Following the triumphant Microsoft Professional Developers Conference (PDC) 2003, at which Microsoft Chairman and Chief Software Architect Bill Gates introduced the next major Windows version (code-named Longhorn) to an eager crowd of 7000 developers, last night Gates did what so many people have done after such an obvious success: He stumbled badly, delivering a sleepy and plodding keynote address to a stunned crowd at the COMDEX 2003 trade show in Las Vegas, Nevada. The speech was Gates's 20th COMDEX keynote address and a time for reflection. But for people who were hoping to hear a bit of Microsoft vision, the keynote was a disappointment, concentrating primarily on the progress Microsoft has made fixing the software problems the company is largely responsible for.
   When Gates took the stage 20 years ago for his first COMDEX keynote address, the world was a vastly different place. Microsoft Multiplan was the company's best-selling software at the time, but Microsoft itself was a small and little-known company. At that time, software products such as Lotus 1-2-3 and WordPerfect dominated the market but slowly retreated under Microsoft Office's crushing power during the next decade. Today, Microsoft is an industry goliath, and most of the companies it competed with in 1983 are long gone. Gates categorized the past two decades by the constraints under which Microsoft has operated: hardware in the 1980s, hardware connections in the 1990s, and software connections today. But I find it difficult not to wonder whether the company's biggest constraints aren't simply legal barriers and the fact that it rarely competes with anyone other than itself. For example, Gates touted the record $6.8 billion his company will spend on R&D during the next year--money Microsoft will use to solve problems such as disconnected data exchanges between companies. And although this question sounds a bit facetious, I have to ask: If it takes $6.8 billion to solve this problem and Microsoft has more than $50 billion in the bank, why can't the company double its R&D budget and solve the problem more quickly?
   So what future technology can we expect from Microsoft? Gates's keynote address concentrated heavily on technologies we've already seen, albeit with new names in some cases--names that Gates often got wrong. For example, a new antispam technology called SmartScreen debuted earlier this year in MSN 8, Hotmail, and Microsoft Office Outlook 2003. Gates said that SmartScreen will also ship in the next version of Microsoft Exchange Server, due in a few months, but the technology will actually ship in an add-on for the current Exchange version. He discussed the next Tablet PC OS, Windows XP Tablet PC Edition 2004, but declined to mention its name. (The good news: The new OS will be a free upgrade for all existing Tablet PC users.) He misidentified Office 2003 as Office XP. And so on.
   Aside from an interesting demonstration of Microsoft Internet Security and Acceleration (ISA) Server 2004, the only bright moment came during a demonstration from Microsoft Research Senior Researcher Susan Dumais, who showed off technology called Stuff I've Seen. The new technology aggregates content you've viewed through email, the Web, and local documents, making it easy to find information you've previously digested even when you can't remember the area in which you found it. Stuff I've Seen will debut in some form in Longhorn, due in 2005, and Dumais's demo was exactly the kind of exciting future technology we've come to expect from a Gates COMDEX keynote address. Sadly, it came 7 minutes before the speech ended.