Tuesday 29 January 2008

Nokia, the OS company

Nokia bought Trolltech for about $150 million, and there's all sorts of speculation online about what it means. Before I get to that, let me quickly summarize what Trolltech does:

Trolltech is a Norwegian company that makes development tools and Linux software. Its best-known products are Qt (a software layer and development tools for writing applications that run across multiple operating systems, including Windows, Mac, and Linux), Qtopia (a user interface and applications layer for Linux), and Qtopia Phone Edition (a Linux software environment for mobile phones).

In the mobile world, Qtopia Phone Edition has been the company's best-known product, although it hasn't exactly been a commercial success. Motorola uses a version of Qt in its Linux mobile phones, but not all of Qtopia. The Sony Mylo mobile device uses Qtopia, as did the Sharp Zaurus PDAs. But Trolltech had so much trouble getting a mainstream phone licensee for Qtopia that it created its own hardware prototype, the Greenphone. (Out of fairness, I should add that Trolltech has a lot of other tiny licensees you've never heard of; you can see the full list here.)

The obvious assumption would be that Nokia bought Trolltech for its phone technology, but that's not what Nokia says. The company's press release says Trolltech will help advance its "cross-platform software strategy for mobile devices and desktop applications, and...Internet services business. With Trolltech, Nokia and third party developers will be able to develop applications that work in the Internet, across Nokia's device portfolio and on PCs."

All About Symbian reinforced that message, reproducing a slide from the Nokia press briefing that showed Qt layered on top of Nokia Series 40, S60, and a variety of desktop PC operating systems (link). The Guardian quoted a Nokia spokesperson as saying the emphasis of the deal is development tools: "This is about Trolltech's fantastic tools. You can much faster develop programmes which can work on multiple platforms." (link).

Vnunet quoted an analyst saying that Nokia will use Qtopia to help deploy its Ovi Internet services cross-platform (link). I don't really see the Internet connection; Qtopia has not been a contender in the net applications world the way that Flash and Silverlight are. But maybe Nokia wants to build it into a contender.

Other analysts suggested other motivations for the purchase. Some of the commentary on Slashdot suggested that Nokia is investing in Linux to counter Google Android (link). ArsTechnica suggested that Nokia might be planning to replace S60 with Qt (link), while others suggested that Nokia plans to use Linux instead of Symbian. Richard Windsor of Nomura pointed out in an e-mail analysis that the purchase of Qt rips the guts out of Motorola's Linux plans, although he guesses that's more of a happy side effect for Nokia than the primary motivation.

But an unsigned article on ZDNet UK had the most sweeping interpretation, basically saying that this spells certain death for all proprietary operating systems (link):

Nokia's bet is that the sheer size of the Qt 4-based market will be a decisive inducement for everyone else, handset makers, operators, and pure applications players alike, and that the explosion in compatibility will amplify the market for everyone much as happened on the desktop when MS-DOS anointed the PC architecture. But unlike then, Qt 4 will break forever the idea that one part of the market can seal itself off as a profitable mini-universe, an idea as archaic in the 21st century as the feudalism it so closely resembles.

As we say here in California, I want some of what he's been smoking.


What does it really mean?

We're all assuming that Nokia actually has a coherent master plan here. Although $150m is a lot of money to me personally, it's mouse nuts to Nokia. Maybe Nokia bought Trolltech just as an experiment, or to keep it from falling into some other company's hands. The fact that Nokia's going to continue to develop its Maemo version of Linux, which is not based on the Trolltech technology, suggests a certain amount of incoherence.

If you want to be really Machiavellian, you could speculate that this purchase is the Nokia mobile phone organization's answer to Maemo -- "you tablet guys keep your version of Linux, now we have our own."

But let's assume there really is a plan, and it's aimed at competitors. About six months ago, I wrote about Nokia's ambitions to be a computer company (link). Now we see them dealing themselves into the operating system competition as well. No matter what you think Nokia's motives are, the fact is that it's now the owner of a respectable cross-platform software layer that runs on PCs and mobile devices. Nokia is now a software layer company, in very direct competition with other layer companies like Microsoft and Adobe and Sun. The deal also makes Nokia a much more important player in the open source community. And it puts Nokia in more direct opposition to the companies with their own operating systems -- Apple and Google and (once again) Microsoft.

That's a huge potential change. I say "potential" because Nokia has a lot more to do if it really wants to compete. The Trolltech team will need more investment (they have been losing money) and Nokia has a lot of work to do in developer evangelism and support to make the challenge real. But the potential is there.

I think that as the implications of the deal become clear, Nokia may have trouble continuing to partner with some of its new competitors. For example, it has spent a lot of time positioning itself as a partner to Adobe Air, but it's hard to see the evolved Qt as anything other than a competitor. Same thing for Google.

As for how this fits with all of Nokia's other products, I'm having a lot of trouble understanding how Qt will cohabit with S60 and Series 40. What exactly are developers supposed to develop for, and which user interface will the phones feature? If Nokia tries to keep all of them going, its phone software is going to look like a petit four, with layers stacked on layers stacked on layers. That makes for a nice pastry, but in a mobile phone it's a recipe for bad performance and short battery life. It's also a certain way to confuse developers.

So a lot depends on Nokia's next steps. Does Qt replace Series 40 and S60? I don't know which group within Nokia made the Trolltech deal, but I wonder if the biggest competitive battle might end up being the one inside the company, between its competing software standards.

Thursday 17 January 2008

MacBook Air: Object of lust or awkward compromise?

It's been interesting watching the reactions to Apple's announcements this week. Probably the most predictable was the disappointment many people expressed (link). After the iPhone announcement last year, almost anything was going to be an anticlimax. At this point Steve Jobs is competing with himself at these keynotes. Never mind that he single-handedly got as much attention as the entirety of CES the week before, if this year's keynote is not more Earth-shaking than the one last year it's a letdown.

Live by the spectacle, die by the spectacle.

To me, the two most interesting announcements were the MacBook Air and the new Apple TV and its associated services. Apple TV is strategic and needs a longer blog post than I have time for tonight. But I'd like to make a quick comment on the Air.


The next PowerBook, or the next PowerBook Duo?

I'm trying to reserve judgment on the Air until I can see one in person. On paper, it makes some uncomfortable compromises. No removable battery, no optical drive...it gives me nasty flashbacks of the PowerBook Duo. Like Air, the Duo was very thin and lightweight for its time, and like the Air it compromised on a lot of features. The Duo had a pretty elegant docking station that allowed you to use it as a full computer at your desk, then take the portable part with you when you traveled. I was working at Apple at the time, and I thought the whole concept was pretty clever.

It didn't sell well.

Turned out most people wanted to take the whole computer with them, not just part of the computer. They traded up to a heavier device with full features.

I worry that they might make the same decision about Air.

On the other hand, I just last week I wrote a post lavishly praising the new iPod Nano because its thin, elegant design more than compensates for its somewhat limited feature set. A lot of people criticized the Nano when it was first announced, in part because the photos couldn't do justice to its elegant design. When you saw it in person, it all made sense.

Maybe it will be the same for Air. I want to see and touch one. Maybe the lust factor will overcome the feature shortcomings. Or maybe personal computers are judged differently from music players. We'll find out.

Sunday 6 January 2008

Mobile Device of the Year, 2007

It's very difficult to say what's the best mobile device in a given year, because different people have different needs and desires. The ideal device for me might be repulsive to you, and vice-versa. But most of the computer publications try to make a call anyway. If you read the end-of-year reviews online, you'll probably conclude that the best mobile product of the year was the iPhone. It was cited by the Washington Post, Wired, Business Week, and Tech Republic (which strangely listed it as a business technology product, alongside Salesforce.com and LinkedIn).

Other mobile products getting mentions from major publications included the Nokia n95, iPod Touch, Razr 2, and Blackberry 8800. Amazon's Kindle was the only one that showed up on both best-of and worst-of lists. The best-ofs generally liked the wireless features and screen, while the worst-ofs disliked the closed business model and "eye-poking" industrial design.

I don't agree with any of those choices.

Since people have different needs, I think the best product of the year ought to be the one that best meets needs the needs of a particular group of users. It should be utterly compelling to its own audience. There are several questions to ask:

How efficient is it? Since people use mobile devices on the go, it should do just what the user needs, without any confusion or unneeded features. But there can't be any critical features missing, either.

How well does it trade off size vs. power? Because it's carried on your person, where size and weight are at a premium, it should balance tiny size with reasonable battery life.

How does it look? Because it's effectively a part of your wardrobe, it must look great (or whatever the target customer thinks of as great).

By that standard, I think the best mobile device of 2007 -- in fact, one of the best mobile products of all time -- was the third generation iPod Nano.

Don't get me wrong, iPhone fans. The iPhone is a very interesting and provocative device. There are some beautiful features in the user interface, and I love the turmoil it's causing in the industry. Several years from now we may look back on it and call it the most influential mobile device of its time. But that doesn't mean it's the best product.

To me, the iPhone is more an intriguing statement of direction than a completed product at this point. The lack of 3G is a huge compromise, and Apple obviously didn't think through the third party application thing. If you want a slow mobile browser that also plays music and videos and doubles as a somewhat awkward phone, then the iPhone is great. But for all of the cool highlights in the iPhone, I don't think it's enough to crush the phone industry in its current version. Future versions, maybe. We'll shortlist the iPhone III for product of the year in 2010.

The n95 is also a remarkable product in its own way, and I know it inspires a lot of technolust, especially in Europe. But in my opinion, it's just the latest Swiss Army Knife of the mobile world. Next year there will be another one from Nokia or Samsung or somebody else that has an even higher-resolution camera or maybe an electric toothpick or something, and people will be fawning all over that one. Like a lot of Japanese consumer electronics products, it's not a marvelous product as much as it is a marvelously ingenious bag of features.

By contrast, in third generation Nano is not just the latest model from Apple, it's an elegant culmination of the design work they've been doing for years.

The Nano doesn't look all that great in photographs. It's wider than its predecessor, which produced some criticism when it was announced (Engadget nicknamed it "fatty," which is asinine when you see it in person). In real life, the Nano's shape is compelling. It's much thinner than you'd expect from the pictures -- shockingly thin for something that has a color screen and plays videos. With its heavily rounded corners and brightly colored case, it feels a bit like a high tech chocolate wafer. You're almost tempted to take a bite out of it.





Physically, the Nano is almost all user interface -- the screen and thumbwheel take up the entire front of the device. Until we get flexible screens, the Nano is about as small as you can possibly make a device with its features. This is the endpoint, a form factor that's going to be with us for a while.

The biggest surprise to me about the Nano is the usability of video on it. When it was announced, I thought video was a throwaway feature -- who would ever want to watch video on a screen that small? But the reality is that when you're sitting down, you'll hold a Nano about 18 inches (45 cm) away from your face. At that distance, the screen is about the same apparent size as a 20-inch television (50 cm) at the other side of the living room. It's not like watching a flat panel monster screen, but it's very usable.

I'm not sure yet how much video will be used on the device, or what sorts of video, but that's a general question about mobile video rather than anything specific about the Nano. What I've observed so far is teenage girls using the Nano to watch music videos together, commenting on how cute the drummer is.

And that's just another sign that Apple made a great design for its target audience.

The new Nano doesn't have Bluetooth built into it, or Wi-Fi, or a camera, or a phone, or a hard drive. That probably accounts for why the technophiles online have been so dismissive of it (link). But to me, it's an almost perfect balance of functionality and art. Come back in ten or twenty years and I think you'll find it in design museums, when most of today's mobile devices will be long-forgotten and mildly embarrassing.

What do you think? Do you agree with my choice? If not, what do you think was the best mobile device of 2007?