possibly a few other things.The Results
While it's ultimately impossible to precisely predict everything that would happen as a result of this, I think several things are likely:
1. Google could no longer be excoriated for being monopolistic since they would be allowing people full control over what was installed, and whatever succeeded would succeed based purely on its merits.
2. Spyware/Adware would be nonexistent because it would be quickly identified and removed and ultimately likely result in that developer's bundle getting boycotted.
3. Bloatware would get eliminated because everything except the absolute essentials could be uninstalled.
4. The differences in user experience would only exist between developer bundles, and the distinction between what is made by Google and what is made by others would be made extremely clear.The Problems
As with any change of such scale, there are obviously potential risks and problems. With the current set up, both Google and the OEMs benefit to a certain extent, but they also hinder each other due to the competition for power over the user's attention.
The benefits to Google with this change are pretty obvious. They would fix all the problems listed above and thereby improve the user experience and resolve a significant PR problem. While it is true that they could potentially lose some market share (with regards to their apps/services) due to the reduced influence over what the user has on their device, I think the benefits would outweigh the costs. They would need to be careful to add some checks to make sure various apps are present if certain apps are required to work properly together (though this is is largely handled by the option to change the default app for some action).
I think if anything would cause this approach to be infeasible it is the fact that Google is generally perceived to have better apps than the OEMs. Therefore, people would probably be far more likely to choose Google over (for example) Samsung apps, unless Samsung spent the time to differentiate its apps from Google's, which it isn't really doing currently because it has no motivation to do so. Keeping the status quo would be the safe bet for the OEMs because that would ensure that they would still be able to have their apps and UI differentiated from everyone else's and placed front and center before the user. They also may not be willing to compete with Google. However, I think they could still differentiate themselves from other OEMs as well as Google and actually increase sales as a result.
They already make camera apps with more extensive features than Google's app. They could extend this further, and also add apps that take advantage of their unique combination of sensors and connectivity chips. The ultimate result would be that competition among the various app bundles would result in those app bundles improving and making the user experience better as a whole while also encouraging the OEMs to focus more on adding more and improving existing sensors and connectivity chips so that they can differentiate from competitors with high quality apps and features.
To say the least, it would involve some risk on everyone's part, but I think the benefits for everyone would far outweigh the costs.
Tagging: +Brandon Lall
, +Michael Webber
, +Jake Weisz
, and +Sundar Pichai
Reference post: http://goo.gl/taZ4S3