The open road

At the recent Symbian Exchange and Expo (SEE) held in London, telecoms.com talked to John Forsyth, leadership team, Symbian Foundation, about the organisation’s new direction and the threat from Linux and Android.

Symbian is already the dominant mobile operating system, where do you see future growth potential?

We had a visit in September from a big delegation of top brass from China Mobile who came to visit Symbian and signed an MOU on collaboration. We have an enormous amount of alignment and obviously TD-SCDMA is high on their agenda – the recently launched 6788 is Nokia’s first Symbian TD-SCDMA handset.

The potential numbers in China are absolutely enormous and this gives China Mobile a degree of influence over what happens in their market, and they would like an open independent platform that is royalty free, open source and not controlled by anyone. And that’s what we are in the process of trying to create.

Symbian’s shift from a commercial entity to an open source organisation is about as radical as can be. How do you adjust to the different mindset?

I used to be VP for strategy at Symbian software when it was a software licensing company. Back then I had a dual outlook on things: Figure out how to maximise the effectiveness of a royalty harvesting machine; but at same time it was clear to me that it wouldn’t be possible to charge a licensing fee for an OS in the future, and that includes the desktop. Value is moving up the stack. People are spending more money on computing but they’re spending it on different things, different experiences, that are delivered at the service level as internet applications.

The change for Symbian was a radical step but look at how we approached it. We could have spun it out, had a dual licensing model, a halfway house, but we didn’t. And as for the association with Nokia and how good and how bad is that for business, well in the mobile world you dream of having Nokia as customer, but if the perception propagates that you are unduly influenced by one customer that can be uncomfortable.

One important thing we have done is make the governance, the whole way we are influenced, totally transparent and equitable. It’s the opposite end of the spectrum from the old Symbian governance model. No matter how big you are you get one vote, so Nokia has the same degree of influence as everyone else.

So has everyone in Symbian accepted the new strategy?

People are individuals with interests and passions and if they’re lucky they end up working or a company that allows them to pursue that. These [Symbian employees] are just software guys, we’re geeks, and most geeks are passionate about open source regardless of the company hat they have to wear during the day. The whole exercise of turning Symbian into an open source platform was the best thing that could happen, because it’s what people are passionate about. And that’s not unusual if you look across the entire software business. Most software guys would rather give their stuff away.

There are a lot of different ways of doing open source and I think the way were doing it hasn’t been done before. It’s hard to say things don’t get done or happen slowly, just look at Apache. They have 50- 100 projects ongoing at any one time and they are still running 90 per cent of the world’s web servers, and nobody’s ever paid a dime for Apache software. Whereas look at Microsoft, which can’t even get five per cent of the server market despite years of developing the IIS webserver. So you have to ask yourself why isn’t Microsoft being adopted? It’s not competitive in terms of features or robustness.

So does open source result in better products? A more agile platform? A more effective ecosystem?

Mostly open source is more productive and efficient than most large software development organisations that do it in a closed fashion. But at the same time there is a complexity to managing it in a way that makes it easy for commercial customers to adopt. And there aren’t the same commitments in open source, such as timelines for products. And that is the problem we’re trying to solve by learning as much as possible from other open source foundations. But what they have not all excelled at is the provision of a clear roadmap, that gives you really clear levels of quality that are going to be reached by certain points in time and how.

So if open source in mobile didn’t have problems we would be sitting round making loads of money from Linux phones. But Motorola practically destroyed itself trying to make Linux phones because it ended up with massive branches of operating systems that tied their resources up.

Isn’t this what Linux players like the LiMo Foundation and Android are trying to sort out?

The thing with the Limo Foundation is that it’s a document factory, they don’t have anyone that produces kits. We have a whole bunch of guys that look at all the code coming in from all different places, integrating it into kits that are validated against reference hardware, so they definitely build and run and there’s one place on the web where you can download something and guarantee it will run on your hardware. Limo hopes one or other of its members will do that stuff. But that puts you back in the world of commercial software because you need guys like Access to integrate different parts of the build and make it work, but they will charge money for that. Then you end up with a commercial software federation.

Android is disruptive but it will be damaging to everyone, including Google if it’s as fragmentary as it seems to be. I don’t think android is in Google’s best interests – Nokia puts out almost half a billion handsets a year and each one of those devices has a home screen, and the potential for a search bar. What are you doing building a mobile platform that Nokia is never going to use? They should be trying to get their search onto Nokia’s handsets.

My first reaction when I saw the Android architecture was  – although there are some interesting things in there, it’s a late 90’s architecture in the sense that they didn’t go for a web programming model but instead went for Java on top of Linux. Personally I think Java is neither here nor there, it’s nicer than programming in native C or C++ language but it’s nowhere near as nice as programming in JavaScript or Flash.

But Symbian has taken some flak as an application development platform…

Application development has not been great for Symbian, there’s no sense in hiding this, and that’s due to a mixture of tools that have been good but not best in class and an environment that’s brilliant for making phones, but too complex for making applications. If you’re a handset manufacturer it gives you a lot of depth and power, but if you’re looking to build a mapping application, then you’re asking a developer to do too much work. So in the roadmap we have a couple of big thrusts: Getting Qt into the platform – there are already 300,000 developers that love Qt, and it’s a fun productive environment; and there’s also the web runtime, so you will be able to program just in JavaScript, HTML, CSS, and write an application that uses all the underlying native services.