New Databases Make Updating Rolodex a Cinch

The old Rolodex is stuffed and disorganized. As you flip through it, trying to remember where you filed Ms. Smith, you wonder: Wouldn't it be easier to put all this on computer?

It's a dangerous thought, because databases can quickly become complex. But a new crop of databases is making the task less daunting than it used to be. And if you keep it simple, the rewards of computerizing your list of contacts can be great.

The simplest way to computerize your Rolodex is to get a flat-file database. These are straightforward products found in most personal-information managers, calendar, and contact-manager software, as well as most of the business-software suites now available. It's hard to pick one program's database over another, because users typically choose such programs for other features. In any case, all their databases use the same approach: each record contains all the information about each individual or item.

Say John Jones is in your database. His record has his name, phone number, title, company, address, and so on. The flat-file database is easier to use than the Rolodex because you can search on any part of a record to find someone. Can't remember Sandra's last name? Search on Sandra instead and you'll find all the Sandras you've typed in. Having trouble spelling Bueurengetti? Search his company name instead. You can even search your database to find all your contacts in a particular state or area code.

This is pretty powerful stuff, and anyone can do it. The challenge comes as your database grows. For example, suppose Mr. Jones's company moves, and it turns out that you have nine other people from that company in your records. With a flat-file database, this means you have to find each person's file and type in the new company address 10 times.

Or maybe you're running an electrical-supply store and want to keep track of who orders what. Since one customer makes several purchases over the year, the flat-file model of one record per person doesn't work very well.

That's why software companies also make relational databases. By relating one file of names with another file of companies, you can enter a company's new address once and it's updated for every employee listed in your database. By relating a table of customers to another table of orders, a relational database can reveal who is buying what when.

With this power, however, comes sometimes-enormous complexity. You have to learn how to relate various tables (which isn't always easy). Try to be a wee bit fancy and you're suddenly thrust into the role of computer programmer.

Fortunately, the latest versions of two of the most powerful relational databases for consumers - Microsoft Access for Windows 95 and Borland Paradox for Windows 95 - shield users from much of this complexity. Using one of their many preformatted models, the average user with a modicum of courage can create some fairly sophisticated and nice-looking databases.

Although a longtime Paradox user, I was more impressed with Microsoft Access. Its tutorial was simpler, and the interface was easier to use in most cases. The program was more flexible, while offering plenty of power. (The latest edition of PC Magazine names Access along with Lotus Approach as its top choices for easy-to-use databases. Approach didn't come in time for my testing.)

But don't get carried away by these programs' ease-of-use claims. Stray off their preformatted path and you quickly run into their complexity. Unless you're a power computer user, the rule with databases is to keep them as simple as you can.

*Send comments to

We want to hear, did we miss an angle we should have covered? Should we come back to this topic? Or just give us a rating for this story. We want to hear from you.