It seems that Microsoft opened its fifth annual MSN Strategic Account Summit recently. The shindig included tons of advertising brass and all sorts of important folks from the Internet end of things, like Terry Semel (chairman and CEO of Yahoo!), and Greg Stuart (president and CEO of the Interactive Advertising Bureau). Yours truly didn’t receive an engraved invitation, alas; it would have been fun, because, to all accounts, it was a great chance to see Microsft’s CEO Steve Ballmer actually admit that his company made a mistake.
Don’t start cheering. The mistake was in outsourcing Web search technology rather than creating it in-house, and Ballmer intends to play catch-up with a vengeance. Within 12 months the company will unveil its own Internet search technology. Microsoft can do it, too; they’ve got billions of dollars invested in research and development.
So what is this new technology going to be like? Well, if it was just going to duplicate what’s already out there, it’d be a waste of time for Microsoft to do it. Granted, MSN gets enough visitors on its own that maybe it only needs to be just as good, not better — but let’s face it, searching the Web really needs improvement. Even Google, the search engine that’s tops in my book, can hardly index more than one percent of what’s out there. It isn’t just that there’s so much out there; it’s that a lot of what’s out there is stored in databases that require users to register — which can range anywhere in price from free to thousands of dollars — and are not open to searching by the general public. So you may not even know that the information you need actually exists somewhere, but you just can’t get to it. In cases where you do get lots of answers to your query, many of them are redundant. It’s feast or famine, and sometimes both at once.
So Microsoft is starting smaller and somewhat specialized, with something called Newsbot. Reportedly in beta in several countries (but not the US — at least, not yet), the program depends on the principles of artificial intelligence and information retrieval. It keeps track of what you’ve already seen, so it doesn’t show you the same stuff all over again. It’s also supposed to get a feel for the stuff you like to read about, so it can make intelligent suggestions for what news to read. Eventually, the company wants to make it possible for users to ask natural language questions — like, say, which companies’ stock went up by more that 15 points today — and get real answers, not garbage.
Microsoft isn’t stopping with the news, though. Something called Blogbot, designed to aggregate content from Web logs, is reportedly in alpha. And — quite honestly, without rancor — I wish them the best of luck. I know for a fact that there’s all sorts of cool information out there in blogs, but I never have the time to investigate it myself. If anybody can come up with a way to let people get at that information easily, reliably, and fairly painlessly, they’d be doing humanity a great service (or at least, the portion of humanity that’s either directly or indirectly online).
Here is where what I said earlier about the non-settlement with the European Union comes in. For those who may not be familiar with that bit of news, seems the European Commission spent five years investigating Microsoft and found it guilty of unfair competition — to wit, taking advantage of its monopoly on the desktop to push its own workgroup servers and multimedia player software, while crowding would-be competitors out of those markets. Seems Microsoft bundled that software in with the Windows operating system, for free — just like it did with its browser. At any rate, last minute attempts to negotiate a settlement did not succeed, so the EC slapped Microsoft with fines and sanctions, and Microsoft is swearing it will appeal the case.
Why did Microsoft and the European Union fail to negotiate a settlement? We’re seeing the why right here. To quote from Commissioner Mario Monti’s March 18th statement about the matter, “We made substantial progress toward resolving the problems which have arisen in the past but we were unable to agree on commitments for future conduct.” In short, it’s the bundling behavior that’s the issue — and what do you want to bet that Microsoft’s all set to bundle its search technology into its Longhorn release of Windows?
In a way, this might not be an entirely bad thing. A search engine on your desktop that, for example, searches both your computer and the Internet for information simultaneously — provided it could filter things properly — could be very useful. Ironically, it would bring us closer to Microsoft rival Sun’s slogan that “The network is the computer!” Of course, if this is how Microsoft does it, it’s easy to see where that would land companies like Google and other Internet search firms. Who’d go online to launch a search if they can use a search engine on their desktop to search everywhere? (And don’t we all wish that could work in meatspace?)
So, will Microsoft get away with it when it happens? Possibly. Even if some company brings charges against them for it, and they’re found guilty (as they were in the case of browser-bundling), the appeals can drag out forever, killing the competition in just a matter of time. Additionally, in this hypothetical antitrust case, Microsoft might not be found guilty; a search function isn’t like a browser or a multimedia player, which is more clearly middleware. Searching is searching, whether you do it on your hard drive, online, or through the three-dimensional clutter in your three-dimensional office. I really hope I’m wrong about this; I’d hate to see Google disappear, and of course a lot will depend on what, exactly, Microsoft manages to come up with. But it might yet win this round.