It's hard for me to conceive of a more useful tool tool than Amazon's recently released Search Within the Book. By using the online bookseller's standard search window, you can now search the full-text of about 120,000 titles, displaying the pages of the books containing your results on-screen. These are current, copyrighted titles, made available through arrangement with their publishers.

If your first thought is that Google already does this, guess again. What Google indexes is the vast profusion of Web pages. But most books, especially those under copyright, cannot be found in full-text form through any Web search engine. Indeed, running a Google search usually provides you with peripheral material, supplemental to the primary sources you would find in books.

And with more and more students now using Google and little else as research for papers and facts, the odd thing about the Web is that it has inspired a great devaluing of information. Careful (and legitimately so) to protect copyright, we have created a parallel information source that offers its own virtues but features only a small subset of the knowledge available in books.

Amazon's CEO Jeff Bezos is taking interesting steps to solve this predicament. I discovered it while writing a recent project about Freeman Dyson and Project Orion, a government attempt to create an atomic spaceship back in the 1960s. Having worked for months, I already knew what the major sources were. But I plugged in "dyson" and "starship" as search terms on Amazon and a wide variety of hits resulted.

For any of these, I could look at the standard Amazon information about the book, such as price, title, publisher, etc. But I also had a listing for an excerpt from any book where my search terms were found. Clicking on the link brought such pages to my screen, with the ability to go forward or back an additional two pages. That's the limit -- four pages, because Bezos' agreements with publishers have been careful to honor copyrighted text.

But what a find those four pages can be. I located one book that I hadn't found before that discussed Orion and its relation to another wild concept, the British Interplanetary Society's Project Daedalus starship design. Even more useful, Amazon's new search led me to bibliographical entries, where I could find lists of books and articles that I could compare to my own research.

In this way, I was able to supplement and, in one case, adjust my presentation of a key fact in the chapter I am writing. Just as handy was the ability to revisit books I already own in physical form. On my shelves, they're obviously hard to search, even with good indices. But I could quickly check a fact or re-scan a bibliography in any of them by running the right terms through Amazon.

This is exciting stuff, because while 120,000 books are a subset of what's out there, the principle being established is that a catalog of online books doesn't have to destroy the sale of the printed volumes. As Amazon presents them, the book searches are restricted, and I wound up buying several titles I knew I would want to read from cover to cover. That's smart bookselling in every way.

American copyright laws keep many titles that are long out of print protected, even though they sell few copies even as used books. A search like this one could open up such a back catalog, reawakening interest in older books and perhaps spurring "print on demand" services. That would be good not only for publishers but also for any information seekers. And it demonstrates that far from being enemies, digital text and the printed word can enhance each other.