Nokia’s problem

Last week Nokia CEO Olli-Pekka Kallasvuo (OPK for short) was replaced by the first non-Finn to lead the company, Stephen Elop, head of Microsoft’s business division (mainly Office). This is big news that might change Nokia’s perception as well as its strategy.

OPK’s tenure as Nokia chief was not lucky. Barely in office he was confronted with the launch of the iPhone, and this issue overshadowed the rest of his career. For the full story of OPK, see, as always, Tomi Ahonen.

After four years of doing little except producing one of the worst touchscreen phones in history, the N97, Nokia was perceived as a loser, and pressure on OPK to resign was growing. Last Friday things came to a head when Elop’s anointment as his successor was announced.

(This article has been translated into Spanish.)


The politics of this move are well thought-out. Elop is a software man, and he is an American — well, technically he’s a Canadian, which makes him more palatable to a European company, but for all practical purposes he comes from the US.

Externally, Elop’s appointment is aimed at placating the US financial world and blogosphere. That seems to have worked: from John Gruber on down Nokia is cautiously given the benefit of the doubt again.

Internally, Elop’s appointment is aimed at the entrenched hardware culture within Nokia,

particularly the 500 VPs with largely radio engineering experience who [...] have responsibility for areas beyond their expertise,

as they are described in the Risku Manifesto a former senior executive wrote. This manifesto calls for the wholesale removal of dead wood in the senior management, such as Marko Ahtisaari, whose job qualifications are that he’s the son of the former Finnish president and Nobel Peace Prize winner; Mary McDowell, who was hired to break open the American market but mysteriously didn’t; and CTO Tero Ojanperä, who released no new technologies in the past two years.

Gruber’s piece contains valuable extra material. A former Nokia software engineer remembers:

It was not uncommon for us [software engineers] to give [the hardware engineers] code that ran perfectly by their own test, only to have them do things like reduce the available memory for the software to 25% the specified allocation, and then point the finger back at software when things failed in the field.

Bottom Line: Nokia is a hardware company that hates software.

That will have to change. Software matters nowadays.

Logic jump

It’s at this point, however, that Gruber makes a curious jump of logic that I can’t follow and he doesn’t explain:

Nokia needs to settle on one software platform for mobile devices, very soon.

Gruber’s likely source, Jean-Louis Gassée, is more explicit:

Today, Nokia pushes devices that use older Symbian S60 stacks, newer Symbian^3 and Symbian^4 engines, as well as a mobile Linux derivative: Meego. Imagine the chuckles in the halls of Cupertino, Mountain View, and Palo Alto. Even with plenty of money and management/engineering talent, updating one software platform is a struggle. Ask Apple, Google, or HP, and the chuckles quickly become groans. Nokia thinks it can stay on the field when it’s playing the game in such a disorganized fashion?

Others make a similar point. The conclusion is that Nokia must embrace either Android or Windows Phone 7. I find this argument distinctly uncompelling, not to say misinformed and out of touch with reality.

(To his credit, Gruber does not draw this conclusion, although he still sees Android or Windows Phone 7 as potential options.)

Two OSs

Why on earth wouldn’t Nokia be able to maintain two operating systems?

Apple does it: Mac OS and iOS. Google does it: Android and ChromeOS. Microsoft does it, 7, Vista, XP, and maybe even older versions. And Windows Phone 7, of course. And I’m sure HP has a few OS skeletons in the contractual closet.

Samsung has firmly settled on a two-OS strategy. It will serve the high-end market with Android, and the mid-range market with its homegrown bada OS. I haven’t heard any chuckling in assorted US halls about the Samsung Galaxy.

HTC has become a very strong player in the high-end Android market, and seems poised to do a repeat performance for Windows Phone 7. There’s a definite two-OS strategy at work here, and I haven’t heard any complaints about it.

Nokia’s only sin seems to be that it doesn’t use an OS that the US chattering class has decided is worth chattering about. In penance it should adopt either Android or Windows Phone. (But not both. It can’t maintain two OSs, after all. Only American companies can do that, although they chuckle about it afterwards in their halls.)

I’ll let you in on a dirty little secret: Nokia has even more OSs than just these two. It has S40, for instance, and I wouldn’t be surprised if they have even more OSs for really cheap feature and basic phones. But talking about S40 is not hip and cool, so let’s not.

Nokia’s OS strategy

Nokia has serious problems, but they do not include its fundamental OS strategy. Rather, the problem lies in execution.


Despite everything Symbian is a feature, and not a bug.

Yes, Symbian is a turd. Its UX was gang-raped by a committee, its architecture is complicated and weird, and its attempts at running a touchscreen horrendously lousy (so far — I haven’t seen Symbian^3 yet).

So you could conclude that Symbian is no competitor for Android or iOS, Windows Phone 7 or webOS. Fine. I agree, actually.

But this conclusion is worthless. Symbian isn’t supposed to compete with those OSs — that will be MeeGo’s job. Instead, Symbian is now firmly aimed at the mid-range market. It must compete against BlackBerry and bada.

Truth to tell, it has a problem there, too. bada is shiny and new — not an iOS-class operating system, but a lot better than Symbian. BlackBerry’s OS6 is heading in the same direction. Symbian is getting old even in its own market.

But the very fact that Symbian is old means that it can run on low-specced, hence cheap, hardware. And Nokia’s ubiquity in the developing countries will make sure that the world’s poor are going to get access to our high-brow mobile ecosystem via Symbian. And not via one of the more sexy OSs.

Yes, Symbian is in a slide from the top of the market to the bottom. But what’s wrong with that? Nokia serves the bottom part of the market, too. Apple doesn’t. HP doesn’t. Android might, but doesn’t right now. Why on earth is Nokia wrong when it serves a market that the US companies ignore because they’re too busy chuckling in their halls?


Nokia desperately needs a high-end smartphone OS, that much is true. And it has one: MeeGo is being groomed for exactly that task. But where, you might ask, are the MeeGo devices? Isn’t it time one or another of them hit the market?

And here we have finally reached Nokia’s true problem: it isn’t fast enough in its software development. Not to say bloody slow. If you want to criticise Nokia for that and hope Elop’s appointment will help here, be my guest.

The Nokia N900, which runs on MeeGo’s precursor Maemo, was released about a year ago. I tested it and it is ... not bad. It’s certainly light years ahead of Symbian, and I can see, in time, and with a lot of effort, a N900 descendant becoming a serious competitor to iOS, Android, webOS, and Windows Phone 7.

But a year has passed and there is still no successor to the N900. That’s a big problem that has to be addressed and solved. Feel free to criticise Nokia here; I certainly do.

Finally, Nokia must take MeeGo seriously as a software platform. This is a subject we can’t say anything about due to lack of data. If you want to be skeptical of Nokia’s software strategy, be my guest. Just remember that something might change after the current shake-up.

Android? Windows Phone 7?

However, embracing either Android or Windows Phone 7 is nonsense. It would lead to another year of doing nothing. Nokia would have to incorporate the new OS, create prototypes, then decide on a device to bring to the market. That takes time; time its competitors will use for bringing out even more advanced phones.

Also, switching OSs would mess up Nokia’s partnerships with Intel and other device vendors such as LG who want to jump on the MeeGo bandwagon. That’s not a very good idea.

Finally, if Nokia would switch to somebody else’s OS it would give away its chance at vertical integration of device, OS, and app store; and this integration is currently the Holy Grail in the mobile world. (That’ll change; these things are dictated by fashion, but for now it’s true.) It would become beholden to Google or Microsoft for a crucial part of its platform.

As business advices go the Android/Windows Phone 7 idea is absolute nonsense. What Nokia should do is clean up its MeeGo act and get phones out there. As to turning the hardware-centric outlook of the company on its head, that has to happen, too, but it will take longer.

Nokia’s problems

Yes, Nokia has very real problems that will have to be solved. Yes, these problems should be discussed. But let’s please concentrate on the actual problems at hand, and not invent new ones that have no base in reality.

Nokia’s basic OS strategy is sound: MeeGo for the high-end, Symbian for the mid-range, and S40 for the low-end. But now it has to actually execute this strategy instead of fooling around.

Nokia, release a MeeGo phone. Before Christmas. But don’t bother with Android or Windows Phone 7.

This is the blog of Peter-Paul Koch, web developer, consultant, and trainer. You can also follow him on Twitter or Mastodon.
Atom RSS

If you like this blog, why not donate a little bit of money to help me pay my bills?



Comments are closed.

1 Posted by Da Scritch on 17 September 2010 | Permalink

This is nearly a good analysis EXCEPT that instead of Google, Microsoft and Apple, mobile phone is the only business of Nokia.

That means that Nokia has only a good consumer experience with mobile phones.

And Nokia can't negociate with mobile phones operator to NOT custom the OS/UI of the phones with their own ... "modding". And those modifications are a real plague to mobile experience.

2 Posted by kl on 17 September 2010 | Permalink

Moore's law will make Symbian market disappear.

In few years hardware for Adroid will be dirt cheap.

Sure, there's a lot of phones to sell in African countries and India, but you can't make fortune by racing to the bottom.

Even today Nokia needs to sell truckload of phones to earn as much as Apple on single iDevice.

3 Posted by AlastairC on 17 September 2010 | Permalink

I doubt that a smart-phone OS will replace the Symbian phones any time soon.

Running iOS/Android takes more than chip power, the whole interface requires (relatively) expensive components such as a large (battery draining) touch screen.

4 Posted by Digital Jedi on 17 September 2010 | Permalink

Kl, are you delusional?! Yes, itoy sells millions, but Nokia sells BILLIONS. With a "B". I'm an Android guy and even I know Nokia isn't going anywhere. Definitely not because of CRApple. Nokia has the #1 OS world wide. CRApple's sh*tty OS is barely a blip on the world market. You must be an iSheep. Sure, Nokia has made some mis-steps but with handsets like the N8, E7 and N9 in the pipe and a software centric guy at the helm, alot of people are going to be eating crow and bandwagon jumping when Nokia makes a "come back". Scratch that, don't call it a "come back". They've been here for years.

5 Posted by Eric on 17 September 2010 | Permalink

Pretty darn good analysis and post. However, I'm not so sure about Symbian. Symbian^3 seems to be actually quite great IF we forget few things: it still has old UI (UX is better now, but certain old-fashioned elements (like the top bar where signal strength, etc. are) are still old-fashioned and not sexy) and with old UI comes few other problems like navigational stuff.
Now, Symbian^4 should change the whole UI and bring quite MeeGo-like UI.

About MeeGo: I'm pretty sure it'll deliver, but things tend to take time. Especially in the open source (and in Nokia) world. Google is developing Android practically by itself so it doesn't have such problems.

Symbian isn't going anywhere. And in general: I just don't get it when people like to talk more about profits than about the actual business. Especially Apple fans seem to be proud how Apple manages to make fortunes (even if they are the ones giving the money). Of course profits are important for investors and companies, but Nokia has been very much profitable even with low-end phones. That's their huge strength.

It bothers a me a bit when you say that you doubt that "smart phone OS" will replace Symbian phones.. I mean, Symbian is very much smartphone OS.

6 Posted by kl on 17 September 2010 | Permalink

@Digital Jedi: ad hominem and "funny" misspellings are not an argument.

Those billions of phones don't translate to equally impressive revenue. Slice of the market that Nokia rules in requires thin margins on cheap hardware.

7 Posted by Travis on 17 September 2010 | Permalink

Brilliant argument, Digital Jedi. That must be why Nokia's smartphone market share is cratering: everyone is an iSheep unable to see how a horrible user interface, pathetic touch implementation and a total lack of modern software add up to an awesome smartphone they have to buy.

8 Posted by Steven on 17 September 2010 | Permalink

What mid-range market? The iPhone 3GS sells for $99. That's you're mid-range market.

Low-end market? Refurbished 3G for $49 from AT&T.

Really low-end market (your phone is so terrible nobody is willing to pay for it) - I'm sure that's the market they are just drooling over.

When smart-phones sell for the same price as dumb-phones, there's no room left for one-off, featureless OSes.

9 Posted by Dan Villiom Podlaski Christiansen on 17 September 2010 | Permalink


The iPhone does not sell for 99 USD anywhere in the world. The only way you can get to pay that price is by binding yourself to a subscription, so that the phone company is certain that the actual cost of the phone — 600 USD — is covered.

For comparison, I got my Nokia 5800 about six months ago for ~250 USD, without involving a telephone company. The 5800 is not a great phone by any measure, but quite decent for the price…

10 Posted by Hamranhansenhansen on 17 September 2010 | Permalink

Apple only has one operating system: OS X.

“iOS” is the mobile (touch) user and application interface on OS X. “Mac OS” is the desktop (mouse) user and application interface on OS X. That enables OS X to run on both kinds of device.

Whether you boot up an iPhone or iPad or iPod or AppleTV or Mac or Xserve or Airport Extreme, you are booting OS X: the xnu kernel is talking to the hardware, CoreGraphics is drawing the interface, CoreAudio is providing the audio, QuickTime is decoding media, Location Services is telling you where you are, and so on. The key benefit of this is Apple adds core features like Exchange to OS X just once, and they show up on all Apple devices. In the case of Exchange, in iOS v3 and Mac OS v10.6. When they improve CoreGraphics or another subsystem for whatever reason, the benefit is seen on all Apple devices.

Microsoft has 2 separate operating systems: NT and CE. Is that working for them? No. That is why they can't make an iPad. NT cannot run on a device that small, and CE cannot run on a device that big. And only NT has significant apps. Fail.

Google has only 1 operating system: Android. It also can't run on a device as big as an iPad. Chrome has not been released and no use has been proven.

11 Posted by Steven on 17 September 2010 | Permalink

@Hamranhansenhansen - wonderful rebuttal of the multiple OS argument. Apple has 6 classes of devices running close variants of the same OS - servers, desktops, laptops, tablets, phones, and set-top boxes. Brilliant. Of course, iPods besides the Touch are the exception to this.

Not sure what you mean about the Airport Extreme though - I've never hard anything about it running OS X.

@Dan - as far as the iPhone not "really" costing $99 - as far as the average consumer is concerned, that's how much the iPhone 3GS costs. Sure, you could pay $600 for one, then pay the exact same $85 a month for service, but what benefit would that be to the average consumer? You just paid $400 extra for the same phone.

The fact that when I choose to upgrade, I can sell my old iPhone on any number of sites for the same price it costs me to buy a new one, basically makes the iPhone is free with contract. It doesn't get any cheaper than that (in the U.S. at least).

I don't think non-contract price has any effect on the mass market. And that's where the money is.

12 Posted by Andy Baird on 18 September 2010 | Permalink

"Yes, Symbian is in a slide from the top of the market to the bottom. But what’s wrong with that?"

Profits--lack thereof. Last time I looked, Nokia was selling ten times as many phones as Apple, yet Apple was making more money while selling a fraction as many phones. (I'm not considering Mac or iPad profits here, just those from iPhones.)

In other words, Nokia is working a lot harder for less money. That's what it's like when you're a "bottom feeder," working the low end of the market. It's a risky niche, where you're constantly in danger of being undercut by somebody with lower labor costs.

As you say, that's where Nokia seems to be heading. Can they survive in this new niche? Perhaps. Is it a desirable direction for the company? Not really.

13 Posted by Shaun on 18 September 2010 | Permalink

@Steven, that's a peculiar thing to the US only. In the UK you can get contracts from about £10-15 including data or PAYG for less even including data. That's where a whole tier of smartphones and featurephones sit that Apple can't touch and Nokia excels at. To get an iPhone with that kind of contract or PAYG you're paying £500 for the iPhone. You can of course get the equivalent of your $99 /$85 split but why would you unless you really MUST have an iPhone and treat the contract like a very bad high interest credit deal.

As to the whole multiple OS thing, I think Nokia has at least FOUR OSs in use. I'm sure S30 is still kicking about at the very low end. S40 is Nokia's mainstream OS with hundreds of millions of phones shipped with it. Gruber compared Symbian to Apple's iPod Pixo OS. He's totally wrong. S40 = Pixo.

I do wish Nokia would kill off the Symbian variants though. It should be S^3 from now on but they've still got S^1 and even S60 3rd Edition devices out there. Nuts. S^3 though is exactly what they needed 12 months ago and it's passable now. Evolution, not revolution.

Meego devices... Well the Meego conference is in November. That's when we'll see something.

Isn't this irrelevant though? - the platform is Qt now.

14 Posted by Tid on 18 September 2010 | Permalink

"Apple does it: Mac OS and iOS."

Kind of, except that Apple does such a good job of merging hardware and software, no user sees it this way. You can't buy iOS, or install it on a Mac or PC. You can buy Mac OS, but you can't install it on any device that runs iOS. Nobody walks out of an Apple store wondering if they got the right OS.

Apple *has* in the past been in between multiple OSs (][/Mac, Lisa/Mac, Mac/A/UX, 9/X), but they were not fun times.

"Google does it: Android and ChromeOS."

The solution to this seems to be that nobody uses (or wants) ChromeOS. :-) I think if consumers had to choose one of these when they were buying a device, it *would* be a much worse situation for Google.

"Microsoft does it, 7, Vista, XP, and maybe even older versions."

Not a very good example: these are versions of the same product. If you count these separately, then Nokia doesn't have 2 OSs, they have about 6!

Microsoft has in the past had multiple parallel versions of their flagship OS (95/NT, Win/DOS, etc.), and it's always been a giant pain for users, developers, and presumably Microsoft themselves. I think Microsoft is a much stronger company since unifying on the NT kernel with an (updated) Win95-style UI.

15 Posted by LL on 18 September 2010 | Permalink

Couple things...

1. 75% of nokia revenue is from s40 phones in the developing world. Not s60.

2. There are about 2000 nokia engineers working on s60 software for s60 devices, and another 2000 working on s40. The me ego guys have a team of 40 working on a integrated software/hardware team. S60 guys are all segmented off in Tampere, and the s40 guys are in Copenhagen. There are a lot of internal politics at play here. meego guys promote themselves as the future of the company, but it will be only over the dead bodies of s60and s40. Meanwhile, see where the money comes from.

3. S60 is failing massively because it is impossible for third parties to write software for it. Have you tried? They barely have the concept of a dll. Nok thinks that they will solve this with qt.

4. Nok is not a hardware co... They are at heart a manufacturing co. Technical expertise is derided, engineers are treated as factory workers.

16 Posted by Travis Butler on 19 September 2010 | Permalink

Speaking as the owner of two Maemo-based Nokia Internet Tablets (the N770 and the N800), I have little confidence in Nokia's ability to execute on Meego.

The problem is that Nokia has been great in developing a mobile OS that appeals to tech geeks and their developer equivalents - but bad at making that OS appeal to average users, and developers who write for average users. I have some Linux skills and could get around Maemo, but doing so was often a pain - particularly when having to drop to the command line on a virtual keyboard hamstrung by the resistive touchscreen. And while there were plenty of programs made for Maemo in the time I used the platform, they were heavily slanted towards geek hobbyists; 'productivity' apps (even something as simple as a contact manager, needed because the built-in one was missing vital elements like physical addresses) were at best ports of desktop Linux apps, usually sloppily done and poorly adapted for mobile use.

I can see why adopting Android - or especially W7 - would be something of a crippling move for Nokia, since they would lose control of a key part of their product. But they have yet to show signs that they really 'get' how to make a good consumer smartphone OS.

17 Posted by Travis Butler on 19 September 2010 | Permalink

One more comment:

"That will have to change. Software matters nowadays."

Exactly. The true lesson of the iPhone, one that Tomi Ahonen seemed to be trying desperately to ignore, is that *features that are too much of a pain to use are irrelevant.* And interface - a software-heavy function - controls how much of a pain a feature is to use. If a feature has a poorly-done interface, it may as well not exist for someone who finds it too much trouble to use.

This is a sliding scale, of course, and tech-savvy users can use interfaces that are too painful for average users. But tech-savvy users are not a mass market.

18 Posted by James Pearce on 19 September 2010 | Permalink

> Externally, Elop’s appointment is aimed at placating the US financial world and blogosphere.

Yes, partly.

Absolutely no.

Hard to believe the board of directors sat around saying "Now then; who would John Gruber like to see?"

CEOs are hired to make a company run brilliantly. One assumes commentators will respect a turnaround in results, whoever's at the helm.