Why Ice Cream Sandwich won't be able to save Android tablets

While more responsive and with an improved browser, Ice Cream Sandwich still doesn't address the fundamental flaws with Android on tablet computers.
Written by Jason Perlow, Senior Contributing Writer

For the last day, I've been tinkering around with Google's Ice Cream Sandwich (ICS) 4.0.3 OS on my Motorola XOOM.

Motorola hasn't yet released the updated software for its tablet officially, but I was able to get the latest version of Android running due to work being done by various open source community teams at XDA-Developers, a popular forums site for Android development and hacking.

I've actually managed to try three separate ICS builds for the XOOM, just to get a sense of what stage the code is currently in. Motorola is reportedly now testing its official software release for their tablets with a pilot group of users, so that means if you own a XOOM, you should be receiving it via an over-the-air (OTA) update within a month.

Right now, all of the builds that are out from the community are essentially "Vanilla" based on code from the Android Open Source Project (AOSP) although they have a few additional tweaks for things such as overclocking if you really want to dive into that sort of stuff.

There are other community-supported unofficial ICS builds out for other tablets such as the Samsung Galaxy Tab 10.1 and the Asus Transformer -- all of which of course require "rooting" your device and installing ClockWorkMod on it, an open source firmware manager utility.

Unless you know what you're doing, I don't suggest going this route with your Android tablet.

Currently, the three "team" builds I have looked at do not have functioning cameras/HDMI ports and have a number of other minor issues, but for the most part the aesthetics and general operation of the software more or less closely reflects what will end up being released within several weeks by Motorola as well as by other Android tablet OEMs.

With the acknowledgement that what I'm playing with right now does have bugs and in no way should be considered a production software release or an officially supported build by Motorola, overall, Ice Cream Sandwich is definitely an improvement over Honeycomb.

The software runs considerably faster, the user interface is more responsive overall and the browser renders pages more fluidly, which has always been one of my major complaints about the OS.

However, while existing Honeycomb tablet owners will see this software as a welcome improvement to what they were using before, I don't see Ice Cream Sandwich as being some sort of magic bullet that is suddenly going to propel Android tablets into major market share territory (with the one major exception being Amazon's Kindle Fire).

Right now, all of the Honeycomb tablets currently in use which are due for the ICS upgrade only occupy about a 3.3% share of the total Android install base. That's not a heck of a lot. Most of the Android that's out there is running on handsets.

Amazon's Kindle Fire, which has supposedly sold millions of devices over the 2011 holiday season, runs on a modified version of 2.3.5, aka Gingerbread. As such it's heavily supplementing the large percentage of 2.3.x smartphone devices that are already in circulation as reported by Google -- most of which will not be upgraded to ICS.

[Note: If you didn't understand that last paragraph, it means that the Kindle Fire is not even being counted in Google's metrics, but by virtue of selling millions of units, it vastly increases the amount of 2.3.x Gingerbread in the wild. Capische?]

Also Read:

The last time I compared Android as a tablet OS versus Android as a smartphone OS side by side back in April of last year, the two implementations were not at version parity. Until now, the most current smartphone implementation was version 2.3.x (Gingerbread) and the most current tablet implementation was Honeycomb (3.x).

Also Read:

The initial release of Honeycomb 3.0 was absolutely rife with problems. The OS had all sorts of application compatibility and overall stability issues which made the original XOOM I owned (and subsequently returned) a nightmare to use.

Over the course of that year 3.1 and later on 3.2 improved stability issues considerably. For testing purposes, I tried my luck again with the Wi-Fi version of the XOOM, when my ZDNet colleague Scott Raymond decided to sell me his for a generous discount when he decided he wanted a thinner, but very similar Samsung Galaxy Tab 10.1 instead.

Boys must have their toys, you know.

Had Scott not decided to give me a nice break on his barely-used XOOM, I would have never spent full price on the tablet, even at the $500.00 price parity with iPad 2 they were selling the devices at the time.

Today, you can pick up a 32GB Wi-Fi XOOM for about $450 and a 16GB Galaxy Tab for about the same. The Asus Transformer which is based on the same nVidia Tegra 2 guts will run you about $400.

The Asus Transformer Prime, which sports the new nVidia Tegra 3 Kal-El quad-core processor, will set you back about $500. Like the XOOM and the Galaxy Tab 10.1 it's also due for a Android 4.0 update shortly.

[EDIT: I had previously stated that the Transformer Prime cost $600, but that was for the 64GB version.]

With the exception of Amazon's Kindle Fire, I still think all Android tabletsareseriouslyoverpriced. Software update or not, that's still going to be a major issue with consumers making decisions about what tablet to buy this year.

At least one major Android tablet manufacturer that I know of is due to release an 8GB 10.1 inch device in the next several days in the sub-$330.00 range. I think that's a good start, but it's not enough.

So then why is everyone so jazzed about the new software update? It's because Google's Ice Cream Sandwich (ICS) version 4.0.3 finally brings Android smartphones and tablets into a single unified codebase.

[Next: Android Code Unification does not equate to Market Expansion]»

One of the prime benefits of this code unification is that there is much more consistency in the user interface between phones and tablets, and obviously both smartphones and tablets that use this version will support the same set of Application Programming Interfaces (APIs).

So over the long term as Gingerbread and Honeycomb devices get updated/discarded and new devices replace them, and the existing catalog of Android apps get ported to the newer APIs from the older ones, the fractionalization/fragmentation, compatability and software update issues will supposedly diminish.


Also Read:

Since I also own a Verizon Galaxy Nexus, I was able to compare it side by side with the ICS builds on Motorola XOOM. I can report that at the very least, the configuration screens look pretty much identical, and for the most part there is a large amount of consistency in the UI dialogs on applications that run on both the smartphone and tablet builds.

Unfortunately, a lot of the things I didn't like about Honeycomb are still present in Ice Cream Sandwich. Even though my phone now runs the same software as my tablet, I still have that sense of "mental context switch" between a tablet interface and the phone interface.

It should be noted that this doesn't happen on Apple iOS devices, which all look and work exactly the same even if you are using an iPhone, an iPod Touch or an iPad.

The launcher UIs used on the smartphone and tablet versions of ICS are different, and it's more than just a little annoying because you can't alter placement of the virtual buttons (Home, Task Switch, Back) or the main "Apps" button which is still on the upper right of the screen, along with the "Options" button that appears under certain application dialogs.

There are legacy launchers that were developed for Gingerbread such asADWLauncher EX and LauncherProand a new ICS launcher that is being developed by the CyanogenMOD 9 project called Trebuchetthat can be used, but most end-users will not opt for these customization solutions.

And while these alternatives solve the App launcher issues, they don't necessarily solve the placement options of the soft buttons which cannot be moved.

And the notifications? Let's just say that on a tablet, they work damn awkward when compared to the smartphone implementation on the Nexus. Frankly, I think even iOS 5 on the iPad has ICS on the tablet beat.

I was really hoping we'd get pull down notifications instead of that weird "stack" on the lower right hand corner which in my opinion is an exercise in pure frustration, especially if you have a lot of notification events from apps piling up. And jumping into the Settings dialog is still a royal pain unless you make a dedicated shortcut to it on your home screen.

Ultimately, this all boils down to ICS being a less intuitive tablet OS which is harder for the average user to understand than Apple's iOS. On a subsidized smartphone device with 4G LTE technology, an end-user is likely to tolerate these kind of geeky issues, but on a $400.00-$600.00 tablet, I'm not so certain.

ICS has done absolutely nothing to make Android less geeky or more user friendly. That alone is worth consideration.

It could be argued by Android evangelists (and this includes myself in that group) that ICS is far more configurable, customizable and more flexible than iOS. This is certainly true. But I'm not sure that necessarily is going to translate into full-size Android tablets moving from a 3.2 percent market share into say, a 20 percent share.

While I expect ICS to do extremely well in the handset space in 2012, there is the small nagging issue that unless we see a significant percentage increase in tablet market share with ICS devices, the tablet-optimized "App Gap" is still going to be a problem for Android for some time.

Android now has over 400,000 apps, but only a tiny percentage are actually designed for Honeycomb and ICS's native APIs and screen formats. The balance of the apps in Google's Android Market have been written for smartphones.

Sure, they'll run on tablets, but the experience on most of them won't necessarily be optimal and a good amount of them will flat out break, even with the new screen-scaling features present in Honeycomb 3.2 and ICS.

Compare this with iOS, which has well over 140,000 native iPad apps, and this app gap becomes far more of an issue when it comes to the platform's attractiveness to consumers relative to the competition.

Sure, Amazon could use ICS in the next Kindle Fire and sell yet millions of more devices, but is that really a win for Google and ICS or is that a win for Amazon, especially if they launch a product that is equally customized as to make Android nearly invisible to the end-user?

And while I'm beating up on full-size Android tablets and ICS -- I might as well say it, because I've always thought this was a stupid choice from the very beginning: why the heck did Google choose a 16:9 aspect ratio for their tablet reference design?

It's nice when you're watching widescreen movies on Netflix, but using these larger devices in portrait mode for book or magazine reading is an exercise in pure suck. With a 7" or 9", it's tolerable. But all those driving games that I love on the iPad's 4:3 aspect ratio would never work well on a 10.1" 16:9 screen because it's completely off-balance.

Say what you want about Apple's walled garden and locked-down OS -- they got the ergonomics of tablets right, and Google has it all wrong.

So between the cost of these devices and the various usability issues and the tablet app gap I've pointed out above, I project that 2012 will not be a good year for Android tablets, unless they are made by Amazon.

Will Ice Cream Sandwich translate into significantly improved sales for Google experience Android tablets, or will they continue to have a lackluster showing in the marketplace? Talk Back and Let Me Know.

Editorial standards