It's official: Google Chrome, the operating system
Well, this answers the question about why no Android for netbooks. In a stunning announcement late Tuesday evening, the company that for years had been suspected of developing an operating system but which had never entirely denied the claim, has come out with it: Yes, Google is making a Linux; yes, it's for netbooks (at least for now); no, it's not Android.
"Google Chrome OS is a new project, separate from Android," reads yesterday's company blog post from VP Sundar Pichai and Engineering Director Linus Upton. "Android was designed from the beginning to work across a variety of devices from phones to set-top boxes to netbooks. Google Chrome OS is being created for people who spend most of their time on the Web, and is being designed to power computers ranging from small netbooks to full-size desktop systems. While there are areas where Google Chrome OS and Android overlap, we believe choice will drive innovation for the benefit of everyone, including Google."
At this point, what we know for certain is in that blog post, and it's not much: Chrome OS is being designed for not just x86 processors but also ARM processors, getting into the embedded devices field, or what ARM itself calls "smartbooks." Initial sales will be through OEMs who are expected to pre-install Google OS on their netbooks. With Acer's contentious history with Microsoft and its already having embraced Android for smartphones, that manufacturer stands perhaps the biggest chance of benefitting from this news. "Acer Aspire Two," anyone?
As a software platform, though, Google's play is bolder than just netbooks: Its aim is clearly to leverage the whole Chrome idea, including the existing Windows-based Web browser, as a foundation for a Web-based software platform to challenge both .NET and Java.
"For application developers, the Web is the platform. All Web-based applications will automatically work and new applications can be written using your favorite Web technologies," reads yesterday's post. "And of course, these apps will run not only on Google Chrome OS, but on any standards-based browser on Windows, Mac, and Linux thereby giving developers the largest user base of any platform."
This explains why Google has been strengthening Chrome's windowing capabilities at the expense of work on other features. It's developing a way for windowed applications that are more functional than Microsoft's, to run cross-platform inside Microsoft's OS, and still have the virtue of running on Mac OS and Google's new brand of Linux. This solves the whole issue of "base" for Chrome OS developers, which is analogous to "audience" for media developers -- it's the perceived group of people for whom software authors write. Historically, a software firm doesn't have much to gain if it writes for the base with the smallest niche, which has been Macintosh's problem since 1984, and Linux' perennial problem on the desktop for commercial software players.
But if developers think of the Web platform on a very high level -- the plateau that Sun/Oracle Java has always hoped to reach, and Adobe Flash (now incorporating Flex) has recently aspired to grasp -- then "reach" is conceivably everyone. Should Google extend its Chrome Web browser, for instance, to Linux distributions other than its own Chrome OS, extending its windowing environment to all the other open source players; and if the Chrome browser also gets its act together on Mac OS; then that is everyone. An app built "for Chrome" would conceivably run on any x86/x64 system in the world, plus a good chunk of ARM processor-based devices. Chrome OS' blanket would extend beyond the reach of Windows and even of Java.
And if Chrome OS runs on x86, after all, what is there to prevent an everyday system owner from installing it on a desktop or notebook instead of Windows?
ADDITION: If writing a Web application is essentially writing "for Chrome," then what makes Google's intention the creation of a "Chrome platform?" After all, Chrome OS opts to be lightweight, in order to keep its profile down for netbooks.
When you see what Google is cooking up with its Chrome browser, you get an understanding of the architectural power play it's working up: Think of a "good" and "better" scenario, where everyday Web apps will run in Chrome and run in Firefox and run in Windows (IE), and everyone's happy and the European Commission is placated and all is right with the world. That's "good."
But then there's "better," a world full of Google Gears and that Chrome OS windowing environment that last night's post mentioned. It's a world where the windowing environment is directly and safely accessible at a low level through remote procedure calls (which in Windows is extremely dangerous), and Web apps that are effectively "Chrome-aware," to coin a phrase, can do more things in more ways than they can when they just run "good." That's the platform play Google is making: Sure, your Web app supports Chrome by default (if you write it "correctly," which translated into English means, "not in Flash or Silverlight"). But to make it better, you add support for Chrome that turns on when Chrome is present. And all the user has to do to make it work is install Chrome.
And when will they do that? Conceivably right when the Web app, running on a non-Chrome OS application, places a call to a resource that Google is hosting. "I see you're not running Google Chrome," says the popup. "Would you like to install it now? It's free!"
It is an extraordinary market play, a gamble for the entire pie, not just a slice of it. And Google is the only player other than Microsoft at this point with both the moxie and the resources to pull it off.
Success for Google OS, however, in any conceivable scenario, would leave a wide path of carnage in its wake. Think of it: Google would uproot Microsoft's entire value proposition for Windows: that it has the biggest platform and the strongest base, so any investment in Windows is a secure one. Google would perhaps indirectly, but certainly inevitably, challenge Apple's position as the turnkey applications distribution channel for iPhone (does anyone think that compiling a Chrome platform attachment for iPhone, if Chrome runs on Mac OS already, would be impossible?). Chrome OS would capture the latent smartbooks market that Microsoft has publicly dismissed as irrelevant, which might conceivably lower the final curtain on Windows Mobile as an embedded platform (if you're an embedded software developer, wouldn't you rather write for the platform that also works on x86?). As an applications distribution platform, Chrome OS would seek to capture the prize that Java originally made relevant, and that Microsoft's .NET sought to "embrace and extend." And as a Web browser (if anyone's still thinking that small), Chrome would leave Internet Explorer and Mozilla Firefox in smoking ruins after a very bloody battle.
That's assuming everything works according to Google's plan, and that's assuming Google really has a plan. As last night's blog post pointed out, Google would only provide more news about Chrome OS "in the fall," which suggests that there's nothing much for it to share with its esteemed community of devoted developers, and may not be much real code to present to them until deep into 2010.
But if Google really does have a plan (and it might), then capturing the king, the flag, the castle, and the whole kingdom -- which is genuinely the intention that Google signaled yesterday -- would require every other player in this industry to sit back and gingerly let it happen. I don't mean just Microsoft, but also Apple, Adobe, Oracle, and Mozilla. There are some who are uncertain just how much fight these old players have in them. But to assume that they will cede their respective claims to the software market without a serious counter-offensive, is to ignore history. If Google does that, then its game is already lost.
Yet it is a game, for the first time in a quarter-century.