The Butt Ugly Weblog

What does the Nokia-Microsoft announcement mean?

I've been mulling over the Nokia-Microsoft announcement for a while to avoid a knee-jerk relation (that I did on Twitter) and here's my take. This is a fairly long post, but I've tried to take a larger view on the topic with lots of detail.

Symbian

Good riddance. Symbian's main problem was always the fact that nobody actually wanted to code for it. It was not built for developers, it was built to be understood by a very small core group of engineers, who would build a few apps on top of it, and that was all.

And I'm not talking about 3rd party developers here. When the system and the programming interfaces are complex, even developing the platform itself and the phone built-in apps costs more money, takes more time, and is more error-prone. Developer productivity is low. This is the reason why Nokia lagged behind in Symbian: it was just too complicated. Yes, technically the kernel is brilliant and power-efficient and fast. Unfortunately, nobody higher up was able to co-ordinate a rescue effort until the idea to buy Qt came along - which would've solved all these issues by increasing developer productivity significantly. Unfortunately it was too late.

Also the fact that Symbian was not fun meant that you would not have a skilled pool of enthusiasts to hire from. People coded for Symbian only because someone paid money for it - and that means that the best, brightest and most productive guys would go to work elsewhere. (Obviously there were exceptions. There always are. But over the years, the most appreciative comment I heard about Symbian from a programmer's point of view was "It's no worse than the others." That tells a lot.)

Windows Phone

Out of the different possibilities, Windows Phone is probably the best choice to adopt. It pains me to say this, but really, it is. Competing in the Android ecosystem with Nokia's overheads against all the Chinese manufacturers? No way. At least with Microsoft there's Microsoft's marketing, traditionally excellent developer tools and relations, and some very good technology and brands (like XBox).

Also, Nokia will probably have much better impact on Windows Phone development than they would for Android, considering that in short order, they will be shipping a massive majority of Windows smartphones. In the Windows ecosystem, they will be the giant.

Windows Phone is -- as S. Elop pointed out in his famous "burning platforms" -memo -- an act of desperation, and the best of the worst options. Now, nobody really likes it, as everyone would've liked to have seen Nokia build something really beautiful and awesome, but they didn't and couldn't, so there you go.

Meego

Meego's situation is interesting, and it's easy to read too much or too little into it. Microsoft hates everything GPL with burning passion, and Linux has been a sore point for them for years. I would not put it past them to hide a special clause in their contract that in exchange for lower Windows license prices, Nokia will stop any Linux development that may threaten them. (Though this is obviously pure guesswork, and nobody in their right mind would ever confirm such a thing.)

In this case, Nokia could not say that they will drop Meego straight away because of existing agreements with Intel and others. But the phrasing "research OS" essentially means "only a handful of people will do any serious work on it." Intel and some others may continue to use it, and it may become an interesting tablet OS at some point. Who knows. But I wouldn't hold my breath waiting for anything. It's no longer on Nokia's critical path for the future, and in fact is in direct conflict with the strategy of their most important partner, and resourcing will reflect that.

This will also have direct impact on Linux in general, since Nokia has been a fairly active contributor to the kernel (2.3% of all kernel contributions, whereas Google is only at 0.7%. Nokia has been contributing over three times more than Google - and Google is actually shipping an Linux-based operating system!).

HOWEVER, it may also be that Nokia's careful wording about Meego is designed not to piss off Microsoft. Every single other major mobile phone manufacturer has their own OS, and as Steve/Apple has many times pointed out, the only way to make a truly great customer experience is to control both the hardware and software. It is entirely possible that Nokia is planning to still roll out Meego as their premiere smartphone OS, and Windows Mobile is only an interim solution. However, it is no longer on the critical path to succeed, so they can afford to continue development in a smaller group that can hone it to perfection. This move gives Nokia time to rebuild their software engineering process by shedding out all of the old baggage - if they are brave enough. This is an opportunity for Nokia to reset themselves and really transform themselves into a proper software company, if they so choose.

Now, Nokia might've been desperate, but they are not stupid. Considering that mobile phone manufacturers have consistently failed to work with Microsoft, Nokia needs a plan B. Now they have a couple of years of time to execute it. If they do not, then it's obvious that they've thrown in the towel and are only waiting for Microsoft to buy them.

I think the way to figure out which way Nokia is going to go can be seen on how they treat Qt. If they choose to keep it, Meego is going to be the future. If they choose to let it wither and die (or sell it off), Meego will become just a curiosity. Qt is the key for Meego developer adoption, and the only way it could become a feasible smartphone platform. In my opinion there's no point to develop Meego just for the tablet market; it would be really hard when you couldn't leverage the same development effort as on the phone side.

At any rate, Meego will lose a number of good people, simply because they are fierce open source advocates who will not work for a company which has sided itself with the Evil Empire. I'm hoping that this is not going to be a large number.

S30 & S40

Not a whole lot will happen here. These guys (who are, by the way, really good at what they do, and totally underappreciated) will continue to ship Nokia's own operating systems and sell billions. Nobody in the tech industry will care because they're not shiny, but a massive amount of the world's population will continue to buy them and for the next billion, the mobile experience will be defined by them.

Even if Nokia is facing a lot of pressure on the low end too, I think that the fact that Nokia no longer has to worry about smartphone development means that they can now dedicate more resources to the feature platorms. Nokia's brand value is still good, the platform is very mature, and they don't have to pretend to keep any sort of backwards compatibility.

Ovi

The mobile industry is still going towards services. Or to be precise, internet services are going to mobile. Microsoft isn't very good in the service business (they're also geared towards shipping boxes), so Nokia will need something. My guess is that they will keep rebranding OEM services under Ovi (like Ovi Chat and Ovi Mail are now run by Yahoo), while concentrating on the few services that are doing well: Ovi Maps and Ovi Music Store, and the necessary infrastructure to run them. I don't think Ovi Store is going to go away as such: Nokia needs an S40 app store; there will still be a significant amount of Symbian devices out there which need a store; and Nokia has an extensive billing system already in place.

If you do, however, see talk about Bing Maps replacing Ovi Maps and Qt being sold, then you know that Microsoft and Nokia are heading for a merger. Or a buyout, which seems more likely at these stock prices.

Java & Oracle

There has probably been much teeth-gnashing and furious typing into Excel sheets at the Oracle boardroom: You see, even though nobody was really paying attention, with every single Symbian phone shipped a Java Runtime. So, in essence, Oracle will lose licensing money from the 100M Symbian phones that Nokia was shipping every year, and will lose the only smartphone platform which had their technology. To compensate, Oracle will need to tighten its screws on other Sun technology they have, possibly upping their effors against Google. Which can be a problem.

Qt

Qt guys are waving the good flag, saying that Qt is still going strong and that they're still hiring. As Symbian is being ramped down, there will be no new projects started for it - and all the old ones are using the native Symbian APIs anyway. There will be no Qt for Windows Phone, as it would compete against Microsoft directly. So Qt for mobile is only with Meego, and as I've said previously - the way that Nokia is going to treat Qt will reveal their intentions for Meego.

However, what is interesting is that Qt is the basis for KDE, which is the other major Linux desktop system. In fact, if Nokia guts Qt development, it will strike a very, very serious blow to Linux as a desktop environment. If Qt goes down, a lot of the Trolls will probably walk out and start a new company that will continue to develop Qt, but since copyrights stay with Nokia, this will be a tougher sell than previously. This would obviously please Microsoft.

If Nokia sells Qt, Google would be a good recipient. They would gain a valuable software asset (and many contributors to Webkit, which is the most important mobile browser engine right now), and could provide a ready-made development environment and tools for their desktop and Android OS. If their discussion with Oracle turns really sour over Java, then they may have to conjure up a good native environment, and Qt could be it. However, more likely the buyer would be either Intel (who would then have a solid continuation for Meego) or another embedded systems manufacturer.

NSN

I'm with Tomi Ahonen on this one - NSN will be sold sooner or later.

Working with Microsoft

I feel a bit sorry for the people who will stay at Nokia. Or to be precise, I feel sorry for their families. Microsoft's HQ is ten timezones away, and that means that teleconferences start at 6 pm for Finns. Or later.

Now, I kinda do like Microsoft's recent efforts. They've been surprisingly nice to people who jailbreak Phone 7, XBox and Kinect are cool, and even Microsoft Office X on my Mac is actually quite pleasurable. So these guys know how to build and ship software, so it'll be okay from that perspective.

But unfortunately, the Earth is spherical, and if Nokia's phone development stays in Finland, it (and the lives of those connected) will be seriously impacted by this. I would not be surprised if Nokia's main phone development were moved eventually to existing locations in the US, leaving only the corporate HQ here - and then it would not be a long time before it too would leave Finland.

Impact for Finland

Short term - this will be devastating to morale and job market once the reality starts sinking in. When Symbian goes, so do all the support functions (HR, WR, etc). And so will many people in the other companies that have been providing subcontracting for Nokia. The job market will be flooded with people with people who don't necessarily even know what it is to be unemployed, or have a skillset that's suitable only for obsolete technology.

Employers (with money) will of course rejoice, since they can now hire good people cheaply out of a large and highly educated pool. Expect average wages to go down. However, there aren't enough employers in Finland to suddenly hire all these people. Some will try entrepreneurship, but few will have the drive and the ideas to carry them anything further than what the separation package carries them. Certainly the best will be snatched by competitors, and everyone with a house loan (which is almost everyone) will seek a steady paycheck rather than take their chances. Most of these people weren't in Symbian because it was fun and exciting - they were there because the money was good. That attitude does not a good entrepreneur make.

With the elections just coming up, no politician is able to do anything other than issue stern statements, and I just don't trust that the Finnish government is capable of quick, decisive action on this. If they do something, it will have impact two years down the line, and while that will be great, the short-term impact of anything they do will be equivalent to handwaving. In fact, I am not even sure whether there is anything that could possibly be done by the government, other than plan for the reduction in tax income.

Finally

Of course, the impact might not be that great, depending on how much custom software Nokia will want to bundle with their new phones. They still need to differentiate, and be able to offer cool apps to counter the larger developer populations of iOS and Android. So there's still hope for good subcontractors and developers.

And, on the other hand, a lot of people will get a very long, paid summer vacation. So perhaps that'll make the country happier, overall :-)

(Phew, there you go. And that - I hope - is the last thing I say on the topic. Unless the great, unwashed masses of the internetz completely misunderstood me, and come flooding here to flame me. Just needed to get this stuff of my chest so that I can purge myself of all things mobile. I'm beginning to be really bored at it - the same way as I was bored at Amiga vs PC vs Mac so many years ago. Discussion about mobile operating systems damages your sanity; the interesting stuff is happening elsewhere now.)

Update: I just want to make it clear that I am not employed by Nokia. I used to be, but I left over nine months ago.