My mind is filled with little green robots… who are they, and where are they going?

This week, I attended O’Reilly Media’s debut Android Open conference, a “big tent” gathering that brought together a wide swath of the open-source ecosystem—from game developers to marketers, from chip manufacturers to venture capitalists. It was an interesting and enjoyable conference, a good place to think about launching an Android-centric business.

Coming up soon, of course, is our own AnDevCon II, which is a larger but more tightly focused conference dedicated to technical classes and workshops on Android development. (Gratuitous plug: It’s Nov. 6–9 in the Bay Area. See www.andevcon.com.)

But back to Android Open. Talking to speakers and attendees, there was a universal concern about exactly what Android is, and what it’s becoming under Google’s direction.

On one hand, it’s pretty clear: Android is an open-source operating system for phones and tablets. Handset makers get a consistent firmware and user experience, while giving them room to innovate and differentiate. Carriers get predictability and a smartphone that they can sell, customize and monetize. Developers get a big enough platform to sell their wares, and more flexibility than they get from Apple.

Google benefits from its investment in several ways, including:

• Commissions on apps in the Android Market.

• Opportunities to bring customers into its cloud ecosystem, such as with Gmail and Google Apps.

• And lots and lots and lots of places for advertising.

Android, however, is widely viewed as a whole lot more than a phone and tablet platform, and as more than an app and ad delivery vehicle. For example, it is viewed as a promising embedded platform. However, because Google can’t benefit directly from embedded development (no apps, no Gmail, no advertising), the company’s position and support are ambiguous at best.

Even within the handset and tablet spaces, the Android user experience and app-store experiences are fragmenting fast. I have two Android handsets: an HTC Evo and a Motorola Atrix. The user experiences are quite different, and on some handsets they even vary from carrier to carrier. Don’t expect an Amazon Kindle Fire to work like a Barnes & Noble Nook Color, and don’t expect either to be like a Samsung Galaxy Tab or a Motorola Xoom.

The fragmentation in app stores is even worse. I can’t buy apps from the Amazon Appstore for Android on my AT&T-hosted Atrix. My understanding is that the Kindle Fire will only run apps that Amazon approves and offers. What’s more, app stores are proliferating like crazy. This doesn’t serve the market!

If Apple has locked the iOS walled garden down too tight, then Google is too loosey-goosey. Device makers (and carriers) have so much room to differentiate that we’re fast losing cohesion.

Remember the days of Windows 95 and Windows NT 4.0, when Microsoft insisted upon a uniform experience and a consistent desktop? While Compaq, Dell, HP and others wanted freedom to differentiate, Microsoft understood that a Windows PC had to be a Windows PC, and it enforced those rules with an iron first and a strong logo program. Of course, once the consumer or enterprise purchased the Windows PC, they could do whatever they wanted, but at least out of the box, Windows was Windows.

So what exactly is Android? Officially, it’s an operating system for handsets and tablets, yes. It’s a set of APIs. But it’s not a defined or consistent user experience for your handset or tablet. It’s not a well-defined embedded platform. It’s not a way to find, purchase and install apps for your device. It’s not a uniform way to sell or distribute apps to your customers or end users.

Android devices are selling like hotcakes—more than a million units activated per month. But without a clear identity, I’m not sure that this is enough.

Alan Zeichick is editorial director of SD Times. Read his blog at ztrek.blogspot.com.