Symbian pathways

On the VisionMobile site today, guest blogger Michael Vakulenko offers a provocative analysis Will Legacy Smartphone Platforms Keep-up with iPhone and Android?

In this article, “legacy smartphone platforms” means “BlackBerry OS, Windows Mobile and Symbian/S60″.  Michael’s conclusion about these platforms is:

…legacy smartphone platforms do a decent job in their respective “comfort zones”.  Nonetheless, when taken out of their natural environment they fall far behind in comparison to iPhone and Android.  These modern platforms were designed for new market requirements without constraints of legacy code or backwards compatibility considerations.

Here’s his summary of Symbian/S60:

Nokia’s decision to open Symbian/S60 source has stalled development of the platform. It will be very difficult for Nokia and its partners to make major improvements to the platform in parallel to moving the platform to an open source model.

The analysis is interesting.  For example, I agree with Michael that speed of implementing new market requirements will be a core differentiator between mobile platforms.  However, you won’t be surprised to hear that I reach a different conclusion.  I have two points to make.

First, the move towards an open source model has not “stalled development” of the Symbian platform.

Plans for open sourcing the platform were announced in June last year.  The release of Symbian OS that was made to device manufacturers and development partners, by Symbian Ltd, towards the end of that same year – Symbian OS v9.5 – broke all internal records in terms of low defect count, keeping the agreed release schedule, and delivering the agreed release scope.  That’s not what you’d expect if the development was “stalled”.

The 9.5 numbering belongs to the world when Symbian OS and S60 were created separately – each with their own version number.  With the Symbian Foundation, these two entities become architecturally unified, into what’s now called “the Symbian platform”.  This new entity has its own new numbering system.  The first version of the platform built from the new code structure – from code kept in Symbian Foundation source repositories – is denoted Symbian^2.

PDK (Platform Development Kit) builds for Symbian^2 have been available to Symbian Foundation members for several months.  PDK builds for Symbian^3 will be available from September.

You can find more details of the contents of Symbian^2 and Symbian^3 releases from the Symbian Developers website.  For example, highlight features of Symbian^2 platform include

• UI Makeover Step 1: Web Runtime offers web design on mobile;
• Personal: All-new home screen supporting user customization, embedded widgets and internet content;
• Adaptable: Support for multiple form factors and input methods (touch, non-touch, flexible aspect ratios and resolutions);
• Dynamic: Location based event framework that allows apps to take action in response to the user’s changing location.

And highlight features of Symbian^3 platform include

• UI makeover step 2: Graphics support for advanced layering and effects (eg semi-transparent content layered over video, complex animated transitions between apps);
• Sounding clear: A new high performance networking architecture enabling broadband speeds, ideal for streaming high definition video and high quality VoIP calls;
• Movie time: Support for files >2GB in size, enough for full length HD movies; Full HDMI support with HDCP;
• Simplifying internet access: Adaptive WLAN background scanning, energy efficient.

This is only the tip of the iceberg.  If you follow the links on the Symbian platform roadmap page, you’ll find there are huge amounts of package-level development taking place.  So, again, I dispute the claim that Symbian platform development is “stalled”.

(For lots more news about items in the Symbian platform roadmap, I recommend attending the Symbian Exchange and Exposition, at the end of October.)

Second – and more significant in the longer term – the open source model provides the means for Symbian platform development to accelerate (as opposed to stall).

This is something that’s becoming increasingly clear to me.  Even though the platform remains, for the time being, only partly open source, some of the benefits of the overall changes in process are already becoming apparent.  These benefits follow from the fact that the ability to build versions of the platform is becoming much more widely distributed than before.  This in turn is due to:

• The creation and distribution of the PDK (Product Development Kit);
• The pending availability of hardware reference platforms.

As development teams throughout the Symbian community reflect on this opportunity, more and more people are warming to the option of a new way to demonstrate the value of experimental new software they have been creating.

Previously, any such software needed to pass through a number of “review tollgates” before having the chance to be integrated into a shipping product.  Product managers are, naturally, often risk-averse when it comes to making time and space on a busy product development schedule for untested new software components.  That can be bad news for developers who have a hunch that their new software could trigger a novel breakthrough, but who don’t get a chance to prove their case.

However, the new Symbian Foundation processes open the possibility for teams (such as research groups inside device manufacturers or network operators) to more easily install their software on specially adapted reference environments.  They can do this without having to queue up for permission from hierarchies of product managers, middle managers, programme managers, and the like.  Instead, they have a new route to demonstrate the suitability (or otherwise!) of their experimental software.  They’ll be able to test this software, get feedback from a sample of real end-users, modify it, test it again, modify it again… and then make the case to product managers to include their work in commercial devices.

In other words: the new Symbian Foundation processes allow more development to take place in parallel – which means that the overall flow of innovation in the platform will be accelerated.

This is in line with something I wrote a few months back – stating the reasons for Symbian moving to open source:

Involving more people will allow more opportunities to be addressed and challenges overcome:

• The number of smart and capable people outside an organisation will always far exceed the number on the inside
• It’s not just about the sheer number of people, but the number of different perspectives.  People in the same organisation are often constrained by common thinking patterns.  By encouraging and enabling contributions from many organisations, we’ll benefit from the fact that there will be many different kinds of approaches
• Freely available standard programming interfaces are an important first step to involving more people – but they only go so far
• Freely available source code, that is open to modification and experiment, enables deeper and more substantial collaboration: collaboration in the evolution and refinement of the underlying software platform, rather than only on the edges of the platform.

I’ve had some ups and downs with the articles on the main Symbian blog site, blog.symbian.org.

Asking readers for suggestions can easily be misconstrued.

Here’s an example.  Back in April, an idea emerged inside Symbian that perhaps we should change aspects of the annual ecosystem get-together, previously known as the Symbian Smartphone Show.  Various aspects of this idea were discussed internally.  I decided to surface aspects of this discussion in a blog posting, Redesigning the Smartphone Show.  I wrote,

The date and venue for the 2009 event has already been set: it will take place on 27-28 October, at Earls Court 2, in central London (the same venue as the 2008 event).

But there’s a lot that is not yet set for the 2009 event:

• The name
• The principal themes and content
• The event branding…

If you have views on what you’d like to see at the 2009 event, please speak up!

For example:

• Which aspects of previous events would you like to see carried forwards, and which changed?
• Which speakers would you like to see at the event?
• What features of the show would make it more likely (or less likely) that you would take part?
• And – last but not least – What should the event be called?

These questions resulted in lots of useful feedback – in blog comments, and via email.  But the questions also resulted in some ridicule.  Bill Ray from The Register (“Biting the hand that feeds IT”) wrote a good-humoured but biting piece called “Symbian show struggles for identity – Just needs name, theme and branding“.  To quote from the article:

This year’s Symbian show has a date; October 27th, and a venue; Earls Court, but is looking for suggestions from the crowd when it comes to a name, theme, branding and content.

The news comes in a blog posting from David Wood, who … explains that the new Foundation has no idea what to call the show-formerly-known-as-the-Symbian-Smartphone-show, or what kind of content would most attract the open-source community that’s going to be vital to Symbian’s future.

Openness in conversations means avoiding reaching and locking in conclusions prematurely.  Even if you’ve already got a good answer in mind, the open approach will push for better answers.  The fourth and fifth answers may well be much better than the first and second answers.

That was the thinking which led us to widen the discussion about the naming and the contents of our annual show, to accept input from the community as a whole.  As we reflected on the feedback and comments we received, we realised that the theme of “Symbian Exchange” was particularly strong .  This eventually turned into SEE – Symbian Exchange and Exposition – with the tagline of “come to SEE the future of mobile”.  It’s a name with a great potential.  And it can be traced to comments to the original blog posting (here and here).

Ridicule is just one of the hazards of an open conversation.  Another risk is that you hurt people, who think that a matter should already be decided, and there’s no need to keep the conversation going.  You can also hurt people by failing to censor comment responses that have all the appearance of being destructive, spiteful, trolls.

Yet another risk is that, in adopting a brainstorming approach, you will say something tentatively and/or provocatively that gets (wrongly) interpreted by readers as a new statement of corporate policy.  For example, a statement such as “if xx and yy don’t happen, then it might be best that zz happens” can be taken out of context and mis-interpreted as “Symbian official spokesperson wants zz to happen”.  Woops.

Given these hazards of open conversation, here are two possible responses:

• Take great care in everything that’s written – to avoid all chance of misunderstanding, but at the cost of slowing down the conversation;
• Seek to explain to the wider audience that a new mode of conversation is happening.

There’s a role for both response.

One reason for creating this new blog – Symbian pathways – is to separate off some writing that is more personal, tentative, opinionated, and provocative; and to generally keep the main Symbian corporate blog for shorter and less contentious items of news.

I’ll continue to try to be clear in my writing, but no doubt I’ll fail from time to time.  That’s a price that’s worth paying, in my view, if it encourages a quicker and fuller exploration of innovative possible pathways to a better mobile future.