As many of you know, I’ve been involved with Tizen for quite some time. I manage Tizen at the Linux Foundation, which means I get a unique perspective on the project. And as anybody who has talked with me about what I do, I am a huge supporter of the Tizen platform and its goals, and am a general optimist (including but not limited to Tizen, of course).
2014 was a big year, with Tizen wearables and cameras hitting the market, Tizen IVI 3.0 achieving GENIVI 7.0 compliance, and a lot of interesting platform work on Tizen:Common. 2015 has really gotten off to a great start as well, with the announcements at CES that all new Samsung Smart TVs released in 2015 will run Tizen, starting in February.
Last week was a special milestone, though. As you may have seen, the Samsung Z1 was released in India. I believe I speak on behalf of many when I say this was a day we’ve been waiting for, and it’s great to see it finally arrive.
As I said, I’m a huge supporter of this platform, not just because I have spent the past three years working on it, but because of what it represents. Let me put this into context.
Linux in general has been very successful in the embedded industry, partly due to its flexibility but also to the general economics of using open source. A lot of this success also is thanks to being in the right place at the right time. Looking back over the last decade and a half, we as a society are quite fortunate that Linux hit a level of maturity right at the time that Unix was getting long in the tooth, when bandwidth was cheap and accessible, and when the capabilities of commodity hardware were really taking off. A large community of really smart people suddenly discovered they had an itch to scratch, and came together to create something that reflected their own current needs, not just those of systems designed years ago. Linux was truly a breath of fresh air for the entire IT and consumer electronics industry, and we’ve all been reaping the benefits of that. Even if you’re using something that doesn’t use open source (and where on earth did you dig up that old fossil?), chances are good they’ve had to scramble to keep relevant as their competitors increasingly use open source components. The economics of open source are unforgiving to those who don’t embrace them. These communities tend to innovate quickly, so you really can’t afford not to use it.
In many ways this mirrors the spirit of Tizen. In the embedded industry there’s a very clear itch – everybody needs to release better products faster, and nobody wants to pay to reinvent the wheel each time. The economics of rolling your own code on a project-by-project basis (even when it’s mostly open source) just are’t enough to guarantee success anymore.
I say Tizen is in the right place at the right time because it was designed to address this specific need, and makes use of components that scale well, from phones to cars to wearables to cameras to TVs and onward.
From the beginning, the goal with Tizen has been to create a flexible, extensible body of code that could be used (and reused) for a huge range of devices, reducing the amount of device-specific adaptation as much as possible. Obviously at the kernel level Linux has been doing this for years, but the goal is to take this model several steps further. One of the main objectives is to identify and integrate a common set of software components above and beyond the kernel (called “Tizen:Common” from Tizen 3.0 onward), and then add specialized layers on top that are specific to the product.
While this may sound like a no-brainer strategy, in the world of embedded device engineering it’s harder to implement than it sounds. The power and performance budgets of embedded devices are usually very tight, capabilities vary greatly, user expectations are very high, and switching costs are low.
And yet power management and performance are where Tizen has done well, to date. This is due in part to the ongoing focus on power management and performance in the Linux kernel, but it also helps greatly that Tizen is a fresh look at what a mobile operating system should be for today’s devices. As a result it is significantly less encumbered by designs that made good sense years ago, but we probably wouldn’t do over again.
Like many large, modern open source projects Tizen is a composite, constructed from other open source software components. We can pick and choose from the best of the best open source components, and design a system that reflects all of this open source innovation from the past few years. And so far, it’s paying off. Across the board, Tizen devices so far have done extremely well against their peers on things users actually care about, like performance and battery life.
The takeaway is that when building an embedded device for the mass market, there’s real value in using the right tool for the job. In this case, the right tool should reflect the realities and constraints of modern embedded system design. It’s far better than continuing to try and pound a square peg into a round hole.
So in light of this, I’m excited. Between Tizen TV, the Samsung Z1, all of the work going into Tizen:Common, a forthcoming release of Tizen IVI, and who knows what else, this is going to be a good year.