What weighs less than one millionth of an ounce, consumes less than a millionth of a cubic inch, holds 4 million bits, and costs less than US$2? What weighs more than 1 pound, is larger than 50 cubic inches, contains less than 4 million bits, and costs more than $20? The same thing: Being Digital stored on an integrated circuit and Being Digital published as a hardcover book.

The most common question I get is, Why, Mr. Digital Fancypants, did you write a book? Books are the province of romantics and humanists, not heartless nerds. The existence of books is solace to those who think the world is turning into a digital dump. The act of writing a book is evidence, you see, that all is not lost for those who read Shakespeare, go to church, play baseball, enjoy ballet, or like a good long walk in the woods. Anyway, who wants to read Michael Crichton's next book, let alone the Bible, on screen? No one. In fact, the consumption of coated and sheet paper in the United States has gone from 142 pounds per capita in 1980 to 214 pounds in 1993.

It's not a book - it's bits
The word is not going anywhere. In fact, it is and has been one of the most powerful forces to shape humankind, for both good and bad. St. Thomas said a few words in southern India almost 2,000 years ago, and today the southern province of Kerala is 25 percent Christian in a country where Christians are less than 1 percent of the population. There is no question that words are powerful, that they always have been and always will be. This back-page column, except for my loathsome picture, has never had anything in it but words.

But just as we seldom carve words in rocks these days, we will probably not print many of them on paper for binding tomorrow. In fact, the cost of paper (which has risen 50 percent in the past year), the amount of human energy required to move it, and the volume of space needed to store it make books as we know them less than the optimum method for delivering bits. In fact, the art of bookmaking is not only less than perfect but will probably be as relevant in 2020 as blacksmithing is today.

It's not bits - it's a book
And yet books win big as an interface medium, a comfortable place where bits and people meet. They look and feel great, they are usually lightweight (lighter than most laptops), relatively low-cost, easy to use, handsomely random-access, and widely available to everyone. Why did I write a book? Because that is the display medium my audience has today. And it is not a bad one.

We can "thumb" through books, annotate and dogear their pages - even sit or stand on them when we need to be a mite taller. I once stepped on my laptop, and the result was awful.

The book was invented 500 years ago by Aldo Manuzio in Venice, Italy. The so-called octavo format was a departure from previous manuscripts because it was handy, portable, and pocket-size. Manuzio even pioneered page numbering. Odd how Gutenberg gets credit while Manuzio is known to only a few. Today's Manuzios are the flock of researchers looking for display materials capable of producing handy, portable, and pocket-size flat-panel displays for PDAs (personal digital assistants, a term coined by John Sculley five years ago and one of the weirdest acronyms to stick). In general, these efforts miss the point of "bookness," because the act of flipping through pages is an indisputable part of the book experience. In 1978 at MIT, we animated flipping pages on a screen and even generated fluttering sounds. Cute, but no cigar. A new effort by Joe Jacobson at the Media Lab involves electronic paper, a high-contrast, low-cost, read/write/erase medium. By binding these pulplike, electronic leaves, lo and behold - you have an electronic book. These are quite literally pages onto which you can download words, in any type, in any size. For the 15 million Americans who want large-print books, this will be a gift from heaven - if Joe succeeds during the next couple of years. So, those of you who don't want to climb into bed with "Intel inside," there is hope. This is the likely future of books.

The model said never to work
When my colleagues and I argue that the mass media of the future will be one that you "pull from" versus one that is "pushed at you," we are told: Poppycock! (Or worse.) These naysayers argue that a "pulling" model cannot be supported because it eclipses advertising. While I am not sure it is even true, let's pretend that it is and ask ourselves: What mass medium today is larger than the American TV and motion picture industries combined, has no advertising, and is truly, as George Gilder puts it, a medium of choice? The answer: Books.

More than 50,000 titles are published in the United States each year. Guess the typical number of copies published per title. A major house considers 5,000 to be about the lowest run it can support economically, while some of the small houses consider 2,000 copies of a title a large run. Yes, more than 12 million copies of John Grisham's novel The Firm were printed, and the first run of Bill Gates's book was 800,000. But the average is much smaller, and these less massive books are not unimportant. They just interest or reach fewer people.

So, the next time you ask yourself about the Web (which is doubling in size every 50 days) and wonder what will economically support so many sites (today, one homepage is added every 4 seconds), just think books. You say to yourself, Surely most of those Web sites will go away - no way. There will be more and more and, like trade books, there will be an audience for all of them. Instead of worrying about the future of the book as a pulp standard, think about it as bits for all: bestseller bits, fewer specialty-seller bits, and no-seller bits for grandparents from grandchild.

Meanwhile, some of us in research are working really hard to make them feel good and be readable - something you can happily curl up with or take to the john.