In this issue, Alan Tupek and Cathryn Dippo point out that making statistical information more easily available via the web supports research activities but also increases the chances that these data will be misunderstood and misused. Thus, making the data available carries a responsibility to educate users, enhancing the promise of increasing what the authors call Americans' "quantitative literacy."
This connection between access to data and education is a subtle and rather interesting point. As institutions, libraries are prominent buildings on college campuses, and retrospective literature searches are features of scholarly discussions in fields as diverse as mathematics and art history. The role of public libraries in support of a national educational mission is certainly routinely stated, although public librarians are generally absent from discussions of research. You would look long and hard, for example, to find much in D-Lib Magazine about public libraries, and few of their representatives come to such venues as the Coalition for Networked Information (CNI) or the Digital Library Federation (DLF). An exception is the new California Digital Library (CDL), which has begun as an initiative among the campuses of the University of California system but calls for interaction with the state's public libraries.
In a research setting, the role of libraries is usually filled by representatives of academic libraries, presumably with the understanding that new technologies will diffuse through the library community starting with the universities and extending outward. I suspect that this model of technology diffusion can be challenged if only on the grounds that the public library audience is substantively different from the research library constituency, different to the point that the one is not a microcosm of the other. Technologies successfully deployed in a research setting, therefore, may not migrate or scale to the heterogeneous environments in which public libraries are found, witness current debates over the use of Internet filters in public libraries. Or the priorities may be different. Indeed, the British Library has found that special libraries are quite different and, as reported by Mark Watson and David Streatfield, has solicited a study on their information technology needs as a first step toward contributing to the research agenda. To the extent that the on-line environment is likely to resemble a public library, future research is likely to require broader participation to understand the multiple constituencies for information technologies.
The close coupling Tupek and Dippo see between data and educational tools also resonates with research on the effect that the information technologies can have on the organization of work -- whether within an organization or among organizations. Research by Thomas Malone of MIT and others on the implications of networked information technologies in industry suggests that there can be a "flattening" of the organization as well as outsourcing of specialized functions to increase overall efficiency. Thus, there is both increased specialization among organizations and reduction in the number of functional levels within an organization. Information technologies can change the distribution of work but do not necessarily reduce the burden.
We can see similar kinds of re-structuring in the types of information technologies associated with digital libraries -- but digital libraries support human institutions and not industrial processes. Certain functions will survive that are independent of the organizational niches and specific technologies that traditionally defined and supported these functions. That is to say, librarians and teachers are more than MARC records and chalk boards; the question is which functions and how?
Certainly the approaches to cross language retrieval that Douglas Oard describes promise to put more data and more tools in users' hands. Yet although the notion of finding and using information across many languages is intrinsically interesting, we potentially lose something in translation. As Shigeo Sugimoto and his colleagues point out in their October 1997 story on Japanese folktales, one of the challenges that translating these folk tales poses is cross cultural -- for example, that the word for "ogre" in Japanese connotes something quite different from the English image. So translation in a cultural vacuum is not to be desired, any more than the misuse of statistics. Rather, access to information in many languages or data in many formats creates both a need and an opportunity for formal, informal, and lifelong learning.
Historically, reference librarians undertook to mediate between users and information and thus to perform this educational function. Although the human presence vanishes in the electronic environment, the functions do not, and losing librarians and teachers from the loop exacts a price. Whether their contributions can be captured in technology is an open question. The answer is probably some version of "sort of", that is, some of these functions can be technologically enabled for some users some of the time. To a man with a hammer, all things are a nail. Digital library technologies are neither as blunt nor as elegantly simple as a hammer. I devoutly hope we do not turn the world into a nail.