The Tortoise and the Hare: The tale of Android evolution
[Android is moving too fast with software releases – too fast for the smartphone ecosystem to follow. At the same time, Android is moving too slow, as CE vendors are taking it outside of its mobile comfort zone with the introduction of form factors from tablets to in-car terminals. Guest author Tsahi Levent-Levi outlines the market forces straining the Android ecosystem and Google, as it moves away from smartphones to additional devices.]
Android is all the rage these days. In my meetings and correspondences with consumer electronic vendors around the world it is as if they have totally forgot about the “old“ “embedded operating systems” – pSOS, VxWorks, MontaVista, Nucleus, OSE, or any of the Linux and Unix variants that people have been using for years now.
While there are a few Meego strongholds and some Embedded Linux developers, most of the market has shifted to using Android. And it’s not just about mobile phones. It’s televisions. And tablets. And media phones. And set-top-boxes. And DECT phones. And DVRs. And Digital Picture Frames. And In Car navigation and entertainment systems. Every device that has a screen is now a prime suspect for migrating to Android.
Chipset vendors have taken notice of Android. Chipset vendors who aren’t catering for mobile devices had no Android in their near future for plans for early 2011. That was 3 months ago. Today, these chipset vendors are joining the bandwagon and are updating their roadmaps and strategy by embracing Android – they have figured that it is better to join the club than to fight the tide.
The Hare: Moving too fast
While this is happening, Google is shifting gears. In 2010 they have shortened the release cycles for many of their products and are raising a new challenge to companies who wish to stay ahead of the game and compete in the market.
With 5 or 6 releases of their operating system in a single year, it may seem that Google is moving too fast with Android. While that is definitely true, Google and Android are also moving too slow at the same time.
If you look at the mobile handset arena, Google is definitely not waiting for anyone.
The sheer amount of releases places handset vendors in an uncomfortable position of being unable to follow suit. Sony Ericsson released their Xperia X10 with Android 1.6 on August 2010. Dell out-did them with Dell Aero running Android 1.5 on August 2010. Older devices were launching with Android 2.1: Motorola Droid X released on July 2010 and HTC EVO released on June 2010 are such examples.
At the same time, Google has had to cope with different implementations of their API set for developers by the different handset vendors through their CTS (Compatibility Test Suite) program.
These changes between Android versions are not only additions – some of them are infrastructure changes that affect developers and break compatibility across versions. Take for example the addition of Stagefright – a new media framework released alongside OpenCore in Android 2.2 – will Google be keeping OpenCore moving forward or will they deprecate it in future releases?
Andy Rubin, VP of Mobile Platforms at Google said in an interview that their launch cycle “will probably end up being once a year when things start settling down”. Is that going to happen any time soon with iOS innovations and the introduction of Windows Phone 7? Unlikely.
The Tortoise: Moving too slow
On the other hand, Google hasn’t been able to address the hockey-stick market demand for the Android platform.
Back in 2007, Google created the OHA (Open Handset Alliance) consortium as a governance framework where Google could establish handset compliance requirements and thereby run the show (see their CTS and CDD requirements recently published. Following the same philosophy, they set up Google TV for Android-powered televisions. The next product category that Google will focus on will be tablets. But what about in-car systems, set-top boxes or media phones? Enter the OESF.
The OESF (Open Embedded Software Foundation) is an open alliance formed in Japan and active throughout Asia Pacific. It is the first non-Google consortium initiative for Android. Its charter is to define new API sets that cover the products that Google doesn’t. In that regard, the OESF has already introduced its own Market Place SDK and is making strides in areas related to home networking, VoIP communication, security stacks, automotive and more.
Google have decided in the past that tablets should be running their Chrome OS – a networked based operating system – and not Android. They also stated that vendors should wait for Honeycomb Android release and not use FroYo or Gingerbread for tablets. Vendors have not been convinced, preferring to use Android instead, with its currently available version. In September 2010, during IFA Berlin , a slew of new Android-based tablets have been introduced: Toshiba Folio 100, E-Noa’s InterPad Android tablet, Elonex eTouch tablet, ViewSonic’s ViewPad 7, Archos’ tablets and Samsung’s Galaxy Tab. Deutsche Bank’s Jonathan Goldberg has compiled a list of 30 tablets planned to launch by the end of this year alone.
The Samsung Galaxy Tab released to the market with much fanfare last month is the first Android tablet that comes from a large vendor and backed by Google through its Android Market. This clearly shows Google’s new stance with tablets. The application layout issues that are expected with this tablet due to different resolutions than those available on mobile phones are going to cause headaches to both users and developers in the short term.
Factor into it the growing hype in China around Android and we are bound to see innovation happening out of Google’s campuses around Android.
Will these issues be solved in Android’s next release – Gingerbread, or only in the one after that – Honeycomb? Will Google try pushing vendors to Chrome OS instead for tablets? These open ended questions show how slow Google is in addressing non-smartphone markets.
This issue of form factors is the second dimension of Android’s fragmentation. There are three more dimensions: implementation fragmentation, user experience fragmentation and codebase fragmentation. If Google wants to retain their control over the Android platform, they will need to solve all of these five dimensions of Android fragmentation.
The crystal ball
Google is moving fast with Android and at the same time are trying to solve fragmentation issues of their platforms: they are working hard on reducing the amount of handsets running older versions of Android, they are trying to solve implementation fragmentation with their CTS suite and they are now focusing on user experience issues.
It is not going to be enough. The Android platform has captured CE vendors of all types. Any device requiring a user interface to operate is either moving to Android or will move to Android soon. By ignoring these devices, Google is leaving a wide door open for other vendors and organizations to cater for their needs: the OESF are doing that on the standardization front, while new entrants to this market such as Amazon may become the ones providing the application stores for such devices.
At the end of the day, Google will be able to focus and control a relatively small number of form factors: smartphones, televisions and maybe tablets. The rest of the market will be using the Android platform without Google’s direct assistance and control; we should see other application stores enter this market, which is a genuine opportunity for the likes of the Amazon app store (Android-based, white label Kindles, anyone?) and all the other service providers out there to compete with Google’s services on Google’s own home turf.
[Tsahi Levent-Levi is Director of Technology and Solution at Radvision. He has been involved with the mobile video telephony market for 8 years, dealing with design, development, standardization, interoperability and marketing of such technologies. You can follow him on twitter or through his personal blog at http://blog.radvision.com/voipsurvivor/.]