According to the latest news, Nokia is likely to wind up encouraging Symbian application development. There are a few things which can be learnt from this move for most of the software companies and businesses.
In a nutshell, Symbian was bought over by Nokia and, which intended to make it an industry wide standard in developing mobile applications – made it a foundation run by member subscriptions. To aid this, Nokia gathered a number of mobile phone vendors which were starving for a mobile operating system. Times passed by, and the Symbian source code was made open source and freely available since last year. Since then, Nokia has experienced a reduced amount of market share and has gone down.
Following is a list of lessons which can be learnt from Nokia and Symbian experience.
Design by Committee
Making Symbian design governed by a committee instead of a central designer, largely contributed to the failure of the Symbian to adopt to the environment. Committee members had their own proprietary implementations of the platform architecture, and then the platform started to starve from lack of usable components.
This was highlighted back in April 2010 on this blog, in the post titled Is Open Source software a classic example of fail for Design by Committee?
Nokia probably had great management people at its top, but lacked any with a technological vision. This is also clearly visible in many software companies engaged in business and enterprise application development. A clear technology visionary would allow any organization to reach its maximum potential with its existing business and technical resources.
In addition to the technological vision, a marketing visionary is also needed within a company to promote its technological efforts to the mass market.
The traditional waterfall software development method along with phased releases used at Symbian and Nokia [not applicable to Series 40 platform] were too destructive for the foundation as a whole, specially at the times where other mobile operating systems were developed and delivered in faster release cycles and the customers were eagerly waiting for bug fixes and new feature additions which can unlock the best features of the device hardware.
Platform, Platform and Platform
As a platform, Symbian was too complicated for the developers to develop applications for. The platform API(Application Programming Interface) has been too complicated for regular and rapid development. This has not been the case for mobile operating systems that were coming from developer centric organizations.
It should be also noted, that the platform has to be maintained by a core-team, with contributions accepted and merged from different products and projects which use the same platform. This has been the case for Qt framework, which Nokia acquired along with Trolltech. Now, Qt has become the only platform that Nokia will ever promote.
Developers, Developers and Developers
Steve Ballmer’s famous lines, has been reiterated at the recent Nokia World summit by the new Nokia Chief Executive Officer, Stephen Elop who previously headed the Microsoft business division. Developer centric organizations rarely go down in business, in spite there would be many management people arguing against the fact, simply because they see developers as inferior to the managers or they have a bad perception of overly expensive developers. Developers are the ones who would be defining the boundaries which a modern organization can expand.
Don’t let the hardware limit the developers
At Nokia, hardware rules, this is no different from many traditional software companies which focus on reducing costs by limiting the hardware and capabilities available to the developers (whether at the runtime environment or the development environment). Give freedom to the developers and a solid foundation which they can simply focus on software and logic, they will do magic. Otherwise, it’s very likely that developers would be compelled to waste time optimizing [or debating the best way to concatenate Strings, which is a classic example] instead of delivering new, interesting and usable features to the end-user.
Product based Matrix Organizations don’t work
Considering Nokia and it’s recent products, one can always see that there is a large amount of different devices which run the same platform, but have different features. It is largely possible that these devices are coming from different teams which are headed by different product or project managers. Very likely that there are developers who excel in different areas of development put into different projects, causing increased fragmentation of the operating system and the device software, making it difficult to maintain. It also reduces the chances of platform improvement by making it difficult to merge individual improvements to the main source code base.
Don’t keep incompetent people
There were loads of people who were working at Nokia from top to bottom (including the CEO and the factory worker) who were simply incompetent for their current role, highlighting a real world occurrence of the Peter Principle. The maximum from them could have been obtained from their competencies if there had been a proper mechanism that would have prevented such. Unfortunately, when the company as a whole goes down, these people are less likely to find a better job and will have to step down and go down another step when joining another company.
Research is important
At all times when these things happened, Nokia Research has been making great advancements in the fields which they conducted research. In the coming times, probably the only Saviour for Nokia would be the product features which can be improved through the in-house research. Microsoft, Google and Facebook are also examples where in-house research has benefited an organization to come out and perform soon after difficult times.
Please feel free to share your views and additions.