The rapidly developing digital library environment continues to present many challenges, not only to those who are just beginning to dabble in digital library initiatives, but also to those with experience. Metadata for Digital Collections is an excellent addition to the growing literature addressing this topic. The author, Steven Miller, is an experienced cataloger and cataloging department manager. This experience, combined with his position teaching courses in metadata, cataloging, and information architecture at the University of Wisconsin-Milwaukee School of Information Studies, makes him ideally suited to address the development and application of metadata to digital collections.

Metadata for Digital Collections is organized into eleven chapters that cover all aspects of creating metadata in a digital library setting. The first chapter begins with the basics: defining metadata, describing types of metadata applied to digital collections, and introducing the reader to metadata standards. Several definitions of metadata are provided and, taken together, they illustrate for the reader how diverse our understanding of metadata can be. Chapter 2 discusses the foundations of resource description, and because of its ubiquity, introduces the Dublin Core (DC) metadata element set. Although prior cataloging knowledge would help the reader put resource description into the context of library databases, prior cataloging experience is not necessary to understand the introductory concepts presented in this chapter.

Chapters 3 and 4 continue the approach of addressing metadata concepts through the application of the DC standard. Chapter 3 explores how resources are identified and how responsibility for creation, contribution, and publication is assigned. Each topic, such as title, identifier, dates, etc., is addressed in a general section, followed by a section devoted to the same topic as defined by DC. Chapter 4 addresses how subject, form, and genre are handled for digital materials. Again, each topic, such as type, genre, and format, is discussed broadly, followed by a description of how that topic is handled in DC.

Chapter 5 makes the case that controlled vocabularies are a crucial aspect of resource description. The role of controlled vocabulary in disambiguation and establishing hierarchical relationships is explained. Many types of controlled vocabulary are discussed, including lists, synonym rings, authority files, taxonomies, and thesauri. The concept of creating a specialized vocabulary is not neglected—references to the American National Standards Institute/National Information Standards Organization (ANSI/NISO) guidelines on creating controlled vocabularies are included.

Metadata created according to a variety of standards, such as DC, Visual Resource Association (VRA) Core, or Metadata Object Description Schema (MODS), can be encoded for storage and transmission using XML. Chapter 6 is devoted to a basic introduction of how the XML encoding standard can be used effectively to store and transmit data. Chapters 7 and 8 address the MODS and VRA Core categories. These chapters are filled with useful examples of MODS and VRA Core records encoded in XML.

Chapter 9 addresses metadata interoperability, sharing, and quality—critical issues in ensuring the long-term viability of metadata created for digital resources. The chapter concludes with suggestions for ways to improve metadata interoperability and quality, such as using DC or another standard element set; including an appropriate amount of contextual information and access points; entering data values that are machine-readable and linkable; distinguishing administrative metadata from descriptive; and documenting local practices. These suggestions are standard practices in traditional cataloging policies and procedures, but have yet to be carried over to the digital realm in a consistent way.

Despite the many metadata schemes already developed and standardized, there are still instances when it is desirable to create a specialized metadata scheme. The steps in developing a metadata scheme are thoroughly outlined in chapter 10. Examples of such schemes also are provided, including DC metadata documentation from the Collaborative Digitization Program, OhioLINK, and Indiana Memory, and MODS documentation from the Digital Library Federation's Aquifer project. These examples are invaluable resources for anyone trying to develop their own documentation.

The final chapter is devoted to a discussion of linked data and the Semantic Web. Although as yet one see few practical applications of linked date, monitoring concepts and development in this emerging field is important.

In addition to the extensive references at the end of each chapter, a robust bibliography and an index appear at the end of the book. The book is generously illustrated with more than one hundred figures and tables. Sidebars illustrating concepts, clarifying definitions, and providing examples are present throughout. The book is clearly written and accessible to students learning about metadata for the first time, but also rich enough to be useful for the experienced practitioner.

Metadata for Digital Collections is well suited for both practicing professionals and students. It provides an excellent grounding in all aspects of applying metadata in a digital library setting and would be a useful addition to any professional library. It also would be appropriate for use in a library or information science course for students who are learning about the organization of information.