Everyone wants what Microsoft's got, namely control of the most widely used computing platform. Or, more accurately, everyone wants the billions and billions of dollars that flow in from the dominance of desktop computing.
Replacing Windows shouldn't be hard. Everyone hates Windows, right?
That seems to be Google's thinking. The company announced this week that its shiny new Chromebooks will become available to order online starting June 15 in the US, UK, France, Germany, the Netherlands, Italy and Spain.
Chromebooks are laptops powered by Google's Chromium, which is an open source, browser-based operating system. The laptops will be built by Samsung and Acer, and they will be priced at £399 if you want 3G capability, with Wi-Fi-only models available for £349.
You can also "rent" Chromebooks. Businesses will pay $28 per month and schools $20. Software updates are constant and automatic. Hardware replacement happens automatically with failures and new versions.
Unlike Chrome, which is Google's browser application, Chromium has a built-in Flash player, a PDF viewer, an automatic self-updater and other features.
Google claims that Chromebooks have multiple advantages over other computers, including USB storage, a limited file manager, offline use of apps and data, superfast boot times and 8.5 hours of battery life on a single charge. The company emphasises that no data is lost when a machine is damaged, lost or stolen.
Google co-founder Sergey Brin offered this analysis: "With Microsoft, and other operating system vendors, I think the complexity of managing your computer is really torturing users. It's torturing everyone in this room. It's a flawed model fundamentally."
It shouldn't be too hard for Chromebooks to compete with "torture."
It's true that Windows computing can be painful and is a flawed model. But there's one major problem with Brin's statement: His sales pitch exists in a theoretical fantasy world where there is no distinction between personal and business computing.
When you think through the implications for these two markets separately, you can see that Chromebooks are best for neither.
Why Chromebooks aren't best for consumers
The idea that cloud-based computing is all about user happiness strains credulity. The whole purpose of cloud computing is to protect organisations from their users. Chromebooks take away user freedom and control. Yet Google is pitching the concept as an attraction to consumers.
Corporations may initially like Google's cloud model because it fences users in and makes it impossible for users to break things. Schools might like Chromebooks because they will make it difficult for students to do things they're not supposed to do, such as download malicious code. But few consumers would choose limitations over freedom.
If consumers actually wanted browser-only computing, they would simply do browser-only computing. Nothing stops anyone from buying a 15-inch laptop, downloading the Chrome browser and then doing all of their computing tasks inside Chrome.
Nobody does that because it's unappealing. People like to install applications, and they will do so if they can.
People who want to replace their "flawed" Windows PCs have an alternative that's superior to Chromebooks: namely, the app model invented by Apple for the iOS, and used by Google Android, HP's TouchPad and RIM's BlackBerry PlayBook. App-based touch tablets solve the "torture" problem Brin highlighted, without the problems inherent in cloud-only computing.
And I'll just come right out and say it: Chromebooks are ugly. The hardware is ugly, and the web is ugly, for the most part.
Thanks to the iPad, consumers have come to expect devices that are aesthetically beautiful, graphically appealing and fun. The Chromebook promises only drab utilitarianism. App-based touch tablets will be more popular for consumers than Chromebooks for the same reason that American Idol is more popular than CNN.
Why Chromebooks aren't best for business
Yes, Windows PCs "torture" users. But so does the cloud. Browsers aren't perfectly secure and reliable, and neither are Internet connections or websites.
Personally, I use three browsers: Google Chrome, Mozilla Firefox and Microsoft Internet Explorer. I do this because each browser has its own limitations and problems.
I've been struggling with the most recent version of Firefox, it crashes and does other weird things. So I've been using Chrome more as an alternative. But Chrome doesn't have my needed Firefox plugins. And it doesn't support RSS. I use each for different tasks. But when I hit a brick wall with both, I use Internet Explorer as my browser of last resort.
Chromebooks wouldn't give you this choice. If Google releases a flawed update, you're stuck with it. If it breaks your plugins, too bad. If it doesn't support your favourite website, tough luck. Besides, if Chrome is so great, why doesn't Google use it on Android?
Internet connections can have problems, too. Both 3G and Wi-Fi have their own sets of challenges. Mobile broadband connections don't connect in many locations. Home Wi-Fi routers can stop working. Of course, these things also happen with regular PCs. And Google is working hard to enable offline browsing. But when my connection is down with a PC, at least I can still use all my applications and files.
Here's the most important thing: The Internet itself can't be trusted to handle 100% of our computing needs.
Google's own Blogger service went down for more than 24 hours this week. To restore service, Google rolled back to an older, backed-up version, which didn't include 30 hours of blog posts for Google's millions of users. As I was writing this column, Google was working to restore the lost posts.
Such disruptions happen all the time, even for cloud-based services that are supposed to be bulletproof. Amazon's EC2 website hosting service, which exists to provide fail-safe totally reliable hosting, experiences catastrophic outages. The most recent outage occurred in April. The glitch took down Foursquare, Reddit, Quora and other major services. It took Amazon four days to return service to normal.
Cloud computing is great, but only in combination with "regular" computing. The only reliable way to manage data is to store and back up locally, and also to the cloud.
The Chromebook model requires the user to have a fully functional machine, browser, connection and web services. Without each and every one of these elements working perfectly, a Chromebook is nothing but a tray for serving snacks. The big question is this: Do you trust Google to keep this programme going?
Google has gained a reputation lately of simply cancelling projects that aren't working out. The company has killed off Wave, Lively, Answers, Dodgeball, Video and other high visibility projects. People start getting excited about a platform, then one day Google makes an announcement, pulls the plug and that's it.
Even more relevant is the cancellation of Google's Nexus One phone. Google came up with a new way to sell phones and rolled it out in January, 2010. The Nexus One was available only online, and unlocked. Google assumed that users would be happy to provide tech support to each other, and so the company didn't set up any way to support users. They assumed wrong. The phone was taken off the market in July.
The question on the minds of businesses considering the Chromebook is this: Will Google end support?
Businesses are used to buying equipment from companies like HP, IBM and others that tend to support their products even if they're not runaway successes. And they provide migration plans if they do end product lines.
If a company is going to buy 300 Chromebooks and reorganise its IT department accordingly, it's going to need assurances that the platform will be supported for at least a decade. I haven't heard Google give that assurance.
The Chromebook's cloudy future
Google says one advantage of Chromebooks is that they don't have applications that need to be patched and updated. But that isn't true. Web-based apps get updated. The difference is that those updates happen without the knowledge, consent or control of either the user or the IT administrator.
This reminds me of the myth that cloud computing doesn't involve servers or hard drives. Of course it does. The Chromebook proposition is not the absence of software, patches, servers and hard drives. It's the removal of these things from your control. Who wants that?
The Chromebook idea sounds cool in theory. But in practice, a cloud-based laptop isn't best for consumers, and it's not best for business. The Chromebook will fail.