A True Open Mobile OS

  • [アーティクル]

Last week Commnexus invited me to participate in a panel to talk about Mobile Operating Systems moderated by Andrew Seybold. This is a follow-up to a similar panel Andy did last year focused on mobile Open OSes, I guess motivated by Google's launch of Android. I blogged after that first panel posting some thoughts about openess.

In the last few weeks I have read a few articles recognizing Windows Mobile as the most open operating system available.Take a look at this review from PC Magazine that ends with the following paragraph: 

Ultimately, Windows Mobile keeps our Editors' Choice because, right now, it's the only truly open OS that embraces the entire marketplace. With Windows Mobile, if you don't like your browser, you can download a different one. If you don't like your carrier, you can switch. If you have a dream for an application, you can program it. That openness, flexibility, and range of choices is what keeps Windows Mobile 6.1 our top candidate for mobile operating systems—for now.

At the end Openess is about balance: on one extreme you have a closed solution where a vendor has full control of the entire experience like the iPhone does and as Jerry Panagrossi from Symbian said, all revenue ends in Cupertino. This model at the other end of the spectrum is the Linux model which leads to fragmentation. Platforms need consistency to be successful: developers need a consistent target they can develop for once and end users need a consistent experience. The trick is how to allow for flexibility while providing a consistent platform. It is about balance. With Windows Mobile you have lots of choices, yet there is a consistent set of APis that allow developers to create a single application that works with every Windows Mobile phone. An example of such application is Windows Live Search - a single CAB file that works with any of the 12.5 million phones we sold in the last six months.

The guy from Google kept talking about how Android will be a really open OS. I know many quaestions about Android have not been answered yet including business models, what will be the core platform, or how will the people who support the platform will make money - none of the Open Handset Allince member companies is a non-for-profit. It's clear how hardware vendors and carriers will make money, I doubt software companies like Packet Video and Esmertec will donate their software to open source for the goodness of openess. Building a phone operating system is not a trivial task, integrating it with specific hardware is hard work for all companies involved. It will be very interesting to see how things evolve for Android and the partners.

What is unclear is what is the value of Android:

  • Cost is not as relevant, a carrier makes thousands of dollars in revenue per year on service a smart device, the cost of integrating and supporting a new platform, even if all the software is free can quickly eat into the savings of $10-20 per unit for a proven, mature mobile OS like Symbian or Windows Mobile. With a fully featured Blackjack 2 available at $99 it is hard to make an argument for cost.
  • Openess. Is relative and not always a good thing. Windows Mobile (and other mobile OSes) allow device manufacturers great flexibility and room for innovation in user interface, installed applciations, radio stacks, etc. Personally, I think the way Microsoft licenses Windows Mobile could be made a bit more strict to provide more consistency.
  • Speed and Stability. In the last two years Windows Mobile has matured as an OS along with the device manufacturer partners. I have been using Windows Mobile 6.1 (a pre release build) for months and have not experienced a single crash. Not once have I needed to remove the battery or do a hard reset. Many Windows moble phones are very fast and responsive. The Q9 Global I am using now is pretty good - it is powered by a 300Mhz processor. I played this week with a reference board from TI using a 600 Mhz processor and was unimmpressed with the speed of the platform (and 600Mhz processors are not cheap).

I have enjoyed doing these panels: the conversation is great and the opportunity to talk to industry coleagues is fantastic. The one thing that was evident to me after the panel was over is that individually we have a lot in common and we are trying to solve the same problems in the industry.

As all my posts in this blog, this reflects my personal opinions which may not be the same as that of my employer. There are no warranties, expressed or impied about the accuracy of the information (I could be wrong).

Comments