The last of the Nokians

There was quite a bit of buzz about the N9 last week, including a lot of gadget lust for the device. There’s a lot of love out there for Nokia yet, and the N9 is a handset to bring this out. Amid the inevitable questions about what could have been if Nokia had stayed with MeeGo as a part of its portfolio, there were even showings of hope that the relegation of MeeGo to a research project wasn’t final. A twitter petition was started and even the occasional  blog post proclaiming a change of heart on the part of Steven Elop were written.

Of course this is nonsense. Even a huge success of the N9 would not change anything at Nokia, not with the current management in place. And then, of course, there is the unfortunate fact that the N9 will not be a success. While some of the reactions to the N9 really were ecstatic, there is no telling how much even of that will keep until the actual release sometime in the fall.

The hardware of the N9 certainly looks gorgeous. Colorful yet simple, with a nicely curved screen, it is something to be desired, and feels at place among the current high-end phones. Once you dig underneath, however, it comes from the same family of SoCs that powered the N900 – which was announced 22 months ago. High-end hardware back then, but now outdated in this segment. Even worse is that performance upgrades can only come from switching to a newer line of SoCs, which needs new OS adaptations under the hood. A reason for using this SoC is that the core OS is the follow-up to Maemo 5, which powered the N900 – and which was canned at the beginning of 2010 when Nokia announced the cooperation with Intel on MeeGo. There’s Qt on top, which as a cross-platform toolkit was developed independently of Maemo and MeeGo, and thus probably not as affected by the recent changes. The entire thing is not fully MeeGo compliant. This strongly indicates that MeeGo was, as a matter of fact, not close to launch-ready, since it was apparently easier to go back to a software/hardware combination from early 2010 than to finish anything currently in development. It also indicates support for current top-end hardware is nowhere there yet.

Thought experiments about reviving MeeGo not only have to take this into consideration, but also that a considerable amount of the development team has left for other companies. Rebuilding the team, and then continuing a product that was further from launch than software abandoned over a year ago – well, it’s hard to see a product launch there before the first WP 7 devices.

So the N9 does not clearly demonstrate  tech leadership on the part of Nokia, as some would have it, but instead speaks of delays and execution problems – once again.

Now users aren’t interested in OSes and only peripherally aware of release schedules. Functionality and ecosystem are what matters to them. The functionality of the N9 out of the box is certainly impressive. All the basics are covered, there is deep integration of social networks, and Nokia Maps on the N9 could well turn out to be the best satnav experience on a smartphone yet. There is even Angry Birds (which was also available for the N900, so this is not that surprising).

Beyond that, once you need an ecosystem to extend this functionality, things aren’t looking so good. There is a small dedicated developer community around the Maemo devices, but the main repository currently only lists 657 applications. A lot of these are going to make it onto the N9, but neither is that an impressive number, nor do most of these have mainstream appeal in their funcionality or UI. Qt as the development toolkit eases porting of newer Symbian applications, and, to be fair, the apps situation there has improved a lot, but again the question is how many developers are actually going to bother to adapt their applications for a single device.

Alien Dalvik is seen as a trump card to solve the application problems by many. It allows Android applications to run on the N9, so potentially the entire Android application catalogue could run on the N9. Aside the point that this is not true for anything that runs native code (read: at least all games that use computing-intensive graphics),  this again requires some developer effort in repackaging their applications. This is not going to happen for the vast majority of the applications, not for a single device. Add to that the problem of discovery, installation and payment absent a central marketplace, and the trump card turns into a dud for mainstream users. Nobody’s willing to hunt the web for install files anymore, or to pay on developer’s websites.

What about the UI, which has gotten people so excited? Could this play a role?  It’s certainly different, it’s fluid, it has interesting ideas. The same was true for webOS. Palm bet the company on it, lost big time, and is now part of HP. So no, the best UI in the world can’t bring about the success of a smartphone in the current market, not without an ecosystem around it.

Not that Nokia is betting the company on the N9. The bet here is without a question Windows Phone 7, and the company management did their best last week to make this clear. First there was the ‘Sea Ray’ leak. A professionally produced video about an internal company presentation of the first Nokia WP7 device, leaked to a Hungarian website. Timed perfectly to land square in the middle of the post-release chatter about the N9. Then Steven Elop followed this with an interview to a Finish newspaper, in which he plainly stated that no sales success of the N9 could change the current course. As he also recently said: “Plan B is to make sure that Plan A works!”

I guess he needn’t worry much about a sales success that would call this decision into question. Nice hardware, with a spec sheet that will be even less stunning by the time the N9 is in shops among a slew of dual-core top-end phones, and with a non-existent ecosystem is unlikely to attract much more than the Linux enthusiasts and Nokia fans who bought the N900. For them the 91,000 units, which Eldar Mutartzin has claimed are the entire production run for the N9, might suffice. It is certainly believable that with no real commitment behind the launch, there are no plans for producing more than a token amount.

Which brings us to the why the N9 was launched at all. The situation here looks murky at best, with several theories out there.

Following the February announcement there were stories that Nokia is contractually obliged to Intel to produce a MeeGo device this year. It is hard to see how a device running on Texas Instruments ARM hardware could benefit Intel directly, but it’s possible they insisted, either to spite Nokia or to get some affordable developer hardware capable of running MeeGo out there. In case of the latter, they might also be willing to accept this as the obligation fulfilled, since it otherwise they could well argue that this is not, in fact, a MeeGo device.

Another line of reasoning has fractions within Nokia still pushing for MeeGo, and the N9 as their baby. I have no idea whether anybody would actually have the pull, e.g. through connections to the board, to get this launched against the resistance of the upper management, but I strongly doubt it.

But then the new management might have found the release to be useful as well, to demonstrate to the world, through its failure in the market, that choosing WP7 over MeeGo was the right decision. The leak of the WP7 video lends a little credence to this theory . Maybe that there are fractions that want this to succeed, and ones that are actively working against this.

Finally, there is the possibility that it is really just there to test some concepts, and as developer hardware for MeeGo as a skunkworks project. The design of the casing, after all, will be reused for the first WP7 device, and there’s even talk of reusing the interface at some point in the future, so the extra work might have been acceptable for this.

Whatever the reason for its launch, in the end it doesn’t matter: It is clear that the N9 is the last device to truly represent Nokia as an integrated hardware and software company – at once intriguing, beautiful,  and doomed. It is the last of the Nokians – and despite, or maybe because of this, I want one. In black, as a beautiful memento of the Nokia that ended on February 11th 2010.



Filed under mobile, Uncategorized

7 responses to “The last of the Nokians

  1. I think you’re focusing too much on the negatives.

    As Urho Konttori describes (, getting the N9 OS done was a major feat. It doesn’t speak to delays, but of a team OVERCOMING those delays… and other obstacles.

    As a former employee (US factory QA engineer for the 770 and N800) I know well what the Maemo and MeeGo teams had to fare with. I also know that if the company had not been so cripplingly conservative, then we would not be having the MeeGo-WP7-Android-etc debate. MeeGo (or Maemo) would have already been proven.

    I can’t see Nokia prospering with WP7. There are more hurdles in the way of success there than there were for MeeGo, and one cannot get around the fact that where smartphones are concerned Nokia really will be an OEM for Microsoft. I just can’t see that as a win for Nokia.

  2. First of all thanks for your comment. I really wish I could have written something more positive.
    The team may well have overcome huge obstacles and delays to get this out of the door. I congratulate them on that, and, as I said at the end, I want a N9. I do think it is a beautiful and interesting device.There is no denying though that this should have come out much earlier. 22 months delay between iterations for an OS in the mobile space is much, much too long. And whichever way you turn it, the hardware platform is old, so a transition to a new generation would have to be the next step forward, and that means further delays.
    I agree that things could have been different with a different release date, and that Nokia should have pushed ahead with Maemo instead of the detour with MeeGo. But things are as they are. Much as I’d have loved to see a competitive, fully-open Linux smartphone OS in the market by now, the N9 isn’t the device to achieve this, and Nokia is not going to be the company to release one in the foreseeable future. For now the best we have here is Android with its limitations, both technically and regarding the openness.
    And finally, the question whether Nokia will prosper with WP7 was entirely outside of the scope of my post. They are presently committed to it as a platform, and this means that the N9 is the last of its kind. They wanted to get it right with the fifth iteration, and they may well have, but company politics killed this one.

  3. Thank you for your article. As a Linux enthusiast, I felt betrayed that this exceptional device would be released, then abandoned; the hush, hush release of the WP7 was very manipulative. I read that the WP7 is not compatible with QT– an doesn’t QT drive some of the great gui we pick up from the demo of the N9? So there is indeed a mystery here. When Microsoft is in the game, I expect money, stock valuation, and market share is more likely the driving force rather than creative development, usability and yes, fun!

  4. Frank

    The fixation on dual-core hardware is rather naiive. Android dual-core devices are showing little advantage from the second core, indeed they exhibit worse battery lifetime. As with a desktop PC, it’s often the addition of more memory that results in a significant performance boost, and the 1GB RAM in the N9 clearly allows it to run a fluid UI with responsive multi-tasking, even on an “old” TI SoC.

    You’re fixation with dual-core is a bit like the fixation with bigger numbers that has blighted smartphones for several years, more megapixels, more megahertz etc.

    If you’ve got a well written OS you don’t need more powerful hardware for the sake of it, and most users would prefer a battery that lasts all day.

    • I’m not one to blindly follow numbers games like the megapixel race, e.g. I’d have much preferred an N8 with 6 Mpx at the same sensor size, and gotten the benefit of a much better low-light performance.
      The N9 seems to run fine on its hardware, and the 1 GB of memory is a huge improvement over the N900. The thing is, most people do care about the spec sheet, and, placed next to dual-core monsters, the N9 loses out on the central high-end spec this year, one that a lot of people do want for the sake of it. (RAM, on the other hand, is hardly ever listed, or considered important.)
      When friends ask for advice about smartphones, it’s things like processors, screen resolution etc. Hardly anybody asks about battery life. It is something that only comes into play once you leave the store with your device and enter the real world. At that point it’s too late to reverse your decision. At least that’s the only explanation I can think of for why people accept what they’re being offered in this respect at the moment.

      • Frank

        So you admit that most people care about the wrong things (big numbers rather than better battery life or better quality photos etc.), yet you perpetuate such fallacies by repeating them in your blog?

        Wouldn’t your blogging serve a better purpose if you were to explain why an older (and now thoroughly understood and optimised) 1GHz SoC with capacious 1GB RAM allied to an efficient OS may actually be as good if not better than a device with a brand new screaming dual-core CPU, often limited memory, no GPU hardware acceleration of the UI (even with dual core the Android UI can still be sluggish) and a rapidly diminishing battery life?

        You say you don’t blindly follow the numbers game, but you pretty much wrote just that for this article. Perhaps in future you’ll be a little more open minded and come to realise that not every OS needs a dual core processor to get the job done (and if I’m honest, Android neither needs a dual-core processor or as an OS does it use hardware resources effectively).

        • My post was not about whether the N9 is a good device in itself. As I said at then end, I’d personally like to have one (and not just as a memento).
          My post was about the perception of the N9 in the marketplace, about what it shows about Nokia’s internal development processes, and the big question of why it was released at all. In this context single- vs. dual core is not about technical merits, but judged by different standards.
          Regarding the marketplace: Saying that people care about the wrong things is a value judgment that you or I are allowed to make. Nokia as a company that wants to sell products in the marketplace has to either go with what the public cares about, or manage to educate the public about what they think is right and change public opinion. Otherwise they fail. Customers care about spec sheets (hence the megapixel race), and so do most journalists who cater to that public. Dual-core makes for a nice headline and a box to tick, while performance judgements are less easy to communicate. When looking at this, then releasing a single-core flagship device to show tech leadership in mid-2011 is at least problematic.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s