Today, invention pre-empts expectation. The University of Houston's College of Engineering presents this series about the machines that make our civilization run, and the people whose ingenuity created them.

This rainy afternoon I'm talking with the head of computer systems in our library. He says, "I don't read my morning paper to learn anything in particular. I read it to answer questions I'd never thought of asking."

He's making a point. I'd asked him to create some network software. He answered with a long list of possibilities. My frustration grew. It was then he talked about reading his paper.

This man has to dance among fast-breaking new information systems like spit on a griddle. Nothing he does ever fits the world of next month quite the way he plans it.

In 1990 he was part of a project to combine the card catalog data from seven libraries on CD-ROM. Now the computer scans millions of books in seconds. I feed it fragments of a title, an incomplete name, an ill-defined subject. It finds what was once unfindable with blinding speed. And it drives his point home.

The CD-ROM system was meant to help us look for things. But, like someone with a newspaper, I'm not looking for anything in particular. This system has become my serendipity machine.

For example, I heard about a new book on women in science. The system turned up another book by the same author. That book was about Baroque stagecraft. It tied stagecraft to alchemy. It also linked alchemy to an ancient text on Roman technology.

Suddenly I saw history in ways I'd never imagined. This system was meant to find things. But it does that so well that it exposes connections you'd never see in a card catalog.

The other day, I looked up an 18th century physiologist. The CD-ROM system told me the fellow had written a very popular volume of poetry. That was a crazy unexpected find. And it sent me off to learn about pre-romantic German poetry and the people who wrote it.

So I use this invention in ways it wasn't meant to be used. We've only had this technology for two years. Yet it's already changed me, and my working methods, radically.

I would never've thought to ask an inventor for this system. I surely doubt that he meant to give me a serendipity machine. But now the deed is done. The future is changed.

So back to our conversation: My friend will eventually give me my network software. When he does, it'll either serve users in ways neither of us can anticipate; or it'll serve nobody and be forgotten. For, in the end, everything that's good about a new technology surprises us. It has to. And that is, at once, both the wonder and the terror of all invention.

I'm John Lienhard at the University of Houston, where we're interested in the way inventive minds work.

(Theme music)

I'm grateful to Tom Wilson, Head of Computer Systems, University of Houston Library, for his contributions to this episode. The CD-ROM catalog system is called HARLiC. HARLiC stands for Houston Area Research Library Consortium.

The Libraries in the Consortium include those at the University of Houston campuses, The Houston Public Library, The University of Texas at Galveston Library, the Medical Center Library, the Texas A & M Library, the Prairie View A & M Library, and the Texas Southern University Libraries. After April 1, 1992, the Rice University Library will appear on the system.