By a process of elimination, the platforms that need to be assessed in greater details, in no particular order are.
- Windows 7 Desktop
- Mac (desktop)
- Linux (Debian) distribution
Each of these platforms will be compared against the evaluation criteria described above.
Note these high, medium, low and none figures below are given as relative measures for the current size of the market, and are qualitative indications only. Our conclusion is that Windows 7 is the widest penetrated platform in the PC space, while Android is the widest penetrated platform in the mobile space. Other than that, all other platforms are partially available across the 5 screens that we consider in the table below.
Quantitative estimates of market figures have been be gleaned from the following source materials
- Windows PC: 350M licenses sold in the last 18 months (http://www.microsoft-watch.com/content/windows_7/windows_7_sales_scrub_vista_curse.html)
- Windows tablet: Windows has less than 5% of tablet sales in Q2 2011 (http://www.androidtablets.net/forum/android-tablet-news/19352-android-tablets-chew-30-out-ipads-marketshare.html)
- Windows Mobile: estimated 5M handsets sold up to end Q2 2011 (http://wmpoweruser.com/canalys-only-1-5-million-windows-phones-shipped-in-q2-2011/)
- Windows automotive: based on industry discussions
- Windows TV: based on industry discussions
- Apple PC: 3.76 Million Macs sold in Q2 2011 (http://www.yourdailymac.net/2011/04/apple-q2-sales-18-65-million-iphone-4-69-million-ipads-3-76-million-macs/)
- Apple tablet: the iPad has consistently over 50% market share and predicted to continue so for 2011 (http://www.ipadforums.net/apple-ipad-news/40786-ipad-marketshare-continues-impress.html)
- Apple mobile: Apple has around 25% share of the Smartphone market (http://www.linuxfordevices.com/c/a/News/ABI-Research-Q2-2011-smartphone-study-plus-HTC-Sensation-teardown/)
- Apple: automotive: based on industry discussions
- Apple: TV: based on industry discussions
- Android PC: initial entry into PC space did not meet with success
- Android tablet: Android has 30% market share of tablets in Q2 2011 (http://www.androidtablets.net/forum/android-tablet-news/19352-android-tablets-chew-30-out-ipads-marketshare.html)
- Android mobile: Android has over 45% of Smartphone market share (http://www.linuxfordevices.com/c/a/News/ABI-Research-Q2-2011-smartphone-study-plus-HTC-Sensation-teardown)
- Android automotive: based on industry discussions
- Android TV: based on industry discussions
- Linux PC: Linux has very low penetration among connect PC users (2%) (http://en.wikipedia.org/wiki/Usage_share_of_operating_systems)
- Linux tablet, mobile, automotive, PC: based on industry discussions
- MeeGo: only 1 mobile device announced, scheduled for Q4 2011. MeeGo can be ported on the N900 which is where the “low” figure on mobile comes from
Strategic view on platform future
|Windows 7||Windows has been the dominant operation system for 20+ years. Even on the most pessimistic trajectory, Windows will be a significant player for many years. Investment in a Win7 webinos stack is well justified|
|Mac||Mac has been going from strength to strength. Not only in PC and laptop penetration, but its derivative mobile and TV instance needs to be monitored carefully|
|Android||Android on phones has been going through phenomenal growth over the past year. Tablets TV, PCs and In-car variants are strongly rumoured|
|Linux||Linux has still not broken through into the consumer market, being largely the preserve of server and techno-uber geeks. However, the consumer centric offerings are improving considerably, and the no PC ecosystems are crying out for a strong competitor to Android and iOS|
|MeeGo||MeeGo can be characterised as a Mobile pruned Linux distribution. The recent Nokia announcements have been unfortunate and have weakened its position considerably. And it has yet to prove itself as a consumer platform with traction. There are strong (but non-provable) rumours that there will be some significant industry movement on these platforms in the next few months|
Our conclusion is that, besides MeeGo whose future is uncertain, all other platforms have an expected longevity that makes them suitable for use in the webinos implementation.
Cross vertical porting
All of these platforms give a strong cross device coverage, which can be summarised in the table below:
Reviewing the platform coverage of each in detail:
Windows 7: Windows 7 is strongly positioned for PC, Tablet and TV (using is Windows Media Centre product). The current mobile proposition, Windows Mobile 7 has been discounted as a priority in the analysis above. However, despite the significant technical differences between desktop and mobile, we can be encouraged by the fact that it comes from the same family of technologies and therefore porting should be possible either at a later date, or by a new consortium partner. A similar argument can be extended to the Windows Embedded. The Windows automotive embedded platform has been de-prioritised as above, but if we develop core desktop assets, motivated parties will be able to reuse and re purpose these assets for this platform at a later date.
Mac: Mac as webinos platform, is deemed very important, primarily because it is the development environment of choice for web developers. As a PC platform, it is therefore essential. As described above the mobile iOS platforms and AppleTV are not priority port considerations, due to commercial limitations on distributions. But the fact that these technologies come from the same technology base (as above) means that if PC Mac is in scope, the effort and cost of porting the Apple flavours or mobile, tablet and TV technologies should be significantly reduced.
Android: is deemed mandatory for its mobile and tablet impact. Although not significantly impacting the market yet, we can presume that TV and in car implementations will make rapid inroads into the market.
Linux: Linux distributions come in many flavours, tweaked to different device types and indeed vertical applications. Developed binaries are not always portable between different Linux distributions, however, proficient Linux developers should be able to port between them relatively easily. Linux, whilst not particularly impressive in terms of absolute numbers in any one of the categories, has a strong technical presence in each of the device types in which we are interested.
MeeGo: MeeGo claims to have a discrete (but highly portable) distribution for each of the device types that we have in scope. Current sentiment is that not all of these platform distributions are product ready
Freedom to implement
Windows 7: unlike Windows mobile 7, even though it is closed source, it is technically highly extensible, with very few ecosystem controls built into the OS. Mac: similar to Windows, in sharp contrast to the mobile variant (iOS) the Mac is open for the technical extensions, and there are no overt Apple control points. Android: of all the above, Android is the most tightly controlled. The marketplace constraints and the exposed APIs can be limiting. Linux: Totally open platform with no ecosystem controls. Technically fully extensible with open source code accessible. MeeGo: based largely on Linux code base, therefore flexible. The MeeGo security scheme implies that ecosystem controls could be introduced, but this is assessed on a device by device basis and does not reflect on the platform itself.
Our conclusion is that most platforms are open with the exception of Android where there are tight marketplace constraints and technical API limitations in place.
Probability (and need) of core platform absorption
Windows, Mac and Android, being all controlled by singular commercial actors in the market, are very difficult to influence. For a third party to get core code changes accepted into the stack is almost impossible
MeeGo and Linux by contrast as genuine open source initiatives, are in principle open to 3rd party contributions. Indeed their long term survival, to some extent, depends on it.
The good news is, as shall be discussed in greater depth in the technical section, the current proposed architecture, does not assume that significant core platform changes are required on the host operating system to port webinos
Consortium member sentiment
In order to provide a more pragmatic view but also in order to capture the real market sentiment webinos has performed an independent survey across the industry (Operators and Manufacturers) partners. Detailed results of this survey can be found in Section 4: Industry partners viewpoint. Here follows an overview of the captured answers to provide a better context within the discussion and justification of the chosen shortlisted platforms.
|Samsung||(minimum) Windows & Linux||Android||Linux based||–|
|Telecom Italia||–||Android or native Linux||–||–|
|Telefónica||–||Android or native Linux||–||–|
Technical effort, capabilities and dependencies:
In order to assess the technical effort to port the webinos framework to the host operating system, we need to first consider the technical architecture.
Based upon the architectural work from WP-3 a set of proposals for implementation technology has been superimposed onto the implementation architecture
This information is to be found in the appendix of this document.