Desktop search: who cares?

• By Mike Gunderloy
• December 15, 2004

You almost certainly saw the announcement, or at least the buzz about it: Microsoft has released the MSN Toolbar Suite Beta With Desktop Search. This competes, of course, with Google Desktop Search. And now Yahoo! has announced they'll be entering the fray with technology licensed from X1. There are other competitors, like Copernic already in this market, too. All in all, there's been enough buzz over desktop search lately to account for a whole swarm of bees. But why?

All of these utilities search your local hard drive for information. They index varying things, but typically it's file contents plus e-mail messages. Most also integrate Web searching, but frankly, I think that's just a meaningless checklist point; everyone I know already has a way to search the Web. The real battle appears to be over who will help you search your own computer. As far as I can tell, there are three reasons why we're seeing a spate of excitement about software in this arena:

• The built-in file search in Windows, frankly, stinks. It arbitrarily skips indexing some files (try searching for content inside of C# code, for example) without telling you, and it's slow as a snail. And if Windows search is lousy, Outlook's "Advanced Find" is, well, the sort of poorly-implemented feature that I personally would be ashamed to ship in software with my name on it.
• People who have been working in the computer field for a long time, particularly at e-mail driven companies like Microsoft, have piled up so many files and messages that their filing systems are breaking down and they have to search for things. So developers are building the tools that they need themselves. This is fine, if you happen to want to target "developers who've been working with the same company for ten years" as a market. But there's a great danger to building the features your team needs and sliding into the assumption that every man, woman, child, and dog on the planet needs those same features. Consider, for example, the intranet features of Office, which have been a great boon to the Office team but which have had little impact in the real world.
• There's a certain amount of chest-thumping, teeth-baring, mindless competition among the major coporations in our industry. More and more Microsoft appears to be considering Google to be The Enemy, who must be Denied Mindspace by Any Means Necessary - or at least by putting out competitive software as quickly as possible.
• The press will write about any software that comes out with a press release and a free beta. Oh, wait, that's me.

None of those, I think, is a terribly good reason to believe that desktop searching software will really turn into the next big thing for your Aunt Mary. Now personally, I wouldn't be without my copy of X1. A quick look at its statistics shows that I've got 533,000 e-mail messages, 45,000 attachments, and 159,000 files indexed this morning. But then, I've got upwind of a terabyte of storage on my home/office network and I've been in this industry for too darned long. But if you're looking for a big consumer market, I'm not really very representative - and if you're reading this, you probably aren't either.

I think that the current round of search utilities will end up getting installed fairly widely, because they're flashy and they're being pushed by big names. But after that, I suspect most people won't actually use them for much. Developers will find them useful when the question is "where did I file that design document that contains the words 'refactor' and 'amplification'?" But Aunt Mary is more likely to ask "Where's that cute picture of Timmy and the twins in front of the gazebo?" and discover that all the indexing in the world won't help her find it until she learns about adding keywords to digital photos. Call me back when that happens, and perhaps I'll be more excited.