The last supper

22. The last supper

For all kinds of reasons, Nokia’s “Seaside” project made good sense. This was the project, communicated to Symbian executives in late May 2008 and announced to the world on 24th June 2008, that would execute a number of overlapping changes in the Symbian world:

  • To merge together the software engineering teams of Symbian Ltd and Nokia S60, into a single entity, with much improved collaboration – avoiding conflicts and duplication of efforts between these two teams
  • To unify the different Symbian UI systems, based around the S60 codeline – avoiding conflicts and duplication of efforts between UIQ, MOAP (used in Japan), and S60
  • To place the source code for the resulting “Symbian platform”, consisting of Symbian OS plus S60, into open source – with the potential benefits of greater community contributions to the platform
  • In this way, to ratchet higher the performance of the Symbian world, with reduced barriers to innovation – as needed in the light of formidable competition from new smartphone operating systems.

Seaside was the outcome of many years of strategic thinking inside Nokia. As covered in the chapter “The near-miss merger”, Nokia had already attempted something similar, twice, in 2004:

  • First, in conjunction with Psion’s exit as a shareholder of Symbian
  • Later in that year, following a very positive set of discussions between Symbian and S60 executives at a retreat at Ruka, northern Finland.

One key difference in 2008 was the idea of placing the platform source code into open source. That step was necessary for the other Symbian shareholders to agree to the Seaside initiative. Otherwise, they feared losing all control over the evolution of the platform. They also feared that Nokia might keep the best new modifications of the Symbian platform to themselves.

One prior example served as an important model in the negotiations between shareholders in the run-up to the Seaside announcement. This was the example of IT giant IBM, and a developer tools platform called Eclipse. The way IBM successfully open sourced Eclipse would serve, it was hoped, as a precedent that Nokia could emulate with the Symbian platform. And, despite the way things turned out for Symbian, it might even serve as a precedent for other technology platforms in the future.

The Eclipse example

Consider a technology asset primarily owned by one company – such as the platform underpinning the set of “VisualAge” tools for developers, owned by IBM, or the S60 user interface running on Symbian OS, owned (or on the point of becoming fully owned) by Nokia. This asset is important to the company, but isn’t the company’s main product. Rather, it is an enabler for other products where the company makes its primary revenues.

Imagine that:

  • The company wishes to accelerate the development of that technology asset
  • There is a long pipeline of potential improvements to the technology asset, all of which would need people to work on them to create them
  • There are many people outside the company who could usefully contribute to that development
  • There are many people outside the company who could benefit from improvements to the technology asset – they could use that asset to enable products that are important to them.

If these conditions apply, the company has the option to allow other people, outside it, both to contribute to the technology asset and to benefit from it. But first, a number of decisions are needed:

  • Will the benefits of a larger development community exceed the cost of any drawbacks from resulting additional competition to the company’s main products – competition that is enabled by sharing the underlying asset more widely?
  • How much effort will be needed, to transform the asset into a form in which outsiders can usefully contribute to it?
  • How much effort will be needed, in the longer term, to oversee the joint development of the asset, in a way that keeps the overall community sufficiently happy with the outcome?

In the case of IBM, their engineers realised they would need to re-architect parts of the platforms underpinning their VisualAge products, before third parties could usefully contribute to this technology. The result became known as the Eclipse platform.

[ SNIP ]

Recent Posts