The second straw is a story in the January 1997 issue of Communications of the ACM. The writer Robert Fox reported that check-out of books increased by a whopping 75 percent in the new "cyber-age" San Francisco Public Library, which offers free and highly popular Internet access (pp. 20-21). San Francisco's "New Main" is not without controversy. But it clearly offers no evidence that Internet access reduces interest in books. Indeed, one of the librarians interviewed makes the argument (as others have) that the appeal of the Internet attracts new readers, who then explore the broader range of services and materials that the library offers.
Many of us have seen these kinds of debates before: What is the role of quantification in the social sciences was a popular one in the 1960s and 1970s? What should constitute the canon was hotly contested in the 1970s and 1980s? To the former, we have found that some questions are amenable to computation and others are not: use computation as appropriate. To the latter, one wise commentator observed that who was in "the canon" mattered less than that a canon -- any canon -- was read.
So what is the bottom line? Not that the debate over what goes into print and what is expressed digitally is sterile, but that framing it as a false dichotomy is. In our lead story, John MacColl will argue that there are some venues for which print is and will likely remain the preferred medium, but that there are others in which digital or a combination of parallel print/digital distribution will be the better solution. And that this is not purely a matter of technology but rather has much to do with users' behavior coupled with the technological options for conveying the information.
To that, we would only add that the fundamental issue is literacy. We believe that the digital medium can contribute to literacy in general and to scientific and technological literacy in particular. This is important. But the job is not ours alone.