Craig S. Mullins

Return to Home Page

September 2007





The DBA Corner
by Craig S. Mullins

Simplified Database Search


Finding “stuff” on the Internet would be extremely daunting if it weren’t for search engines like Google and Yahoo. These search engines offer an interface so simple that even your Grandmother can use them. I mean, who doesn’t understand a text box with a big button next to it labeled “SEARCH”?

It is this ubiquitous search capability, more than anything else, that has improved the usability of the Web to make it a useful business tool. Without search engines, all of that information may still be “out there” on the Web, but we would not know where to go to find it. Without the indexing of all of that information on the Web, coupled with a very simple interface for searching those indexes to find the information we are looking for, the Web would not be the roaring success it is today.

OK, so why am I talking about web search in a column about databases? Well, think about the potential power in marrying the two technologies. Databases are the mechanism of choice for storing and accessing large volumes of critical business data. Search engines are the mechanism of choice for locating useful information from within a huge store of data. What if we could offer a search engine like capability on top of our database data? Think about it.

How do we find data in a database today? For the most part, we write SQL queries. But this is a talent that is mostly geared to IT people. Oh, there are those power users out there that can write some SQL, but they are more the exception than the rule. But everyone can “develop” a search engine query.

So, envision a box with a “SEARCH” button next to it. And you could type whatever you wanted into the box and receive a result based on querying your databases. Let’s type, say, “What were our gross sales for blue widgets in the second quarter of 2007”… and here comes the data. That would be easier than having to type something like this:

AND    DATE BETWEEN ′2007-04-01′ AND ′2007-06-30′;

Let’s face it, most business users don’t know how to write such a query, and don’t want to know how to write such a query. So instead, they call up their IT contact and a programmer or DBA writes the appropriate queries. And this creates a bottleneck as business people wait for IT people to write code to get them the data they need.

But guess what? There is a solution out there that allows corporate databases to be searched this way. The company is called Enth, Inc. ( Enth provides a search-like experience to database data.  Instead of directing searchers to web pages, Enth will create new content in response to a search request by converting the request into dynamic SQL requests that are issued against databases.

You can see this technology at work for yourself if you visit Enth’s web site. They have partnered with Stats, Inc. to provide access to National Football League and Major League Baseball data archives. So you can use the simple search interface you are used to for searching the web, to aggregate, filter and sort sports statistics that are stored in a structured database. Try it yourself. Once you get to the web site go to the “Getting Started” tab. Once there, I typed in “How many home runs did Willie Stargell hit in 1971” and I got back 48.

I think this is something that holds a great deal of promise to opening up the vast treasure troves of business data to non-technical users. And that should improve productivity and perhaps even profits... and that is a good thing, right?






From Database Trends and Applications, September 2007.

2007 Craig S. Mullins,  All rights reserved.