reviews Book Reviews 289 Book Reviews Arms, William Y. Digital Libraries. Cam­ bridge, Mass.: MIT Pr., 2000. 287p. $45, alk. paper (ISBN 0262011808). LC 99­ 14773. This important, but difficult-to-read, book is likely to elicit a broad range of re­ sponses and reactions from its readers. A tradition-minded librarian might easily be tempted to think at once about early retirement, whereas someone eager to explore the ever-changing panorama of new technologies will excitedly welcome its vision of the future. The first chapter, a broad gauge tuto­ rial, probably designed mainly for aca­ demic administrators or university gov­ erning boards, is unlikely to tell active li­ brarians anything they do not already know. The following three chapters de­ scribe the development of the Internet, detail with considerable accuracy the sometimes-contentious relationships be­ tween libraries and publishers, and dis­ cuss contemporary innovation and re­ search. Arms correctly points out that many people who grew up with the Internet believe they have “discovered” online information systems and are com­ pletely unaware that the concepts form­ ing the basis of digital libraries were long ago researched and developed by the li­ brary community. Librarians are not likely to receive well the author’s statement in chapter four— “Computer scientists take research seri­ ously ”—with its clear, gratuitous pre­ sumption that librarians do not. Computer scientists, writes Arms, are the hares of digital libraries, whereas large libraries— and by extension, one can infer, their li­ brarians—are the tortoises. Yet, there is little discussion about why librarianship was for so many years averse to research, an attitude constantly criticized by early pioneers of library research such as Pierce Butler and Jesse Shera. There is no discus­ sion of the obvious arrogance among some computer scientists who, for years, held back the develop­ ment of practical, user-friendly information systems. Missing from the work is any substantial consideration of the research role of the former Council on Library Resources (CLR); its work is alluded to only in passing. Arms suggests that re­ search in information science is “at least 30 years old,” but it is actually almost twice that age, dating back to the 1940s when Ralph Shaw developed and demonstrated his RapidSelector. Arms does not mention the pioneering theoretical research done at MIT in the mid-1950s by Calvin Mooers, perhaps the first person to connect a type­ writer terminal to a mainframe, nor does he discuss the important developmental work on automatic indexing done over an entire lifetime by Gerard Salton at Harvard and Cornell. Throughout the work, Arms discusses in substantial detail a variety of impor­ tant, major issues. The interoperability of various computerized information sys­ tems, the costs of standardization, and the need to maintain legacy systems are given valuable and thoroughgoing discussion. Chapter five, devoted to personnel, us­ ers, and organizations, is one of the best treatments anywhere of how challenging it is to weld together disparate interests into a fundamentally restructured, uni­ fied service mechanism. The author prop­ erly rejects as naive the idea that technol­ ogy “eliminates” the need for professional management of information. Quite the opposite, Arms maintains that technology more than ever demands managerial tal­ ent strong enough to administer person­ nel, equipment, and services that vary widely in every dimension. Librarians may not want to hear about the univer­ sity library’s declining importance rela­ tive to other information resources, but Arms does not shrink from stating this 289 290 College & Research Libraries fact. Citing an anonymous University of California professor, Arms acknowledges baldly that for many students, the Internet is the library. Chapter six on economic and legal issues and chapter seven on access management and security are models of clarity and detail. Chapter twelve, which deals with object models, identifiers, and structural metadata, could profitably have come much earlier in the book, for it lays an excellent foundation for the technical details underlying the concept of the digi­ tal library. The permanence and durabil­ ity, or more accurately, the impermanence and evanescence, of digital data are top­ ics that Arms discusses at length through­ out the book. Just ten pages from the end may be the work’s most important point about digital libraries and permanence, a quotation from Michael Wettengel, a Ger­ man archivist: “Computer technology is made for information processing, not for long-term storage.” There are problems with this book that make it difficult to read. Overall, a choppy style lends an aura of incoherence to parts of the work; sometimes paragraphs con­ tain simple declarative sentences so short that the reader is left wondering about the connections between them. Similarly, some chapters have weak transitions and do not segue well into each other. It is disappointing to find grammatical and spelling errors that a competent proof­ reader should have caught immediately. Did author and editor rely on the notori­ ously undependable grammar checkers that come with word processors? A maddening feature of the book stems from its graphic design. Throughout the work are dozens of sidebar-like “panels,” each containing a succinct summary of some extraordinarily important or essen­ tial technical matter (e.g., Java, the Dublin Core, Real Audio). Unfortunately, the designers chose to print these panels onto an overdense, half-tone background. Con­ sequently, the panels are virtually impos­ sible to read with any degree of speed or comfort. Furthermore, the arrangement almost guarantees lack of easy compre­ hension. Only a dedicated (or perhaps May 2001 masochistic) researcher will get through the panels, without which the full import of the author’s work cannot be appreci­ ated. It is astonishing that so distin­ guished a publisher would permit this intrusion into the communication process and quite ironic that it occurs in a book whose chief subject is communication over the ages. If the work is reprinted or revised, correction of its editorial and pro­ duction flaws is mandatory. In a bizarre and embarrassing error in panel 9.1, which deals with the Oxford English Dictionary, Arms credits its cre­ ation to “James Morris and his col­ leagues.” It is universally known that Sir James Murray (1837–1915) almost single­ handedly created the OED, having spent the last thirty-six years of his life on the project. (The OED’s home page lists all editorial personnel connected with the project from its inception; there is no ref­ erence to James Morris.) A generally excellent glossary helps the reader understand numerous techni­ cal terms and acronyms, although some acronyms are left undecoded and entered into neither index nor glossary. For ex­ ample, CNRI (Corporation for National Research Initiatives) first appears on page 90, again on page 102, but is not explained until one reaches panel 12.8 on page 240. The index itself is seriously deficient, lack­ ing entries for obvious topics such as the CLIR (or its predecessor, the CLR), Li­ brary of Congress, NISO, RLG, RLIN, and UNIX, and excluding a host of contem­ porary personal names important to the development of digital libraries (e.g., Henriette D. Avram). In fact, the entire index contains only two personal names: Vannevar Bush and J.C.R. Licklider. Quite a number of index entries even lack their full complement of locators. In sum, Digital Libraries is a mixed bag, marred by some serious errors and ques­ tionable claims, but serving as a power­ ful stimulus to thinking that departs from tradition. It integrates into a single work detailed information on a vast range of modern technologies and inventories a cluster of unpleasant problems besetting Book Reviews 291 contemporary librarianship. Digital Li­ braries is a disturbing, distressing book, as it should be. Modern librarianship can benefit from a gadfly, and Arms plays that role with admirable effectiveness.—Allen B. Veaner, Tucson, Arizona Authenticity in a Digital Environment. Washington, D.C.: Council on Library and Information Resources, 2000. 76pp. $20, alk. paper (ISBN 1-887334-77-7). On January 24, 2000, the Council on Li­ brary and Information Resources (CLIR) convened a group of recognized experts to ponder the questions: What is an au­ thentic digital object? How do the stan­ dards of archival preservation apply to digital artifacts? Where can archivists position themselves, as the keepers of the intellectual record, to ensure that future generations will be able to study certifi­ ably genuine digital documents? In an effort to get the discussion started, CLIR asked five leaders in the fields of archiving and digitization to write posi­ tion papers on various aspects of the topic of ensuring authenticity of the digital record. This publication contains those five essays. Charles T. Cullen, president and li­ brarian of the Newberry Library, writes in his essay, “Authentication of Digital Objects: Lessons from a Historian’s Re­ search,” of the difficulty of affirming the provenance of paper objects, let alone that of digital objects. He expresses chagrin at the lack of real signature markings that would prove beyond doubt that a docu­ ment is truly the work of the assumed author. He touches on the ease with which changes can be made without note in digi­ tal work, even when an honest transcrip­ tion is attempted. Who can vouch for the fidelity of the transcript to the original? Finally, Cullen urges librarians, publish­ ers, and authors to push forward to find methods to mark digital works with iden­ tifiers that will prove authorship. Peter B. Hirtle, codirector of the Cornell Institute for Digital Collections, in his essay, “Archival Authenticity in a Digital Age,” focuses on the records used for authentication of an object. As an ex­ ample, he uses the USS Constellation, a wooden-hulled navy vessel moored in Baltimore Harbor. The Constellation was recorded as being Baltimore built in 1797, and all historic documents regarding her construction and launch were consistent in this regard. However, it was discov­ ered that she was actually built in Nor­ folk, Virginia, in 1854, and constructed to look like the much-admired USS Consti­ tution, but commissioned for duty during the Civil War, not the Revolutionary War. The documentation had been “adjusted” in 1909 to reflect the earlier building date, although records prior to 1909 showed her accurate construction date. Hirtle’s point is that if paper records can be so manipulated and still appear authentic, archivists will need to exert great effort to maintain accurate provenance for digi­ tal objects. He proposes that one collec­ tive method will be “social mechanisms of control,” or diplomatics, a body of prac­ tices that has long been accepted as a rig­ orous validation technique. Trust in the repository will continue to be a key con­ cept. David M. Levy, a consultant on docu­ ments, digital libraries, and publishing, uses a humorous example in his essay, “Where’s Waldo? Reflections on Copies and Authenticity in a Digital Environ­ ment,” to make the point that in a digital world, in which a copy can be identical to the original (or nearly so), the ascer­ taining of authenticity of the original be­ comes almost impossible. First establish­ ing that it may not be possible to create stable digital objects, he suggests that audit trails may be useful in defining the “original.” He posits that our first step is to understand what it is we want to ac­ complish, then to discern what is possible to accomplish. Clifford A. Lynch, executive director of the Coalition for Networked Informa­ tion, in his essay, “Authenticity and In­ tegrity in the Digital Environment: An Exploratory Analysis of the Central Role of Trust,” sets forth many propositions