There is a tendency to be a fan of brand new versions of Linux, especially if they have a big brand behind them, and which have only just had the wrapping taken off their Software Development Toolkits. We put it down to a sort of boredom with all things that are established and a love of revolution among the programming, as well as marketing, cognoscenti.
It is this type of emotional ‘rooting’ for one version or another that brand is based on. We all love Google Chrome, not because it is proven, not because we have seen it, but because it comes from Google and it will surely be as open and user friendly as ‘ well as the search engine.
Of course we have no such belief in reality, and while Android may well create the kind of stir as a handset operating system, which the Apple iPhone made as a device, Google has told us that it is only meant for smaller devices without huge amounts of storage and that Chrome, the operating system version which has yet to be delivered, is the real OS for larger devices.
Is there any optimism that these two operating systems will have a common development environment? That we can treat them almost as if they were the same target platform? The great appeal, according to Google is that there will be services out there in the Cloud which can be harnessed by either, because the ‘It’s about the browser stupid.’
It was a few months after the emergence of Android, just as soon as its first SDK version had been delivered, that we heard that someone was using it for a set top, but we haven’t heard that since, and all talk of the set top market shifting ‘en masse’ to Google has since evaporated. However all that talk began again after the announcement at Mobile World Congress that Nokia’s Maemo and Intel’s Moblin were to merge to form MeeGo, with set top vendors immediately considering this as a potential move.
So instead of an untried OS and an untried SDK from a company that is likely to have a secret agenda and who wants to bully set top vendors, there is the option of a properly open source environment, okay we accept that it is both new and not available until next year, but it has one clear advantage, the QT (pronounced CUTE) development framework.
What Google is saying is probably right, that different device types need a different OS. But that’s been the QT approach from way back in the early 1990s, it’s a development environment which can target multiple Linux-like OS.
When Nokia first merged its Linux options into one with Maemo, it almost forgot that it had QT tucked away and the announcement did not guarantee the link between QT and Maemo. Complaints were made and that was rapidly and aggressively addressed.
In essence QT is one of those, write once, deploy many, pieces of software which has far more technology under the bonnet that most people realize. It is cross-platform in nature from the ground up, it includes a separate user interface layer so that the UI can be preserved from device to device or amended or adapted so that it makes sense for the size of device. And it is, as Chrome will be, all about applications being web-enabled. It specifically targets desktop, mobile and embedded operating systems without any rewriting of the source code and is built using the powerful object oriented C++ programming language.
By now there are extensive C++ class libraries which mean that a proficient user can pretty much find code that has done any specific job in the past, and done it well, and re-use it. Another aspect is that the embedded version is by now mature and takes up a tiny footprint on embedded devices.
On a set top there are a lot of fairly intricate jobs going on, parsing a variety of data streams, recognizing different types of video formats and codecs, decoding conditional access encryption ‘ just changing a channel on say an IPTV set top, is a major exercise and virtually no set top does it exactly the same from implementation to implementation. A fast integrated development environment is an absolute necessity when you are adapting a set top to a new environment.
But imagine the glee when set top makers are now being asked to also help in the creation of three screen video services, which have to be played out on a wide spread of devices, from an intelligent remote control, to a PC or a handset, as well as a set top itself, and all the TVs they are attached to. The UI consistency and portability can suddenly become a real feather in the cap of the set top vendor and the operator, wanting three screen video, can become a convert to the technology overnight.
Okay we grant there’s more to it than that. You may have to go back to an earlier version of the video, scale it down on a server, change its dimensions to or from wide screen and then re-issue it from another place, but the consistency of the code, the homogeneity of the support experience all offer the brand something when they transfer to a three screen strategy.
At present the praise for MeeGo may only be the preference of a quirky set top developer, heard in the corridor of a set top company, but it could easily amount to a trend, and that trend could provide a natural bridge between the ‘cheap as chips’ Intel Atoms and the ubiquity of ARM based processors in portable environments, and that advantage could be its making.
The alternatives are to go with new Google or existing Microsoft offerings, or Apple, which exerts iron control over which applications are distributed onto its devices. Any of these decisions will trap more and more devices into those architectures, complete with the dominance that gives these companies and any secret, advertising agendas which they have. But the agendas of Nokia and Intel are for the most part neutral, at least to set top makers.
There’s nothing to stop another variant of Linux, such as LiMo, making a strong bid to dominate handsets, but it has no edge to spill out onto tablets and set tops and TVs, so it could be that MeeGo is going to become a de facto media OS in the not too distance future, simply by the assets it brings to set tops.