Steve Ballmer laughs off Google's Chrome OS threat
In a speech followed by a question session in front of the Microsoft Worldwide Partner Conference audience, Ballmer said, "to me the Chrome OS thing is highly interesting," but remarked that "it won't happen for a year and a half and they already announced an operating system," speaking of Google's Android.
Android, first officially announced four months after the original iPhone went on sale, is a conventional smartphone operating system based on the Linux kernel and using a Java Virtual Machine-like runtime for building applications. The company has floated plans to enable hardware makers to build not just Android smartphones, but also Android-based netbooks and other devices.
Google's recently announced Chrome OS also targets the netbook market. However, rather than being a general purpose, conventional desktop operating system, Chrome OS appears to simply be Google's Chrome browser running with the minimum software below it, which is also based on the Linux kernel.
This would restrict devices using the new system to running HTML 5 applications, similar to the WebOS Palm developed for the Palm Pre, but with nothing tying applications to the platform itself; in other words, the Chrome OS wouldn't run special "Chrome web applets" like the Pre but full blown web apps that can also run in any other standards-compliant browser, from Safari to Firefox to the iPhone.
Two client operating systems?
Taking aim at Google's new strategy, Ballmer told the crowd, "I don't know if they can't make up their mind or what the problem is over there, but the last time I checked, you don't need two client operating systems. It's good to have one."
Ballmer may have intended to be speaking from experience about Microsoft's travails in trying to push its DOS-based Windows 95/98/Me users to its NT-based operating system, a struggle that began in 1993 when the company debuted Windows NT 3.1 with a version number that made it the heir apparent to 1990's Windows 3.0 and an early precursor of Bill Gates' promises of an advanced, object-oriented post-NT Cairo operating system that could rival Steve Jobs' futuristic NeXTSTEP.
Instead, Windows NT 3.1 ended up only suitable for use on servers and high end workstations, leaving Microsoft to continue work on the interim, DOS-based Windows 3.1 and 3.11, released at the end of 1993. Two years later, NT still wasn't ready for the mass market, resulting in the DOS-based Windows 95. Another two years later, Windows 98 appeared, followed by 98 SE and then Windows Millennium Edition, the DOS-based doppleganger to the NT-based Windows 2000 which was still only recommended for business users and servers.
It wasn't until 2001 that Microsoft finally released a client PC operating system based on its NT kernel to the mainstream consumer audience: Windows XP. However, the eight years of struggle toward NT seems like a cakewalk compared to the last eight years of Microsoft's attempts to move Windows XP users toward a more modern operating system with advanced graphics compositing and well designed, object oriented frameworks more like Mac OS X. Microsoft is still struggling to promote adoption of Windows Vista and Windows 7 in a PC world dominated by users who prefer Windows XP.
Microsoft is still selling dual operating systems
Despite all that history behind the company, Ballmer's more curious logical leap stems from the fact that Microsoft's smartphone operating system most comparable to Android, called Windows Mobile, is based upon Windows CE, which uses an entirely different kernel than the company's NT-based desktop operating systems. The only direct commonality between the two is that both use the same brand name.
If Ballmer was ignoring the kernel (both Android and Chrome OS will use the same Linux kernel under the surface) and directing his "two client operating system" attack upon the fact that Android and Chrome OS will be different development platforms, then another fact floats to the surface: the Windows Mobile platform is not at all compatible with the Windows XP/Vista/7 desktop platform. Most Windows Mobile software is still written in Win32, whereas Microsoft is working to push desktop and web developers to .NET (it's also hopeful of moving Windows Mobile developers there, but there's less new interest); these two development platforms are somewhat comparable to Apple's classic Mac OS APIs and the company's more modern Cocoa APIs used in Mac OS X.
However Ballmer would like to slice it, there's no way to say that his company has a cohesive platform that bridges mobile devices and the PC desktop. Instead, the company's powerful desktop operating system monopoly has done next to nothing to advance Windows Mobile over the last ten years.
Instead, Microsoft has lost its share of the smartphone market, from a high of 24% in 2004 to today's 12% (according to Q4 2008 figures published by Gartner). RIM has a 19% share and Apple took a nearly 11% share in just a year and a half of trying. Android has yet to register as a significant blip in the smartphone market, but its launch was targeted directly at neutralizing Windows Mobile. Rather than building an integrated phone like the BlackBerry, Palm Pre, or iPhone, Google set out to counter Microsoft's general purpose operating system that any manufacturer could license.
Google takes on Microsoft in the OS business
Without waiting for Android to take off as a Windows Mobile substitute, Google has announced Chrome OS as a new, free offering for netbook manufacturers that will enable them to sell their hardware without a Windows license, just like Android in the smartphone market.
When asked if Microsoft needed to follow the same strategy, Ballmer answered, "We don't need a new operating system. What we do need to do is to continue to evolve Windows, Windows Applications, IE [Internet Explorer], the way IE works in totality with Windows, and how we build applications like Office."
Part of that strategy involves floating a web-based version of Office to defend Microsoft's productivity suite from encroachment by Google's web-based Docs. Google is also hitting at IE with its Chrome browser, in tandem with Mozilla's Firefox and Apple's Safari, all three of which are working to support HTML 5, a standard Microsoft is currently betting against in its efforts to resist further erosion of the company's control over the web with IE and the company's Flash-like Silverlight plugin that replaces open web documents with a proprietary, closed binary.
First they ignore you, then they laugh at you
However, Microsoft's active IE browser share is shrinking dramatically on the desktop, from near totality just a few years ago to something closer to 60% despite being bundled with every new Windows PC sold worldwide. That makes Microsoft's ability to hold up progress on new interoperable standards much more difficult.
Among mobile devices, Pocket IE's flouting of web standards is simply irrelevant, with HTML 5 proponent Apple now owning more than half of all mobile web traffic in Mobile Safari, and many other smartphone platforms, including Android, BlackBerry, and Palm also adopting a WebKit-based browser.
Google's Chrome OS will leverage its WebKit commonality with these mobile devices to expand the number of platforms reachable by HTML 5, making it a common denominator in reaching a wide variety of clients. The history of computing shows that usable common denominators have a powerful effect in marginalizing smaller alternative platforms.
The widespread use of IBM PCs enabled DOS to eradicate a variety of alternatives for booting desktop PCs. Windows killed off the market for alternative graphical systems due to its ubiquitous use on desktop PCs. The web killed off proprietary dial-up services despite Microsoft's attempts to leverage its monopoly to promote MSN. The iPod's use of MPEG MP3/AAC standards killed off Microsoft's ability to push its own Windows Media Audio format.
Microsoft's other anti-mainstream positions have also failed to gain traction, from the AAF media container to its Windows Media Video codecs to its HD-DVD format. Eight years of technologies associated with Vista have also failed to set the pace in market that Microsoft itself monopolizes, making it evident that the dual-platform crisis that Ballmer wishes upon Google is really more of a problem for Microsoft itself.
Then they fight you, then you win
There's also no shortage of irony in Ballmer's laughing off a rival product announcement after the embarrassment Ballmer suffered in claiming that Apple would never amount to more than 2-3% of the market with the iPhone, and that the new device was the most expensive phone in the world despite the fact Microsoft's primary partner was already selling more expensive Windows Mobile phones than the iPhone.
Ballmer's company is now scrambling to catch up with the success of last year's iPhone 2.0 App Store, promising a mobile store of its own toward the end of the year. The problem for Microsoft is that its store will only support Windows Mobile 6 or later devices, which already represents a smaller installed base than the iPhone and iPod touch.
Further, Microsoft's efforts to peddle software to Windows Mobile users will be stymied by rival efforts by manufacturers and mobile carriers. LG is already trying to market software to its own phone users, some of which are based on Windows Mobile, and Verizon Wireless has insisted that it will attempt to control sales of mobile software to its subscribers. That slashes in half the readily reachable market Microsoft has available to it in the US, which is by far the largest market for Windows Mobile phones.
The move to mobile
Ballmer no longer laughs at the iPhone. Next year, he may also need to reevaluate his impression of the threat Chrome OS will have in pushing an HTML 5 platform of web apps on low end netbooks. Earlier this year, Microsoft's position was that Windows-based netbooks only needed to run three apps concurrently because they're only used for basic text entry, browsing, and other simple functions.
If users can get what they want from a web-based netbook without Windows, the desktop market for low end Windows PCs will only continue to implode as cheap netbooks replace low end Windows PCs with open source appliances capable of running anyone's web applications.
That's good news for Apple's iPhone and iPod touch, which are also HTML 5 savvy, as well as the emerging crop of WebKit-based smartphones with screens large enough to interact with web applications. But its the biggest challenge yet for Microsoft's heavyweight desktop PC operating system, which is sold in large volumes across low end hardware. Losing any significant portion of its existing Windows PC base will be disastrous for Microsoft's monopoly business model.
That, in turn, will force Microsoft to choose whether to release a simple web-based operating system of its own for light duty, low end machines in a battle with Google's free Chrome OS offering, or to attempt to upscale the Windows PC back into the high end consumer/prosumer market Apple has established for the Macintosh, something that HP, Dell, and Acer have been unable to do as the bottom drops out of the conventional PC market. Or attempt to take on a war with two fronts.
Microsoft could also take on Apple in delivering its own higher-end mobile device, although the company is late to the gate. Sources within Microsoft report that the launch of the iPhone 3GS blew Microsoft's own internal plans out of the water. "I read the specs on the iPhone 3G S," one Microsoft developer recent wrote, "and realized it has already surpassed the thing we're trying to do that won't ship for a large number of months."
Perhaps laughter is just how Ballmer shows he's nervous.