First You Search for the Database, Then You Search for the Data

Developer News Microsoft Shows Off Silverlight 4, IE9 Plans Metasploit Expands Vulnerability Test Framework HyperCard Reborn?

Northern Light: The Future of Finding Journal Articles?

There's another problem. Even if you have access to a collection of high-quality databases because of your association with a university or a big company, the databases tend to be physically distinct, with distinct search interfaces. For instance, take a look at the scholarly databases provided to members of the Michigan State University community. A huge range of high-quality materials is offered.

Frankly, most users probably find the list of available databases rather befuddling. The names of the databases -- words like "First Search" or "ERL" or "Agricola" may mean something to trained information professionals, but not to most folks. Unless you already have a pretty good idea what database you need by name, you either pick one randomly, or you stare at the screen and give up.

Of course, part of becoming an expert in a discipline means learning the literature, and today the literature is largely in databases. Virtually all researchers in medicine know they can't do research without Medline, for instance. But anyone new to a field, or doing a search outside one's field, is likely to be baffled by the choice of databases.

That was the big problem with WAIS, or Wide Area Information Servers -- the first popular Internet-based, multi-topic database tool, created by Brewster Kahle in the early 1990s. You had to search for a "source" (or database) and then you did your actual search.

Most people are too lazy to do both steps. In some cases, we're willing to take note of a special-purpose database. For instance, we know that Switchboard is the place to look up phone numbers for individuals or businesses. But for most other searches, most people don't know OCLC's First Search from Lexis-Nexis from Infoseek. With dozens of brand names for databases, most of us don't want to wade through and learn which ones specialize in what kinds of content. Instead, we want one place to type our query, and we want one hit list of high-quality answers.