An Open and Dangerous Playground

Mobile Mar 24, 2011 No Comments

Ladies and gentlemen, RIM just announced that they are going to allow Android apps to run on the Blackberry Playbook. Let that soak in for a moment. Your favorite Android tablet apps, every single one of them, will be able to run in emulation on the Playbook. All of you out there who are dying to get your hands on a Playbook and make it your own, you, along with all of those many Android tablet owners, will be able to run those Android apps. You won’t have to go out and buy a Xoom or a Galaxy Tab to buy into that ecosystem. RIM is providing a way for you to have that Playbook AND the apps you really want. It’s like the best of both worlds, assuming that you use a Blackberry as your main phone and will have it in order to use the email and calendar clients.

Or, if you’re like just about everyone else out there, this is a completely meaningless development that is worth noting not for it’s own merits but as an indicator of just how sad a state RIM finds itself in.  For months now, people have been wondering if RIM would do something like this.  Since Android is built on Java, it’s always been theoretically possible to run Android apps in an “Alien Dalvik” layer that runs apps in virtualization.  Running the apps in virtualization is not as unsatisfactory as it might sound, since all of Android is a virtual machine and the apps are essentially “virtualized.”  The advantage of allowing the Playbook to run Android apps is that there would theoretically be an ecosystem of 200,000 existing Android apps that could be easily ported over and submitted to the Blackberry App World.  There are a myriad of problems with this plan, however, from the perspectives of Android developers and even RIM itself.

For starters, the opening of this article is actually a bit misleading.  You can’t currently port Android tablet apps over, because the guidlines outlined by RIM only mention Android 2.3 apps.  Android 3.0 Honeycomb is the first true tablet OS from Google and includes its own unique SDK for apps.  The experiment of putting Android 2.2 (designed for a phone) on a tablet was already tried by Samsung, resulting in a lackluster Galaxy Tab and a complete about-face as soon as they had Honeycomb with which to begin building their Galaxy Tab 10.1.  Ordinary Android apps don’t work well for a tablet, even on a diminutive 7-inch tablet like the Playbook.  So right out of the gate, you have the entire purpose of using Android apps drawn into question.  It seems certain that RIM will eventually allow Honeycomb apps to work on the Playbook, but that time hasn’t yet come, and the entire point of this strategy was to have useful apps at launch.

What happens, then, when RIM does build in the tools to use Android 3.0 apps on the Playbook?  Won’t that fix all the problems that were just mentioned?  That all depends on whether or not there are going to be many useful Honeycomb apps.  The Motorola Xoom just launched and has barely any apps or developer support behind it.  If development for Android tablets follows the same path that development for Android phones took, then it isn’t going to be until a significant amount of hardware sells before developers take it seriously.  Right now, there isn’t a single Android tablet that looks to be remotely competitive to the iPad. So where does that leave the state of tablet-specific Android apps?  In the starting blocks, just like the Playbook itself.

Furthermore, as RIM adds yet another way to build apps that the Playbook supports, what is the clear message in terms of what frameworks developers should use to write Playbook apps?  They’re allowing people to write in Adobe Air and sort of get by with Android apps.  Where is the incentive to write natively for the Playbook?  It doesn’t exist, and probably for a more fundamental reason than just having too many options.  The tools that RIM is providing for their new QNX framework must not be that compelling if they’ve already lost confidence in their ability to convince developers that it’s an environment worth investing in.

And finally, the reason this is a bad idea is the thing that should have been considered first.  What kind of experience is this really going to be for consumers?  Playbook owners, however few there might be, will be greeted with a complete mashup of Blackberry Java, Adobe Air, and Android apps, none of which will be particularly strong on their own merit.  And since the Playbook does actually do real multitasking, how is a virtualized app, designed for a different OS and hardware, going to perform and affect battery life?  Battery life has already been identified as a weak spot for the Playbook before they can even get it to market, but RIM doesn’t seem concerned that these Android apps will likely tax the system in ways that true native development wouldn’t.  How can anyone respect this device if RIM can’t show any true ownership of it even at launch?

The reality is that nobody can.  The Playbook absolutely demos well and has shown tons of potential in its interface, but like so many other products, the experience of actually using it over time will likely tell the true story.  Since RIM went out of their way to steal so much from Palm’s WebOS, it seems only fitting that after an initial flurry of hype, this device is also likely to trip out of the starting blocks and watch the competition run out of sight.

No Responses to “An Open and Dangerous Playground”

Leave a Reply