MNB Reader Stories: The Burning Symbian Blog

Here is a story sent to MNB by Janne who has decided to collect some of the thoughts and discussions over the past week or so. We occasionally get articles sent in and most of the time it just goes under our own post, but now I’m going to start putting them into their own writer column, “MNB ReaderGenerated”.  If you have articles you want to share with MNB’s readers, send them in to tips[at]mynokiablog[dot]com. We can also discuss for those who’d like to share Nokia news directly on the longer term, but that’s another topic. Let’s get onto Janne’s post

Thanks folks!

BR,

Jay

________________

Last weekend the mobile blogosphere noted the first anniversary of Nokia’s February 11th 2011 announcements and reminisced on the famous Burning Platform memo by Nokia CEO Stephen Elop. Understandably the entire spectrum of human emotion and technocrat opinion still linger over these events. One of the more prominent, or at least lengthy, analysis of the Burning Platform memo and subsequent events came from Tomi Ahonen in his Communities Dominate Brands blog, where also I myself commented on his analysis. Later I quoted and discussed my comments here atMyNokiaBlog, where “arts” inspired me to collect my thoughts into a separate article.

The big question I, like many, have asked themselves when analyzing the post-Feb 11th Nokia strategy is: Why? The previous Nokia Qt strategy was solid, potentially spanning a compatible software ecosystem from “the next billion” Meltemi, midrange market dominating Symbian and high-end MeeGo – price points from $100 to $1000. Why drop it all for the unproven Windows Phone? The obvious answer for many has been Stephen Elop and his Microsoft background. While for an objective observer it is impossible to completely dismiss any such backroom shenanigans, there are reasons to consider Elop’s trojan play unlikely. Not the least of which is the sheer number of old-guard Nokia executives and board directors approving the new approach. In fact, if the public version of the events is to be believed, it was Nokia’s Kai Öistämö who raised the flag to him saying he fears the MeeGo strategy will not be enough. The now-famous “Oh shit” moment.

In his Communities Dominate Brands blog Tomi Ahonen makes a commendable effort of taking us through the numbers involved in the pre- and post-Feb 11th Nokia. No doubt about it, since February 11th, Nokia’s position has crashed. The company went from smartphone dominance and profits to losing the leadership and recording losses, all the while shedding tens of thousands of employees bordering a national tragedy in Finland and local tragedies in many other places. Make no mistake: When Nokia jumped off that burning platform, hitting those North Atlantic waters clearly hurt it bad. Stephen Elop has admitted as much at least on two separate occasions. The effect on Symbian sales was more severe than anticipated. First signs came during the Q2/2011 results and the verdict during the Q4/2011 results:

“The challenges we are facing during our strategic transformation manifested in a greater than expected way in Q2 2011.”

“Specifically, changing market conditions are putting increased pressure on Symbian. In certain markets, there has been an acceleration of the anticipated trend towards lower-priced smartphones with specifications that are different from Symbian’s traditional strengths. As a result of the changing market conditions, combined with our increased focus on Lumia, we now believe that we will sell fewer Symbian devices than we previously anticipated.”

Arguably part of the damage stems from the way Symbian was effectively end-of-lifed on February 11th, and the rest from the competitive landscape. I have no quarrel with those who say the Symbian announcement reminds them of the Osborne effect and that it was ill-considered. It is a compelling argument to make and certainly one worthy of consideration. I am confident that corporate communications experts and organizations will continue discussing that for years to come. As they should. But what about that competitive landscape? Over the course of the February 11th weekend and at the Nokia general meeting in May 2011 Stephen Elop and later Nokia Chairman Jorma Ollila made a series of interesting comments on the state of Symbian.

Two of the comments around February 11th that stuck with me, were from Stephen Elop explaining how he had discussed the time needed to modernize Symbian vs. the time needed to adapt Windows Phone with his teams, and they had come to the conclusion – apparently even a head of the Symbian program – that going Windows Phone would be faster than fixing Symbian. Speed being the essence, this apparently was an important consideration for Nokia. In the second notable comment, and Elop reiterated this in a couple of presentations and interviews in different ways, he explained that they had found it hard to adapt Symbian to some future technologies. He did not elaborate what these technologies were, but it is not very hard to guess as the industry is on the verge of moving wholeheartedly to multi-core processors and fourth generation communications technologies.

I don’t have the exact quotes from the Nokia general meeting, but I was present in the audience and according to my notes Jorma Ollilla used a part of his speech to discuss how the previous Qt strategy change at Nokia had been well-timed in his opinion, but had failed due to two reasons: failures in management and the poor state of Symbian’s codebase. I remember thinking that it must be significant when Symbian’s code, something so technical, is actually brought up – by the Chairman to shareholders no less – in conjunction with leadership woes as one of the two major reasons of strategic failure at Nokia. Later, in the CEOs speech, Stephen Elop again reiterated this point and explained that Symbian had become, quote, “fragile” and that changes were costing too much and taking too long. Clearly, reading between the lines, Nokia management was telling a compelling story here.

What went wrong with Symbian? We know about its eccentric British roots in form of EPOC, a quaint legacy that bugged Symbian C++ developers for better part of the platforms heyday. While frugal, it was hard to develop and to develop for Symbian, which meant things took more time and the code wasn’t as easily compatible elsewhere. Qt was going to solve this issue for application development, but who knows how the plumbing beneath still looks like and what it was like to develop it further. Some have argued that as a hardware and telecommunications organization first, Nokia’s software credentials were lacking. Indeed, before the cuts Nokia was employing far more people to implement Symbian than, say, Apple or Microsoft had working on their mobile operating systems (perhaps even more than their desktop operating systems) with arguably much less results to show. Even before Stephen Elop’s watch, the wait for Nokia N8’s Symbian^3 to release and mature was long and unsatisfying, not to mention the Nokia N97 debacle.

Clearly, not all was well in the land of Symbian. But I would like to especially revisit those Stephen Elop’s comments from around February 11th above. Nokia specifically seemed concerned about two things: the speed at which Symbian could be developed reliably as well as difficulties adopting future hardware features. Compared to contemporary mobile technology, which approaches desktop and gaming console levels of performance and complexity, Symbian (and EPOC before it) was originally designed with a very different kind of mobile device in mind – one that had very limited resources and none of this multi-core hoopla. Symbian is comparably frugal, but also quite alien compared to other smartphone operating systems that have adopted internals more akin to those in desktops and servers. Different internals also mean, that less reference drivers and software from chip manufacturers is readily available for Symbian (unlike, say, for Windows or Linux-based systems), and more needs to be done internally.

And I think, at the end of the day, this probably was the reason to ditch Symbian. If recent leaks are to be believed, Nokia is having trouble with their sole dual-core Symbian design and are considering abandoning it completely. These are rumors, of course, but to find more confirmation for this line of thinking, one doesn’t have to look any further than towards the current crop of Nokia Belle devices on the market (e.g. 701). Released last fall at a time when the competition was boasting dual-core processors, modern resolutions and specifications, Symbian’s finest were released with a single-core processor, and an aging ARM11 one at that, upped to a gigahertz and otherwise specs not that much different from the Symbian^3 brethren of yesteryear. These devices were likely in the pipeline before Stephen Elop started, or had a chance to impact their development, so any conspiracy theories probably must go out the window. I doubt this was just inventory or bill of materials control either. Symbian has been technically lagging for a long time. Much has been explained by its frugality, cost savings and not needing more, but combined with the quotes above I doubt that has been the only reason. Much of the reason why we often critique Nokia and Symbian for delays and disappointments seem to result from the fact that Symbian is “fragile”.

Would the axed Symbian^4 have made a difference? While boasting a compatibility-breaking user-interface that might have been an improvement, it would have been another departure from Qt development for Nokia (it was not written in Qt) and I doubt it would have made a difference for the fundamentals. And I think it was the fundamentals, the plumbing, the codebase and the development that was so different from the other major players, that made Symbian an impractical proposition going forward. Nokia could have tried to modernize it, but it was losing the pace, the battle – and felt that it would lose the war if nothing dramatic was done. And this was not happening only in the long-run, but was already evidenced in the past 12-24 months of smartphone advances that Symbian had missed.

With the emergence of dual-core and quad-core processors as well as a host of new faster, fourth generation communications technologies (not to mention user experience improvements and whatever ecosystems and services the future would require), Nokia would have had to adapt Symbian to techno-realities and architectures quite distant from its origins. Symbian Donna and dual-core may still emerge of course, and Nokia probably has some really nice Symbians still up its sleeve in the camera domain, but the writing was on the wall for the software. Symbian will be sufficient for the short-term, but the long-term would have been too big of a hill to climb. Software architectures on mobile devices will keep on getting more and more complex and a more modern approach is necessary. Even RIM/BlackBerry realized this and is moving from their original to a completely different operating system for their multi-core mobile phones.

The implications of this Symbian realization must have been quite significant, because the whole Qt strategy relied on Symbian being the base upon which to build. It was to be the dominant force, which would help MeeGo enter and eventually expand. It was to be the Qt mass-market that would support the MeeGo niche while it grew. With that base crumbling faster than expected, MeeGo would have had to go more of it alone – and to build an ecosystem which nowadays needs to include everything from mobile apps to televisions to books to media content to office to gaming to advertising. Not only wasn’t MeeGo ready to cover all pricepoints, MeeGo was continuously late as well. Again it came down to speed of execution. On February 11th, Stephen Elop put it like this in their media Q&A: “Our belief, and this was an important driver on our decision-making, is that we can move faster in the introduction of new devices across the price-range… through this partnership [with Microsoft] than what we’ve ever done before.”

In hindsight, Nokia probably should have started the move away from Symbian already back in 2007 and done everything to advance Maemo as a sort of Symbian NT – as one commenter so pointedly put it. In the beginning of 2011 it was, or at least Nokia felt it was, too late for that. The rest is history of course. Nokia chose Windows Phone as their new primary smartphone operating system after concluding they could best be competitive by focusing and helping push that ecosystem, and by being a preferred player within that ecosystem – instead of joining the crowd at Android or trying to build an ecosystem alone with MeeGo. Short-term pain for long-term gain, goes the hope. The state of Symbian seems to have been a major factor behind that choice.

Update:

Symbian^3 resurgence myth. How Nokia Q4 2010 results show smartphone sales collapse well in progress
http://www.staska.net/2011/07/25/symbian3-resurgence-myth-how-nokia-q4-2010-results-show-smartphone-sales-collapse-well-in-progress/

Why did Nokia Symbian smartphone sales crash this year? Infographic
http://www.unwiredview.com/2011/07/22/why-did-nokia-symbian-smartphone-sales-crash-this-year-infographic/

Tags: featured

Category: Maemo, MeeGo, Nokia, Symbian, Windows Phone