In Search of Better Search

Opinion: Despite King Google, frustration still reigns, but there are signs of a better future.

Everybody loves to search. I know I do my fair share of Web searching every day.

And Im not alone. Search is, after all, the No. 2 activity on the Internet, according to a recent study by the Pew Internet & American Life Project.

Web searching was just behind e-mailing on the list of top Internet activities. (Reading the news was third.) How did I find this study? By using a search engine, of course.

But is using a search engine really an activity? Isnt saying that performing searches is one of the main reasons people use the Internet kind of like saying that riding in taxis is one of the main reasons people go to New York City?

But, come to think of it, using a search engine is a lot like taking a taxi: Its simply a way to get to a destination that you want to go to. And, also like a taxi, Web searching often doesnt take the most direct and obvious route.

/zimages/2/28571.gifClick here to read about Microsofts experimental online marketplace, code-named Fremont.

Search is the current rage, among many over-hyped technologies. Google is the search king; Microsoft and others want to be king; and lots of businesses are in mortal terror of what Google and search technologies will do to their business models.

But is search really the technology to rule all other technologies? Color me skeptical. Im certainly not on the side that thinks Google will take over everything and well all be surfing on the GoogleNet. But Im also not among the now-somewhat-fashionable contrarians who think Google is poised for a Netscape-like fall.

I think search is a valuable technology and Google will be an important company for a long time. But it is still basically a middleman technology, and one with its fair share of flaws.

If you doubt me, ask yourself this: Are you truly happy with the results you get and the experiences you have in your regular Web searches, whether using Google or competing search engines?

Was your search as quick, painless and, most important, accurate as it could be? For me, the answer is too often "no."

I know some of you are thinking, "Hold on there—Google is great, and it makes research and much of my life much easier." I agree that theres still a somewhat-magical quality to typing a search term and having potential answers appear. But if you think about it, youll likely agree that, when you do a search, the site you really want to find is several pages deep—or maybe doesnt come up at all.

/zimages/2/28571.gifGoogle brings local coverage to its shopping service, Froogle. Read more here.

I know search engines frequently choke on one of my most common searches—namely, to see if a specific restaurant has a Web site. Most of the time, it seems, I type in a restaurant name and the town its in, and no result pointing to a Web site comes up. Then I decide to just try the model "," and up comes the site.

Im sure you can think of specific cases where a search failed you in a similar way. Sometimes, especially when Im looking for a specific site, I prefer the old Yahoo-style hierarchical category taxonomy. Thats because those Yahoo listings were built by people who knew that if you were looking for all newspapers in the Baltimore area, then you just wanted links to those papers Web sites, not links to other sites talking about Baltimore newspapers. This is where those marvelous search engine algorithms often come up short.

But there is hope, and it doesnt come from more ingenious search algorithms. It comes from the idea of people saying what something is—namely, the recent popularity of tagging.

Anyone who has used the image service Flickr or the bookmarking site knows how this works: The sites basically rely on people looking at something and then adding tags that say what that something is.

Amazon recently announced that it will pay users to tag content in a similar way, and others hopefully will follow.

To me, this will bring a whole new dimension to search and make it possible to find actual things, not just sites that talk about the things. This model may even help to bring about the Semantic Web, where everything understands what everything else is actually about.

So we dont really need smarter search engines, just smarter Web content.

Labs Director Jim Rapoza can be reached at

/zimages/2/28571.gifCheck out eWEEK.coms for the latest news, views and analysis on enterprise search technology.