5 min read

Dialing For Dollars

An awful lot of people -- including yours truly -- assume that when the dust settles in the smartphone market, Linux will either dominate its competitors or at least claim a handsome share of the spoils. Are we dealing in reality here, or is Linux strolling into a war zone armed with little more than its own delusions of grandeur?
About a month ago, we ran an eye-opening feature covering the next great battleground in the mobile industry: the smartphone market.

I learned several important things from Dave Haskin's article. First, I discovered that smartphones aren't yet able to dispatch killer robots capable of hunting down telemarketers like rabid dogs -- although they will bring together many of the features spread today across mobile phones, PDAs, Tablet PCs and laptops, all in a single powerful, highly convenient package.

Second, I learned that the smartphone market is still a wide open space where it's impossible to pick clear winners or losers -- an incredible situation in a market that has been growing like a weed and is likely someday to confer the kind of wealth that will make today's mobile moguls look like a bunch of dirt-poor hillbillies.

Third, I realize now that an awful lot of people -- including yours truly -- assume that when the dust settles, Linux will either dominate the smartphone market or at least claim a handsome share of the spoils.

Over the past several weeks, we've seen plenty of news dealing with this fast-growing, viciously competitive new market. As a result, I've decided that it's time to revisit my assumption -- or perhaps I should say, our assumptions -- that Linux is destined to win the lion's share of the smartphone spoils.

Are we dealing in reality here, or is Linux strolling into a war zone armed with little more than its own delusions of grandeur?

When I tackle this topic, I start with one key assumption: Smartphone technology isn't about new features and services -- it's about putting the existing services in a smaller, more convenient, better integrated package.. We've already added a staggering number of new technologies to our lives over the past decade, and frankly, I think even the most tech-savvy mobile gearheads are ready to take a breather and to focus for a while on quality rather than quantity.

So, instead of peddling their wares with the help of telemarketer-hunting hit 'droids, the eventual victor in the Smartphone Wars will have to focus on the basics: power, performance, reliability, and usability. It'll be no small feat, given that within a few years, the typical smartphone will include a feature set combining those available on two or three different devices today.

What will it take to power a platform capable of getting the job done? In a word, efficiency: These devices will have to incorporate kernels capable of managing complex, high-bandwidth network operations, all without wasting scarce CPU cycles or system memory.

They'll also have to be as good as bulletproof: System software that emits gurgling noises and spits up teeny-tiny little Blue Screens of Death when you try to call Mom and surf the Web at the same time need not apply for smartphone duty.

And of course, if you're a smartphone platform vendor, your handset-manufacturing partners will have a few not-so-modest demands of their own. They'll expect a software platform that enables short development cycles, low production costs, a superb user interface model -- oh, and and the ability to ape whatever the competition is doing, exactly five minutes after they do it, without charging a penny more.

I'm missing a few things, but you get the point.

Given all of these demands, is Linux the perfect smartphone platform? Not a chance.

In fact, Linux is a long way from delivering an ideal smartphone solution. Instead of an innovative user interface, for example, all Linux can offer smartphone vendors is a good, hard beating with the ugly stick of their choice.

Fortunately, "perfect" isn't the goal -- "better than everyone else, and constantly improving" is the goal. What Linux lacks in UI pizzaz, it makes up with rock-solid technical chops and amazing adaptability -- two huge advantages in a market likely to feature tiny margins, appalling customer churn rates, and zero tolerance for time-to-market screw-ups.

Yet Linux has another handicap to overcome if it hopes to win the smartphone wars: a distinct lack of here-and-now market share. To overcome this disadvantage, Linux will need a to build a constructive, if not always cooperative, relationship with the current mobile platform leader: Symbian. That's no easy task, since Symbian currently runs half of the world's handsets, and its biggest customers -- Nokia, Samsung, and Sony Ericsson -- are also its majority-owners.

Even without taking on Symbian head-to-head, however, I think Linux can still build a solid base -- if today's press releases turn into tomorrow's production decisions.NTT DoCoMo, Japan's biggest wireless operator, has committed to using both Symbian and Linux -- but not Microsoft. And PalmSource is in the midst of an effort to rebuild the Palm OS interface layer on top of a customized Linux kernel it acquired along with its purchase of China MobileSoft last year.

If both of these branches bear fruit, Linux will have a solid start in the smartphone market. Better yet, it will have a superb start in the Asian smartphone market -- the same market where I'm betting Microsoft is headed for disaster at the hands of open-source competitors and indifferent consumers.

So there's my line of reasoning -- and based on the results of our recent poll asking which mobile player you saw winning the smartphone market, many of you must have a similar scenario in mind. Or do you? Pass along your thoughts on how you see Linux playing the smartphone game, and we'll see how else we can spin what is rapidly becoming one of my favorite Linux topics.

Matt McKenzie is editor of Linux Pipeline. A permanent link to this article is available here.