Memo to Google’s Android team: Tie your shoelaces, kids

Is Google the new Microsoft? Maybe — but that’s the wrong question. More to the point is this: Can Google avoid tripping over its own feet as it strives to make Android a serious mobile platform? As my trusty Magic Eight Ball used to say, “All signs point to no.”

OK, that’s not exactly what the Magic Eight Ball used to say, but it’s close enough. Google doesn’t know what to do with the Android platform: Should it be open, or should it be controlled by Google? Either alternative could work, but the halfway solution we’ve seen so far is a setup for failure.

“Google wants it both ways,” says Michael Gartenberg, a partner in the Altimeter Group. “They want the perception of being open and the idea of being able to support everyone, but they also want their own vision for the hardware and the platform,” he tells me. As a result, device makers and developers are confused, customers are disappointed, and — worst of all — the Android platform is fragmenting.

Things in Android land are so bad, you have to wonder if the only company that has a chance to give Apple a run for its mobile money is, believe it or not, Microsoft and its forthcoming Windows Phone 7.

Google’s Android is fragmented as Google stumbles about Google’s modus operandi is to whip out software at a lightning pace, then fix the bugs and add the missing features later. While that strategy has succeeded extraordinarily well on the Web, it doesn’t work at all for consumer hardware and the software that runs on it. Google is so clueless about consumer electronics that it launched the Nexus One without a plan to provide customer support. What’s more, the mobile world works on a pace that makes Moore’s Law seem snail-like.

Device makers can’t move as fast as Google’s software developers. It takes time to optimize hardware and the software that runs on it. Once that task is complete, the vendor isn’t going to throw out the work before getting some sort of payoff. One example is the Motorola Backflip, which runs version 1.1 of Android, notes Gartenberg, while other smartphones run version 2.1. Still others run version 1.6 or 2.0. That would be like going to the store and seeing PCs for sale that run Windows 98 next to Vista and Windows 7 machines.

Among device makers, “there’s a growing frustration of licensees that Google is shipping latest and greatest while they stuck with old,” says Gartenberg.

Not only do vendors have to rush to keep up with the platform, they have to find ways to differentiate their Android products to avoid commoditization. And because the Android platform itself isn’t finished (there are few security or multitouch capabilities, for example), the device makers figure they have to do the last 10 or 20 percent themselves, leading to uneven quality and devices that support some apps but not others. Android, says Gartenberg, is following the path of desktop Linux, which has never taken hold because it fragmented “into different versions with different core feature sets, different users experiences, and that run different applications.”

For example, Google recently touted the latest mobile version of Google Earth, a cool app you can’t use unless you’re running Android OS 2.1. On the other hand, “I recently tried to install one of the few good Android games and found it won’t work on Nexus One as it has a nonstandard screen resolution,” Gartenberg wrote in his column on Engadget.

Meanwhile, Google acted very Apple-ish (that is, like a control freak) when it told developers of an Android tablet that they couldn’t place the Apps Marketplace on it because it isn’t a phone, Gartenberg says. Google’s inability to choose between an open platform and a closed one is an example of corporate cognitive dissidence, a condition antithetical to business success.

A more adult choice: Windows Phone 7 breaks with the past Even Steve Ballmer admits that Windows Mobile was a turkey from the get-go. So when Microsoft started talking about a new mobile OS, lots of us rolled our eyes. It’s still very early days, but to quote the old Magic Eight Ball again, “Signs point to yes,” as in yes, it could well be interesting.

Microsoft’s willingness to throw out a lot of old code and get a fresh start is a huge and significant change, as my colleague Galen Gruman wrote after a meeting with the company’s Windows Phone 7 group last week. Until recently, Microsoft’s mobile OS development efforts were too much like its desktop operating system development process. It carried forward a huge code base to ensure backward compatibility.

We could argue about whether that’s a good idea on the desktop, but there’s no argument that much of the Windows Mobile code base needs a stake in the heart. Indeed, backward compatibility with a stable of mundane apps used by few consumers isn’t much of an upside when the trade-off is code bloat and kludgyness. So when the Microsofties told Gruman that the new mobile OS will break Windows Mobile apps that can’t be rewritten, he saw it as big plus, not a minus — and I agree.

It’s also interesting — and rather ironic — that Microsoft is going to exert much more control over the hardware than it ever has on either the desktop or the mobile side of the business. That’s what Apple has always done. And it’s exactly what Google is not doing.

Ultimately, Google will have to choose a closed or an open model. As much as I respect the open source model, I don’t think it translates well in the smartphone world. Google needs to grow up, set real standards, and tie its shoelaces.

I welcome your comments, tips, and suggestions. Post them here so all our readers can share them, or reach me at bill.snyder@sbcglobal.net.

This article, “Memo to Google’s Android team: Tie your shoelaces, kids,” was originally published at InfoWorld.com. Read more of Bill Snyder’s Tech’s Bottom Line blog at InfoWorld.com.