College and Research Libraries Book Reviews Robinson, Peter. The Digitization of Pri- mary Textual Sources. Oxford: Office for Humanities Communication, 1993. viii, 104p. $5 (ISBN 1-897791-05-4). Robinson, Peter. The Transcription of Pri- mary Textual Sources Using SGML. Ox- ford: Office for Humanities Communi- cation, 1994. viii, 136p. $5 (ISBN 1- 897791-07 -0). The Office for Humanities Communi- cation is part of Oxford University Com- puting Services, but is funded by the British Library Research and Develop- ment Department. Its series of Publica- tions, of which these are volumes 4 and 6, aims to introduce humanities scholars to ways in which computers impact on their teachi,ng and research. The issues that these two complementary works address also concern many librarians, either as potential creators of electronic information, or as customers for the growing number of commercially pro- duced text and image databases. Both books include many of the same elements: an introduction discussing why one might undertake such a project and some of the decisions it will require; brief descriptions of some existing proj- ects, their unique problems and the so- lutions found; and a clear introduction to the technical issues involved. Never- theless, the emphasis is quite different. Digitization of Primary Textual Sources is concerned with producing digital images of existing materials. Although Robinson's primary focus is on manu- scripts, he also describes projects in- volving paintings, sculpture, and the preservation of printed materials. Much of the book is given over to a discussion of the various technical choices (e.g., resolution, file formats, compression, image capture with scanners and digital 568 cameras, image storage) and copyright issues. His presentation of these issues will be helpful to those considering any sort of digitizing project. Robinson has clear preferences and does not shy away from specific recommendations, but he also recognizes that local needs and abilities may dictate less than ideal solu- tions. He is careful to point out the strengths and weaknesses of each ap- proach and gives scaled recommenda- tions for treatment of various types of materials. The rather specific title of the second book, Transcription of Primary Textual Sources Using SGML suggests the differ- ence in its approach. Although Robinson discusses tagging schemes that have been developed for particular materials and admits that for a given, limited proj- ect an alternative tagging scheme may provide an effective and efficient solu- tion, this book is less concerned with choosing among alternatives than with introducing the reader to Standard Gen- eralized Markup Language (SGML), particularly as it has been elaborated .by the Text Encoding Initiative. The needs of publishers, printers, and large organi- zations for a standardized markup sys- tem independent of language, software, and hardware have brought stand- ardization in this area further along than for digital images. Robinson argues that use of SGML not only saves a great deal of time and energy for scholars, but it also frees future generations from de- pendence on obsolete hardware and software. "Content-aware" markup of a text is a complex matter requiring many very specific decisions. Robinson's de- scription of markup will seem overly detailed only to those who have not yet looked at the TEl Guidelines. Neverthe- less, Robinson offers a useful introduc- tion to the process of markup, as always emphasizing the decisions to be made, and provides useful examples of the ways in which thoughtful markup can aid analysis of a text. Robinson has gone to some effort to ensure that these books are up to date (e.g., working from a proof copy of the most recent revision of the TEl Guide- lines), and even Digitization (completed in summer 1993) shows little sign of age a year later. A major reason for this is Robinson's success at identifying the emerging standards for treatment of electronic materials. His books are use- ful introductions to the issues, whatever decisions the reader may make about a particular project. They are of special importance for librarians, who have to worry about more than the needs of one project or about who will want to buy a CD-ROM this year. By urging scholars and publishers to use standardized for- mats, librarians can provide users with standard interfaces for many different da- tabases, reduce their investment in hard- ware and software, and, perhaps most importantly, ensure that the best of today' s scholarship is available tomorrow, so that we do not have to redigitize or retag hun- dreds of thousands of images and pages of manuscript because we can no longer in- terpret them.-James Campbell, Univer- sity of Virginia, Charlottesville. Chartier, Roger. The Order of Books: Read- ers, Authors, and Libraries in Europe be- tween the Fourteenth and Eighteenth Centuries. Translated by Lydia G. Co- chrane. Stanford, Calif.: Stanford Univ. Pr., 1994. 126p. alk. paper, $35 (ISBN 0- 8047-2266-8). At first glance, Chartier's new book would seem to be of extraordinary inter- est to academic librarians who like to explore the theoretical underpinnings of their profession. We recognize in the author's name a historian of ideas whose chief focus is the history of books and reading. The word order in its title offers provocative layers of meaning, whether applied to individual texts, to book pro- duction, to library shelves, to bibliog- raphies. The enticing titles of the three Book Reviews 569 essays comprising the text read "Commu- nities of Readers," "Figures of the Author," and "Libraries without Walls." The twenty- one pages of footnotes refer to a wide range of studies in various languages that one might wish to follow up. In the essays, originally published in French in 1992, Chartier questions, elu- cidates, and theorizes about ideas put forth by himself and others on such sub- jects as the relationships between text, print, and reader; changing perceptions of the connection between author and text, literary property, and censorship; and the various meanings of library (bib- ยท liotheque) as space, compilation/ conden- sation, or book catalog. As might be expected, many of his examples or expli- cations originate out of early modern French culture; for instance, in two of the essays he contrasts Bibliotheques by La Croix du Maine (published in Paris in 1584) and by Du Verdier (published in Lyons in 1585). The essays are ostensibly tied together by a preface and an epi- logue, which obliquely suggest what may be the author's taking-off point: Chartier's restating of the "complex, subtle, shifting relationships" between books and their readers in earlier times as a paradigm for his uneasy coming to terms with the world of electronic texts. In the first essay Chartier emphasizes how meanings of texts depend upon the forms in which they reach readers. Put- ting the French text of his own book beside the English confirms this point all too ironically. The English-printed and designed in Great Britain although pub- lished in this country-is presented with large, clumsy, poorly spaced type. The eight plates (nine in the French) have been unsettlingly resized as well. The appendix containing valuable original language versions of translations within the French text is replaced by an index in the English version. The French version includes a note on Chartier's scholarly interests and a list of his previous books, neither of which is carried over into the English. Within the text Roger Chartier himself refers to the essays as a "few reflections" or "historian's musings" (les detours his-