College and Research Libraries Book Reviews Tufte, Edward R. Envisioning Informa- tion. Cheshire, Conn.: Graphics Pr., 1990. 126p. $48 (ISBN 0-9613921-1-8). LC 90-166920. Tufte, Edward R. The Visual Display of Quantitative Information. Cheshire, Conn.: Graphics Pr., 1983. 197p. $36 (ISBN 0-9613921-0-X). LC 83-156861. Envisioning Information is Edward R. Tufte's second book on information de- sign. His 1983 Visual Display of Quantita- tive Information is already a classic; the new book is sure to become one as well. Visual Display deals with statistical graphics, charts, maps, and tables, set- ting forth principles for design and criti- cism, applied to a rich variety of good and bad historical examples, and illustrated with new designs. Envisioning Information goes mtich further, covering visual displays of information of all kinds, from the Vietnam Veterans Memorial to railway timetables, electrocardiograms, guidebooks, scientific visualizations, maps, computer screens, dance notations, and much, much more. The two books complement each other, and both should be studied. The new work does not add much to the theory of the first, _but extends its range of applica- tion-notably in the direction of color, which is treated rather grudgingly in the first book, but celebrated in the second: color used to label, to measure, to repre- sent, as well as to enliven and decorate. (Envisioning Information is printed in six colors, except for the chapter on color, which is twelve colors on eleven-that is, twenty-three printing units.) The first book is about "the use of abstract, non- representational pictures to show num- . bers"; the new book is about ways of representing complex and multidimen- sional information of any sort. Tufte says it is about the "escape from flatland," his 382 way of describing the attempt to repre- sent the dynamic and ·multidimensional in two dimensions-the flat, static page. Tufte teaches statistics, political econ- omy, and graphic design at Yale, but he can as well be called an information sci- entist and these books contribute to in- formation theory. Consider these to be works on alternatives to discourse. For many librarians and information scien- tists, as for many cognitive scientists, the word is primary, the sentence (or its ab- stract shadow, the proposition) is the basic unit of thought and communica- tion, and discourse, connected strings of sentences, is the basic vehicle for reason- ing and instruction. Graphics are merely illustrative or decorative, but not essen- tial and not efficient. These books argue against such views. In the words of its author, Visual Display is about "how to communicate information through the simultaneous presentation of words, numbers, and pictures." It catalogs the basic structures for showing data: sen- tences, text tables, tables, semigraphics, and graphics. Which is best for what purpose is one of the basic tactical prob- lems of the information designer. For in- stance, tables are often preferable to graphics and to prose for small data sets. But "often the most effective way to de- scribe, explore, and summarize a set of numbers ... is to look at pictures of those numbers." Tufte will have nothing to do with the idea that simplification in the presenta- tion of information is necessary to avoid information overload, boredom, or in- comprehension. "What is to be sought in designs for the display of information is the clear portrayal of complexity." High information density, data-rich displays are what people want, and adding detail may often clarify. "We thrive in informa- - ---------------------------------------------------------------~ tion-thick worlds because of our marvel- ous and everyday capacities to select, edit, single out, structure, highlight, group, pair, merge, harmonize, synthesize, focus, organize, condense, reduce, boil down, choose, categorize, catalog, classify, refine, abstract, scan, look into .... Clutter and confusion are failures of design, not attri- butes of information." Tufte's books are profound meditations on ways in which information can be envisioned, "in order to reason about, communicate, document, and preserve that knowledge." They are exhibits of, and reflections on, works of "cognitive art," beautiful works of useful information, works at the intersection of image, word, number, and art. Anyone involved in the design of in- formation displays, from library hand- outs to computer interfaces, should take time to study these books. But I would also put these books on the reading lists for library school courses in reference and bibliography (especially the evalua- tion of reference works), collection de- velopment, communication, cataloging, information retrieval theory, and infor- mation systems design, and would rec- ommend them to anyone seriously interested in any of those subjects, not just to clarify their ideas about what makes for good or bad visual displays of informa- tion, but as instruments for thought about thought, communication, and information. These are source books, vivid demon- strations of graphic power, that have the potential to change an individual's view of information: away from the view that discourse is primary and graphics are simply illustration, toward the view that discourse is often problematic and that methods for the graphic display of infor- mation are for mariy purposes superior in the portrayal of density, complexity, and dimensionality. Tufte offers us deep considerations on the limits of discourse, with implications for how we think about communication and the storage and retrieval of information. Envisioning Information is, not at all in- cidentally, irresistibly beautiful. It is cheap at its price. Both books must be in any decent academic library; many li- brarians and information scientists will Book Reviews 383 insist on having their own copies as welL-Patrick Wilson, University of Cali- fornia-Berkeley. Advances in Library Resource Sharing. Cargill, Jennifer, and Diane J. Graves, eds. V.l, Westport, Conn.: Meckler, 1990. 238p. alk. paper, $55 per year (ISSN 1052-262X; ISBN 0-88736-490-X). In one of the best essays in this book, Marsha Ra makes a credible case that "resource sharing as we now understand it will probably cease to exist." We almost certainly are looking at a paradigm shift in libraries. Whether "advances" -the opti- mistic word used in the title of this col- lection of essays-is the right word for this shift is profoundly uncertain. This is not a good book. It is cluttered with too many essays that were written without evident purpose. We do not need yet another account of the Center for Research Libraries, or an article on the economics of resource sharing that contains no economic analysis, or a set of unthoughtful reports on regional resource sharing, or a complaint about library services from a faculty person who is myopic, uninformed, and cranky. There is not much in this book to suggest that its compilers had a definable edito- rial purpose (other than to produce a book) or took much care to create a vol- ume of value and merit. The compilers promise an annual volume on resource sharing. Let us hope for other things. Amid this dross, there are some essays that merit attention. Richard M. Dough- erty and Carol Hughes issue the now familiar call for libraries to shift their mission from owning information to providing access to it, to shift from deliv- ering bibliographic units to delivering information, and to do this in ways that are speedy, convenient, and customized to the individual reader's needs. Marsha Ra picks up this theme and observes that electronic networks, uniform communi- . cation standards, expert systems, and workstations will soon permit resource sharing with little direct involvement of librarians. Some of the transformations in authorship and publishing that elec- tronic media will require, if we are to