Smartphone OS Installed Base

communitiesdominatebrands.gifThe indomitable Tomi Ahonen shares some stats from his Phone Book 2012. Of particular interest is the installed base of smartphones, by operating system.

Note that this is the installed base, not recent sales, which means the numbers are more useful for people interested in trying to determine what platforms they should support. Further, they are world numbers which is important as some US-only research, not clearly disclosed as such, tends to be misleading.

  

2012SmartphoneInstalledBase.png

Nokia and Microsoft in the Last Chance Saloon

reuters.pngReuters have an interesting article "Nokia, Microsoft head for Last Chance Saloon". It describes the poor uptake of Windows Phone Devices (3.7 percent of the global smartphone market) and how the next incarnation, Windows Phone 8 "is very high stakes" for Microsoft and Nokia.

I was never a fan of Nokia’s strategy announced on Feb 11 2011. I am not anti-Nokia or anti-Microsoft. In fact, I developed for Windows CE/Mobile for a decade followed by Nokia/Symbian for half a decade and wanted to see both companies succeed. For the last few years I have only developed for Android and a small amount of Windows Mobile for a few very long term clients. With Windows 8 we enter a new era. As a developer, is this an area worth pursuing?

The existing Windows Phone 7 is based on Windows CE which runs on ARM-based devices. The existing Windows 7 (desktop) runs on Intel architectures. With Windows 8, Microsoft have allowed Windows (desktop) to run on ARM devices. This has two consequences for mobile…

  1. It’s possible to have less power-hungry (ARM rather than Intel) tablets that run Windows 8.
  2. Windows Phone 8 no longer needs to use Windows CE and now uses the same underlying architecture as the desktop.

But there’s a catch – in fact there are two catches. Windows 8 on ARM (called WindowsRT) is not the same as Windows 8 on Intel. Both are separate and you won’t be able to run applications compiled for one on the other. This means existing Windows apps won’t run the new ARM based tablets unless the respective developers recompile their code.

The second catch is that while existing Windows Phone 7 users have been promised app compatibility, there will be no existing device OS upgrades. This might cause some existing Windows Phone fans to become disillusioned.

There will be two ‘classes’ of Windows tablet. Intel/AMD ones will be expensive, as now. The new WindowsRT will be less expensive, priced like Android tablets, but there will be a lack of apps and initially few manufacturers. As a developer, you can see this as an opportunity (a new market) or a limitation (initially a small market).

Here’s my take on this. I can’t yet see how Windows Phone 8 will compete with iOS or Android. Similarly, I can’t see how the new ARM-based Windows 8 tablets will compete with the iPad or Android tablets – at least in the consumer space. However, the tablets might be more attractive for businesses that already use Microsoft technologies and there might be scope for B2B applications.

One thing I don’t understand is why .NET didn’t shield an app from the underlying platform (ARM vs Intel). This was one advantage of .NET, being able to run managed code that’s independent of the underlying architecture. I suspect it’s like Java ME, Nokia’s Qt and many other cross platform development tools. Difficulty, mis-management or neglect has prevented these tools from realising their aspirations.

Mobile Stats, Android and Fragmentation

gartner136.gifThe theme today seems to be mobile stats and different ways to view them. First take a look at Gartner’s Q1/2012 stats released today. The interesting part for mobile developers is the breakdown by operating system…

gartnerq12012.png 

Android has made a large gain in market share. Symbian is dying fast and Microsoft has less share than Bada that, according to rumour, will be retired soon. If the Gartner stats aren’t enough for you, take a look at the Google sponsored Our Mobile Planet, that allows you to create your own charts based on consumer stats. Finally, if your focus is Android then visit the opensignalmaps fascinating analysis of Android fragmentation and then read how Google is shifting tack to try to reduce fragmentation.

The Enterprise and Mobile Platform Security

dimensionalresearch.png

Dimensional Research has a recent free report on The Impact of Mobile Devices on Information Security (pdf) that highlights some of the issues related to enterprise device security. The report is based on a survey of 768 respondents responsible for securing company access in the United States, Canada, United Kingdom, Germany and Japan.

enterpriseincreasedpersonaldevices.png

The report shows the dramatic growth in the use of personal devices at work, how iOS is the most used mobile platform, Android is the most risky platform and that mobile devices are driving an increase in security risk. The main problems are seen to be lack of employee awareness, insecure web browing and use of insecure WiFi. Careless employees are seen to be a greater risk than hackers.

Of particular interest to mobile developers targeting the enterprise is a chart showing the most used mobile platforms connecting to corporate networks…

enterpriseplatforms.png 

The top platforms are iOS, BlackBerry and Android. This is a relatively large, quick turnaround from when, only several years ago, Windows Mobile and Symbian (and BlackBerry) used to be the top platforms in enterprise.

Graphics Download for Mobile Apps

imageformats.pngOne of the areas I commonly see startups underestimate is the download of non-photographic graphics. In its simplest form, download via http is trivial on all mobile platforms. However, the complexity manifests itself when you start to consider that different devices have different screen sizes and resizing graphics on the device is undesirable either because upsizing causes fuzzyness, downsizing loses small line detail in shapes and, in any case, downloading larger graphics for later reduction is extremely wasteful on both server resources and data.

Things get slightly more complex when you start to think about the server sending a particular size based on the device. The device type has to be sent to the server, the server has to keep different sizes of the same image and these images have to be created somehow, usually by a human if it’s a non-photographic graphic. What’s more, the server needs to know about all types of device and update these into the future. An additional problem on iOS is avoiding undesirably large downloads via 3G

All these things should influence your initial design from the outset. Keep the graphic design as simple as possible. Every dynamic graphic you add will increase the complexity. Where possible, consider using scalar vector graphics rather than bitmap-based formats. Possibly have different screen layouts for different screen resolutions so you can re-use a same-size graphic in different ways. Think about which devices on which platforms you wish to support as the more you have, the greater the graphical complexity. To spice up app, consider using fixed brand-coloured buttons, list backgrounds and other controls rather than insisting on dynamic bitmaps.

Gartner Q4 2011 Stats

gartner136.gifGartner have just released their worldwide smartphone stats for Q4 2012. Total smartphone sales in 2011 reached 472 million units and accounted for 31 percent of all mobile devices sales, up 58 percent from 2010. Of interest to developers is the table showing platform market share…

gartnerq42012.png 

These figures are interesting because they are world figures, not just US numbers as shown by some companies and analysts. The order of magnitude of the various numbers backs up those given by Canalys last week.

Canalys 2011 Full Year Stats

canalys.gifCanalys released their full year 2011 smartphone global sales stats last Friday. An amazing 488 million smartphone were shipped in 2011. While sales by individual companies are interesting, the table most useful for mobile developers is the split by platform…

canalysye2011.png 

Android now has more than double the market share of iOS. Symbian is dying fast. Windows Phone is dying even faster than Symbian. Personally, these stats reflect my decision last year to become an Android only developer having previously been multi-platform.

These stats and Nokia’s, so far, relatively unsuccessful take-up of Windows Phone also seems to support my belief last year that Nokia was making a bad decison that would be a fatal diversion. However, that’s another story that I hope to write about later this week, nearer the 1 year anniversary of that fateful day.